diff --git a/develop/doc_cn/_sources/faq/build_and_install/index_cn.rst.txt b/develop/doc_cn/_sources/faq/build_and_install/index_cn.rst.txt
new file mode 100644
index 0000000000000000000000000000000000000000..f1677e216f31d79b53ac29a0afbf6fbb886a0dcd
--- /dev/null
+++ b/develop/doc_cn/_sources/faq/build_and_install/index_cn.rst.txt
@@ -0,0 +1,111 @@
+###################
+编译安装与单元测试
+###################
+
+..  contents::
+
+1. 运行Docker GPU镜像出现 "CUDA driver version is insufficient"
+----------------------------------------------------------------
+
+用户在使用PaddlePaddle GPU的Docker镜像的时候，常常出现 `Cuda Error: CUDA driver version is insufficient for CUDA runtime version`, 原因在于没有把机器上CUDA相关的驱动和库映射到容器内部。
+具体的解决方法是：
+
+..  code-block:: bash
+
+    $ export CUDA_SO="$(\ls usr/lib64/libcuda* | xargs -I{} echo '-v {}:{}') $(\ls /usr/lib64/libnvidia* | xargs -I{} echo '-v {}:{}')"
+    $ export DEVICES=$(\ls /dev/nvidia* | xargs -I{} echo '--device {}:{}')
+    $ docker run ${CUDA_SO} ${DEVICES} -it paddledev/paddlepaddle:latest-gpu
+
+更多关于Docker的安装与使用, 请参考 `PaddlePaddle Docker 文档 <http://www.paddlepaddle.org/doc_cn/build_and_install/install/docker_install.html>`_ 。
+
+
+2. CMake源码编译, 找到的PythonLibs和PythonInterp版本不一致
+----------------------------------------------------------------
+
+这是目前CMake寻找Python的逻辑存在缺陷，如果系统安装了多个Python版本，CMake找到的Python库和Python解释器版本可能有不一致现象，导致编译PaddlePaddle失败。正确的解决方法是，
+用户强制指定特定的Python版本，具体操作如下：
+
+    ..  code-block:: bash
+
+        cmake .. -DPYTHON_EXECUTABLE=<exc_path> -DPYTHON_LIBRARY=<lib_path>  -DPYTHON_INCLUDE_DIR=<inc_path>
+
+用户需要指定本机上Python的路径：``<exc_path>``, ``<lib_path>``, ``<inc_path>``
+
+3. CMake源码编译，Paddle版本号为0.0.0
+--------------------------------------
+
+如果运行 :code:`paddle version`, 出现 :code:`PaddlePaddle 0.0.0`；或者运行 :code:`cmake ..`，出现
+
+..  code-block:: bash
+
+    CMake Warning at cmake/version.cmake:20 (message):
+      Cannot add paddle version from git tag
+
+那么用户需要拉取所有的远程分支到本机，命令为 :code:`git fetch upstream`，然后重新cmake即可。
+
+4. paddlepaddle\*.whl is not a supported wheel on this platform.
+------------------------------------------------------------------------
+
+出现这个问题的主要原因是，没有找到和当前系统匹配的paddlepaddle安装包。最新的paddlepaddle python安装包支持Linux x86_64和MacOS 10.12操作系统，并安装了python 2.7和pip 9.0.1。
+
+更新 :code:`pip` 包的方法是\:
+
+..  code-block:: bash
+
+    pip install --upgrade pip
+
+如果还不行，可以执行 :code:`python -c "import pip; print(pip.pep425tags.get_supported())"` 获取当前系统支持的python包的后缀，
+并对比是否和正在安装的后缀一致。
+
+如果系统支持的是 :code:`linux_x86_64` 而安装包是 :code:`manylinux1_x86_64` ，需要升级pip版本到最新；
+如果系统支持 :code:`manylinux1_x86_64` 而安装包（本地）是 :code:`linux_x86_64` ，可以重命名这个whl包为 :code:`manylinux1_x86_64` 再安装。
+
+5. 编译安装后执行 import paddle.v2 as paddle 报ImportError: No module named v2
+------------------------------------------------------------------------------------------
+先查看一下是否曾经安装过paddle v1版本，有的话需要先卸载：
+
+pip uninstall py_paddle paddle
+
+然后安装paddle的python环境, 在build目录下执行
+
+pip install python/dist/paddle*.whl && pip install ../paddle/dist/py_paddle*.whl
+
+6. 遇到“非法指令”或者是“illegal instruction”
+--------------------------------------------
+
+PaddlePaddle使用avx SIMD指令提高cpu执行效率，因此错误的使用二进制发行版可能会导致这种错误，请选择正确的版本。
+
+7.  python相关的单元测试都过不了
+--------------------------------
+
+如果出现以下python相关的单元测试都过不了的情况：
+
+..  code-block:: bash
+
+    24 - test_PyDataProvider (Failed)
+    26 - test_RecurrentGradientMachine (Failed)
+    27 - test_NetworkCompare (Failed)
+    28 - test_PyDataProvider2 (Failed)
+    32 - test_Prediction (Failed)
+    33 - test_Compare (Failed)
+    34 - test_Trainer (Failed)
+    35 - test_TrainerOnePass (Failed)
+    36 - test_CompareTwoNets (Failed)
+    37 - test_CompareTwoOpts (Failed)
+    38 - test_CompareSparse (Failed)
+    39 - test_recurrent_machine_generation (Failed)
+    40 - test_PyDataProviderWrapper (Failed)
+    41 - test_config_parser (Failed)
+    42 - test_swig_api (Failed)
+    43 - layers_test (Failed)
+
+并且查询PaddlePaddle单元测试的日志，提示：
+
+..  code-block:: bash
+
+    paddle package is already in your PYTHONPATH. But unittest need a clean environment.
+    Please uninstall paddle package before start unittest. Try to 'pip uninstall paddle'.
+
+解决办法是：
+
+* 卸载PaddlePaddle包 :code:`pip uninstall paddle`, 清理掉老旧的PaddlePaddle安装包，使得单元测试有一个干净的环境。如果PaddlePaddle包已经在python的site-packages里面，单元测试会引用site-packages里面的python包，而不是源码目录里 :code:`/python` 目录下的python包。同时，即便设置 :code:`PYTHONPATH` 到 :code:`/python` 也没用，因为python的搜索路径是优先已经安装的python包。
diff --git a/develop/doc_cn/_sources/faq/cluster/index_cn.rst.txt b/develop/doc_cn/_sources/faq/cluster/index_cn.rst.txt
new file mode 100644
index 0000000000000000000000000000000000000000..e59c1e1a54a0c876d1e6e89f88030de59fb9fc1a
--- /dev/null
+++ b/develop/doc_cn/_sources/faq/cluster/index_cn.rst.txt
@@ -0,0 +1,17 @@
+###############
+集群训练与预测
+###############
+
+..  contents::
+
+1. 集群多节点训练，日志中保存均为网络通信类错误
+------------------------------------------------
+
+集群多节点训练，日志报错为网络通信类错误，比如 :code:`Connection reset by peer` 等。
+此类报错通常是由于某一个节点的错误导致这个节点的训练进程退出，从而引发其他节点无法连接导致，可以参考下面的步骤排查：
+
+* 从 :code:`train.log` ， :code:`server.log` 找到最早报错的地方，查看是否是其他错误引发的报错（比如FPE，内存不足，磁盘空间不足等）。
+
+* 如果发现最早的报错就是网络通信的问题，很有可能是非独占方式执行导致的端口冲突，可以联系OP，看当前MPI集群是否支持resource=full参数提交，如果支持增加此参数提交，并更换job 端口。
+
+* 如果当前MPI集群并不支持任务独占模式，可以联系OP是否可以更换集群或升级当前集群。
diff --git a/develop/doc_cn/_sources/faq/index_cn.rst.txt b/develop/doc_cn/_sources/faq/index_cn.rst.txt
index d69d111917ca7a79bc65b051c8eefaba165d77bd..9929767cac212237b3e2c3a547ba9a3c9d5f0979 100644
--- a/develop/doc_cn/_sources/faq/index_cn.rst.txt
+++ b/develop/doc_cn/_sources/faq/index_cn.rst.txt
@@ -1,592 +1,11 @@
-####################
 FAQ
-####################
+====
 
-..  contents::
+..  toctree::
+  :maxdepth: 1
 
-1. 如何减少内存占用
----------------------------------
-
-神经网络的训练本身是一个非常消耗内存和显存的工作，经常会消耗数10GB的内存和数GB的显存。
-PaddlePaddle的内存占用主要分为如下几个方面\:
-
-* DataProvider缓冲池内存（只针对内存）
-* 神经元激活内存（针对内存和显存）
-* 参数内存 （针对内存和显存）
-* 其他内存杂项
-
-其中，其他内存杂项是指PaddlePaddle本身所用的一些内存，包括字符串分配，临时变量等等，暂不考虑在内。
-
-减少DataProvider缓冲池内存
-++++++++++++++++++++++++++
-
-PyDataProvider使用的是异步加载，同时在内存里直接随即选取数据来做Shuffle。即
-
-..  graphviz::
-
-    digraph {
-        rankdir=LR;
-        数据文件 -> 内存池 -> PaddlePaddle训练
-    }
-
-所以，减小这个内存池即可减小内存占用，同时也可以加速开始训练前数据载入的过程。但是，这
-个内存池实际上决定了shuffle的粒度。所以，如果将这个内存池减小，又要保证数据是随机的，
-那么最好将数据文件在每次读取之前做一次shuffle。可能的代码为
-
-..  literalinclude:: src/reduce_min_pool_size.py
-
-这样做可以极大的减少内存占用，并且可能会加速训练过程，详细文档参考 :ref:`api_pydataprovider2` 。
-
-神经元激活内存
-++++++++++++++
-
-神经网络在训练的时候，会对每一个激活暂存一些数据，如神经元激活值等。
-在反向传递的时候，这些数据会被用来更新参数。这些数据使用的内存主要和两个参数有关系，
-一是batch size，另一个是每条序列(Sequence)长度。所以，其实也是和每个mini-batch中包含
-的时间步信息成正比。
-
-所以做法可以有两种：
-
-* 减小batch size。 即在网络配置中 :code:`settings(batch_size=1000)` 设置成一个小一些的值。但是batch size本身是神经网络的超参数，减小batch size可能会对训练结果产生影响。
-* 减小序列的长度，或者直接扔掉非常长的序列。比如，一个数据集大部分序列长度是100-200,
-  但是突然有一个10000长的序列，就很容易导致内存超限，特别是在LSTM等RNN中。
-
-参数内存
-++++++++
-
-PaddlePaddle支持非常多的优化算法(Optimizer)，不同的优化算法需要使用不同大小的内存。
-例如使用 :code:`adadelta` 算法，则需要使用等于权重参数规模大约5倍的内存。举例，如果参数保存下来的模型目录
-文件为 :code:`100M`， 那么该优化算法至少需要 :code:`500M` 的内存。
-
-可以考虑使用一些优化算法，例如 :code:`momentum`。
-
-2. 如何加速PaddlePaddle的训练速度
----------------------------------
-
-加速PaddlePaddle训练可以考虑从以下几个方面\：
-
-* 减少数据载入的耗时
-* 加速训练速度
-* 利用分布式训练驾驭更多的计算资源
-
-减少数据载入的耗时
-++++++++++++++++++
-
-使用\ :code:`pydataprovider`\ 时，可以减少缓存池的大小，同时设置内存缓存功能，即可以极大的加速数据载入流程。
-:code:`DataProvider` 缓存池的减小，和之前减小通过减小缓存池来减小内存占用的原理一致。
-
-..  literalinclude:: src/reduce_min_pool_size.py
-
-同时 :code:`@provider` 接口有一个 :code:`cache` 参数来控制缓存方法，将其设置成 :code:`CacheType.CACHE_PASS_IN_MEM` 的话，会将第一个 :code:`pass` (过完所有训练数据即为一个pass)生成的数据缓存在内存里，在之后的 :code:`pass` 中，不会再从 :code:`python` 端读取数据，而是直接从内存的缓存里读取数据。这也会极大减少数据读入的耗时。
-
-
-加速训练速度
-++++++++++++
-
-PaddlePaddle支持Sparse的训练，sparse训练需要训练特征是 :code:`sparse_binary_vector` 、 :code:`sparse_vector` 、或者 :code:`integer_value` 的任一一种。同时，与这个训练数据交互的Layer，需要将其Parameter设置成 sparse 更新模式，即设置 :code:`sparse_update=True`
-
-这里使用简单的 :code:`word2vec` 训练语言模型距离，具体使用方法为\:
-
-使用一个词前两个词和后两个词，来预测这个中间的词。这个任务的DataProvider为\:
-
-..  literalinclude:: src/word2vec_dataprovider.py
-
-这个任务的配置为\:
-
-..  literalinclude:: src/word2vec_config.py
-
-
-利用更多的计算资源
-++++++++++++++++++
-
-利用更多的计算资源可以分为一下几个方式来进行\:
-
-* 单机CPU训练
-
-  * 使用多线程训练。设置命令行参数 :code:`trainer_count`。
-
-* 单机GPU训练
-
-  * 使用显卡训练。设置命令行参数 :code:`use_gpu`。
-  * 使用多块显卡训练。设置命令行参数 :code:`use_gpu` 和 :code:`trainer_count` 。
-
-* 多机训练
-
-  * 请参考 :ref:`cluster_train` 。
-
-
-3. 遇到“非法指令”或者是“illegal instruction”
---------------------------------------------
-
-PaddlePaddle使用avx SIMD指令提高cpu执行效率，因此错误的使用二进制发行版可能会导致这种错误，请选择正确的版本。
-
-4. 如何选择SGD算法的学习率
---------------------------
-
-在采用sgd/async_sgd进行训练时，一个重要的问题是选择正确的learning_rate。如果learning_rate太大，那么训练有可能不收敛，如果learning_rate太小，那么收敛可能很慢，导致训练时间过长。
-
-通常做法是从一个比较大的learning_rate开始试，如果不收敛，那减少学习率10倍继续试验，直到训练收敛为止。那么如何判断训练不收敛呢？可以估计出如果模型采用不变的输出最小的cost0是多少。
-
-如果训练过程的的cost明显高于这个常数输出的cost，那么我们可以判断为训练不收敛。举一个例子，假如我们是三分类问题，采用multi-class-cross-entropy作为cost，数据中0,1,2三类的比例为 :code:`0.2, 0.5, 0.3` , 那么常数输出所能达到的最小cost是 :code:`-(0.2*log(0.2)+0.5*log(0.5)+0.3*log(0.3))=1.03` 。如果训练一个pass（或者更早）后，cost还大于这个数，那么可以认为训练不收敛，应该降低学习率。
-
-
-5. 如何初始化参数
------------------
-
-默认情况下，PaddlePaddle使用均值0，标准差为 :math:`\frac{1}{\sqrt{d}}` 来初始化参数。其中 :math:`d` 为参数矩阵的宽度。这种初始化方式在一般情况下不会产生很差的结果。如果用户想要自定义初始化方式，PaddlePaddle目前提供两种参数初始化的方式\:
-
-* 高斯分布。将 :code:`param_attr` 设置成 :code:`param_attr=ParamAttr(initial_mean=0.0, initial_std=1.0)`
-* 均匀分布。将 :code:`param_attr` 设置成 :code:`param_attr=ParamAttr(initial_max=1.0, initial_min=-1.0)`
-
-比如设置一个全连接层的参数初始化方式和bias初始化方式，可以使用如下代码。
-
-..  code-block:: python
-
-    hidden = fc_layer(input=ipt, param_attr=ParamAttr(initial_max=1.0, initial_min=-1.0),
-                      bias_attr=ParamAttr(initial_mean=1.0, initial_std=0.0))
-
-上述代码将bias全部初始化为1.0, 同时将参数初始化为 :code:`[1.0, -1.0]` 的均匀分布。
-
-6. 如何共享参数
----------------
-
-PaddlePaddle的参数使用名字 :code:`name` 作为参数的ID，相同名字的参数，会共享参数。设置参数的名字，可以使用 :code:`ParamAttr(name="YOUR_PARAM_NAME")` 来设置。更方便的设置方式，是使得要共享的参数使用同样的 :code:`ParamAttr` 对象。
-
-简单的全连接网络，参数共享的配置示例为\:
-
-..  literalinclude:: ../../python/paddle/trainer_config_helpers/tests/configs/shared_fc.py
-
-这里 :code:`hidden_a` 和 :code:`hidden_b` 使用了同样的parameter和bias。并且softmax层的两个输入也使用了同样的参数 :code:`softmax_param`。
-
-7. paddlepaddle\*.whl is not a supported wheel on this platform.
-------------------------------------------------------------------------
-
-出现这个问题的主要原因是，没有找到和当前系统匹配的paddlepaddle安装包。最新的paddlepaddle python安装包支持Linux x86_64和MacOS 10.12操作系统，并安装了python 2.7和pip 9.0.1。
-
-更新 :code:`pip` 包的方法是\:
-
-..  code-block:: bash
-
-    pip install --upgrade pip
-
-如果还不行，可以执行 :code:`python -c "import pip; print(pip.pep425tags.get_supported())"` 获取当前系统支持的python包的后缀，
-并对比是否和正在安装的后缀一致。
-
-如果系统支持的是 :code:`linux_x86_64` 而安装包是 :code:`manylinux1_x86_64` ，需要升级pip版本到最新；
-如果系统支持 :code:`manylinux1_x86_64` 而安装包（本地）是 :code:`linux_x86_64` ，可以重命名这个whl包为 :code:`manylinux1_x86_64` 再安装。
-
-8.  python相关的单元测试都过不了
---------------------------------
-
-如果出现以下python相关的单元测试都过不了的情况：
-
-..  code-block:: bash
-
-    24 - test_PyDataProvider (Failed)
-    26 - test_RecurrentGradientMachine (Failed)
-    27 - test_NetworkCompare (Failed)
-    28 - test_PyDataProvider2 (Failed)
-    32 - test_Prediction (Failed)
-    33 - test_Compare (Failed)
-    34 - test_Trainer (Failed)
-    35 - test_TrainerOnePass (Failed)
-    36 - test_CompareTwoNets (Failed)
-    37 - test_CompareTwoOpts (Failed)
-    38 - test_CompareSparse (Failed)
-    39 - test_recurrent_machine_generation (Failed)
-    40 - test_PyDataProviderWrapper (Failed)
-    41 - test_config_parser (Failed)
-    42 - test_swig_api (Failed)
-    43 - layers_test (Failed)
-
-并且查询PaddlePaddle单元测试的日志，提示：
-
-..  code-block:: bash
-
-    paddle package is already in your PYTHONPATH. But unittest need a clean environment.
-    Please uninstall paddle package before start unittest. Try to 'pip uninstall paddle'.
-
-解决办法是：
-
-* 卸载PaddlePaddle包 :code:`pip uninstall paddle`, 清理掉老旧的PaddlePaddle安装包，使得单元测试有一个干净的环境。如果PaddlePaddle包已经在python的site-packages里面，单元测试会引用site-packages里面的python包，而不是源码目录里 :code:`/python` 目录下的python包。同时，即便设置 :code:`PYTHONPATH` 到 :code:`/python` 也没用，因为python的搜索路径是优先已经安装的python包。
-
-
-9. 运行Docker GPU镜像出现 "CUDA driver version is insufficient"
-----------------------------------------------------------------
-
-用户在使用PaddlePaddle GPU的Docker镜像的时候，常常出现 `Cuda Error: CUDA driver version is insufficient for CUDA runtime version`, 原因在于没有把机器上CUDA相关的驱动和库映射到容器内部。
-具体的解决方法是：
-
-..  code-block:: bash
-
-    $ export CUDA_SO="$(\ls usr/lib64/libcuda* | xargs -I{} echo '-v {}:{}') $(\ls /usr/lib64/libnvidia* | xargs -I{} echo '-v {}:{}')"
-    $ export DEVICES=$(\ls /dev/nvidia* | xargs -I{} echo '--device {}:{}')
-    $ docker run ${CUDA_SO} ${DEVICES} -it paddledev/paddlepaddle:latest-gpu
-
-更多关于Docker的安装与使用, 请参考 `PaddlePaddle Docker 文档 <http://www.paddlepaddle.org/doc_cn/build_and_install/install/docker_install.html>`_ 。
-
-
-10. CMake源码编译, 找到的PythonLibs和PythonInterp版本不一致
-----------------------------------------------------------------
-
-这是目前CMake寻找Python的逻辑存在缺陷，如果系统安装了多个Python版本，CMake找到的Python库和Python解释器版本可能有不一致现象，导致编译PaddlePaddle失败。正确的解决方法是，
-用户强制指定特定的Python版本，具体操作如下：
-
-    ..  code-block:: bash
-
-        cmake .. -DPYTHON_EXECUTABLE=<exc_path> -DPYTHON_LIBRARY=<lib_path>  -DPYTHON_INCLUDE_DIR=<inc_path>
-
-用户需要指定本机上Python的路径：``<exc_path>``, ``<lib_path>``, ``<inc_path>``
-
-11. CMake源码编译，Paddle版本号为0.0.0
---------------------------------------
-
-如果运行 :code:`paddle version`, 出现 :code:`PaddlePaddle 0.0.0`；或者运行 :code:`cmake ..`，出现
-
-..  code-block:: bash
-
-    CMake Warning at cmake/version.cmake:20 (message):
-      Cannot add paddle version from git tag
-
-那么用户需要拉取所有的远程分支到本机，命令为 :code:`git fetch upstream`，然后重新cmake即可。
-
-12. A protocol message was rejected because it was too big
-------------------------------------------------------------
-
-如果在训练NLP相关模型时，出现以下错误：
-
-..  code-block:: bash
-
-    [libprotobuf ERROR google/protobuf/io/coded_stream.cc:171] A protocol message was rejected because it was too big (more than 67108864 bytes).  To increase the limit (or to disable these warnings), see CodedInputStream::SetTotalBytesLimit() in google/protobuf/io/coded_stream.h.
-    F1205 14:59:50.295174 14703 TrainerConfigHelper.cpp:59] Check failed: m->conf.ParseFromString(configProtoStr)
-
-可能的原因是：传给dataprovider的某一个args过大，一般是由于直接传递大字典导致的。错误的define_py_data_sources2类似：
-
-..  code-block:: python
-
-     src_dict = dict()
-     for line_count, line in enumerate(open(src_dict_path, "r")):
-        src_dict[line.strip()] = line_count
-
-     define_py_data_sources2(
-        train_list,
-        test_list,
-        module="dataprovider",
-        obj="process",
-        args={"src_dict": src_dict})
-
-解决方案是：将字典的地址作为args传给dataprovider，然后在dataprovider里面根据该地址加载字典。即define_py_data_sources2应改为：
-
-..  code-block:: python
-
-     define_py_data_sources2(
-        train_list,
-        test_list,
-        module="dataprovider",
-        obj="process",
-        args={"src_dict_path": src_dict_path})
-
-完整源码可参考 `seqToseq <https://github.com/PaddlePaddle/Paddle/tree/develop/demo/seqToseq>`_ 示例。
-
-13. 如何指定GPU设备
--------------------
-
-例如机器上有4块GPU，编号从0开始，指定使用2、3号GPU：
-
-* 方式1：通过 `CUDA_VISIBLE_DEVICES <http://www.acceleware.com/blog/cudavisibledevices-masking-gpus>`_ 环境变量来指定特定的GPU。
-
-..      code-block:: bash
-
-        env CUDA_VISIBLE_DEVICES=2,3 paddle train --use_gpu=true --trainer_count=2
-
-* 方式2：通过命令行参数 ``--gpu_id`` 指定。
-
-..      code-block:: bash
-
-        paddle train --use_gpu=true --trainer_count=2 --gpu_id=2
-
-
-14. 训练过程中出现 :code:`Floating point exception`, 训练因此退出怎么办?
-------------------------------------------------------------------------
-
-Paddle二进制在运行时捕获了浮点数异常，只要出现浮点数异常(即训练过程中出现NaN或者Inf)，立刻退出。浮点异常通常的原因是浮点数溢出、除零等问题。
-主要原因包括两个方面:
-
-* 训练过程中参数或者训练过程中的梯度尺度过大，导致参数累加，乘除等时候，导致了浮点数溢出。
-* 模型一直不收敛，发散到了一个数值特别大的地方。
-* 训练数据有问题，导致参数收敛到了一些奇异的情况。或者输入数据尺度过大，有些特征的取值达到数百万，这时进行矩阵乘法运算就可能导致浮点数溢出。
-
-这里有两种有效的解决方法：
-
-1. 设置 :code:`gradient_clipping_threshold` 参数，示例代码如下：
-
-..  code-block:: python
-
-optimizer = paddle.optimizer.RMSProp(
-    learning_rate=1e-3,
-    gradient_clipping_threshold=10.0,
-    regularization=paddle.optimizer.L2Regularization(rate=8e-4))
-
-具体可以参考  `nmt_without_attention  <https://github.com/PaddlePaddle/models/blob/develop/nmt_without_attention/train.py#L35>`_ 示例。
-
-2. 设置 :code:`error_clipping_threshold` 参数，示例代码如下：
-
-..  code-block:: python
-
-decoder_inputs = paddle.layer.fc(
-    act=paddle.activation.Linear(),
-    size=decoder_size * 3,
-    bias_attr=False,
-    input=[context, current_word],
-    layer_attr=paddle.attr.ExtraLayerAttribute(
-        error_clipping_threshold=100.0))
-
-完整代码可以参考示例 `machine translation <https://github.com/PaddlePaddle/book/blob/develop/08.machine_translation/train.py#L66>`_ 。
-
-两种方法的区别：
-
-1. 两者都是对梯度的截断，但截断时机不同，前者在 :code:`optimzier` 更新网络参数时应用；后者在激活函数反向计算时被调用；
-2. 截断对象不同：前者截断可学习参数的梯度，后者截断回传给前层的梯度;
-
-除此之外，还可以通过减小学习律或者对数据进行归一化处理来解决这类问题。
-
-15. 编译安装后执行 import paddle.v2 as paddle 报ImportError: No module named v2
-------------------------------------------------------------------------------------------
-先查看一下是否曾经安装过paddle v1版本，有的话需要先卸载：
-
-pip uninstall py_paddle paddle
-
-然后安装paddle的python环境, 在build目录下执行
-
-pip install python/dist/paddle*.whl && pip install ../paddle/dist/py_paddle*.whl
-
-16. PaddlePaddle存储的参数格式是什么，如何和明文进行相互转化
----------------------------------------------------------------------
-
-PaddlePaddle保存的模型参数文件内容由16字节头信息和网络参数两部分组成。头信息中，1~4字节表示PaddlePaddle版本信息，请直接填充0；5~8字节表示每个参数占用的字节数，当保存的网络参数为float类型时为4，double类型时为8；9~16字节表示保存的参数总个数。
-
-将PaddlePaddle保存的模型参数还原回明文时，可以使用相应数据类型的 :code:`numpy.array` 加载具体网络参数，此时可以跳过PaddlePaddle模型参数文件的头信息。若在PaddlePaddle编译时，未指定按照double精度编译，默认情况下按照float精度计算，保存的参数也是float类型。这时在使用 :code:`numpy.array` 时，一般设置 :code:`dtype=float32` 。示例如下：
-
-..  code-block:: python
-
-    def read_parameter(fname, width):
-        s = open(fname).read()
-        # skip header
-        vec = np.fromstring(s[16:], dtype=np.float32)
-        # width is the size of the corresponding layer
-        np.savetxt(fname + ".csv", vec.reshape(width, -1),
-                fmt="%.6f", delimiter=",")
-
-
-将明文参数转化为PaddlePaddle可加载的模型参数时，首先构造头信息，再写入网络参数。下面的代码将随机生成的矩阵转化为可以被PaddlePaddle加载的模型参数。
-
-..  code-block:: python
-
-    def gen_rand_param(param_file, width, height, need_trans):
-        np.random.seed()
-        header = struct.pack("iil", 0, 4, height * width)
-        param = np.float32(np.random.rand(height, width))
-        with open(param_file, "w") as fparam:
-            fparam.write(header + param.tostring())
-
-17. 如何加载预训练参数
-------------------------------
-
-* 对加载预训练参数的层，设置其参数属性 :code:`is_static=True`，使该层的参数在训练过程中保持不变。以embedding层为例，代码如下：
-
-..  code-block:: python
-
-    emb_para = paddle.attr.Param(name='emb', is_static=True)
-    paddle.layer.embedding(size=word_dim, input=x, param_attr=emb_para)
-
-
-* 从模型文件将预训练参数载入 :code:`numpy.array`，在创建parameters后，使用 :code:`parameters.set()` 加载预训练参数。PaddlePaddle保存的模型参数文件前16字节为头信息，用户将参数载入 :code:`numpy.array` 时须从第17字节开始。以embedding层为例，代码如下：
-
-..  code-block:: python
-
-    def load_parameter(file_name, h, w):
-        with open(file_name, 'rb') as f:
-            f.read(16)  # skip header.
-            return np.fromfile(f, dtype=np.float32).reshape(h, w)
-
-    parameters = paddle.parameters.create(my_cost)
-    parameters.set('emb', load_parameter(emb_param_file, 30000, 256))
-
-18. 集群多节点训练，日志中保存均为网络通信类错误
------------------------------------------------------------
-
-集群多节点训练，日志报错为网络通信类错误，比如 :code:`Connection reset by peer` 等。
-此类报错通常是由于某一个节点的错误导致这个节点的训练进程退出，从而引发其他节点无法连接导致，可以参考下面的步骤排查：
-
-* 从 :code:`train.log` ， :code:`server.log` 找到最早报错的地方，查看是否是其他错误引发的报错（比如FPE，内存不足，磁盘空间不足等）。
-
-* 如果发现最早的报错就是网络通信的问题，很有可能是非独占方式执行导致的端口冲突，可以联系OP，看当前MPI集群是否支持resource=full参数提交，如果支持增加此参数提交，并更换job 端口。
-
-* 如果当前MPI集群并不支持任务独占模式，可以联系OP是否可以更换集群或升级当前集群。
-
-19.  如何调用 infer 接口输出多个layer的预测结果
------------------------------------------------------------
-
-* 将需要输出的层作为 :code:`paddle.inference.Inference()` 接口的 :code:`output_layer` 参数输入，代码如下：
-
-..  code-block:: python
-
-    inferer = paddle.inference.Inference(output_layer=[layer1, layer2], parameters=parameters)
-
-* 指定要输出的字段进行输出。以输出 :code:`value` 字段为例，代码如下：
-
-..  code-block:: python
-
-    out = inferer.infer(input=data_batch, field=["value"])
-
-需要注意的是：
-
-* 如果指定了2个layer作为输出层，实际上需要的输出结果是两个矩阵；
-* 假设第一个layer的输出A是一个 N1 * M1 的矩阵，第二个 Layer 的输出B是一个 N2 * M2 的矩阵；
-* paddle.v2 默认会将A和B 横向拼接，当N1 和 N2 大小不一样时，会报如下的错误：
-
-..      code-block:: python
-
-    ValueError: all the input array dimensions except for the concatenation axis must match exactly
-
-多个层的输出矩阵的高度不一致导致拼接失败，这种情况常常发生在：
-
-* 同时输出序列层和非序列层；
-* 多个输出层处理多个不同长度的序列;
-
-此时可以在调用infer接口时通过设置 :code:`flatten_result=False` , 跳过“拼接”步骤，来解决上面的问题。这时，infer接口的返回值是一个python list:
-
-* list 中元素的个数等于网络中输出层的个数；
-* list 中每个元素是一个layer的输出结果矩阵，类型是numpy的ndarray；
-* 每一个layer输出矩阵的高度，在非序列输入时：等于样本数；序列输入时等于：输入序列中元素的总数；宽度等于配置中layer的size；
-
-20. :code:`paddle.layer.memory` 的参数 :code:`name` 如何使用
--------------------------------------------------------------
-
-* :code:`paddle.layer.memory` 用于获取特定layer上一时间步的输出，该layer是通过参数 :code:`name` 指定，即，:code:`paddle.layer.memory` 会关联参数 :code:`name` 取值相同的layer，并将该layer上一时间步的输出作为自身当前时间步的输出。
-
-* PaddlePaddle的所有layer都有唯一的name，用户通过参数 :code:`name` 设定，当用户没有显式设定时，PaddlePaddle会自动设定。而 :code:`paddle.layer.memory` 不是真正的layer，其name由参数 :code:`memory_name` 设定，当用户没有显式设定时，PaddlePaddle会自动设定。:code:`paddle.layer.memory` 的参数 :code:`name` 用于指定其要关联的layer，需要用户显式设定。
-
-21. 两种使用 drop_out 的方法有何区别？
------------------------------------------------------
-
-* 在PaddlePaddle中使用dropout有两种方式
-
-  * 在相应layer的 :code:`layer_atter` 设置 :code:`drop_rate`，以 :code:`paddle.layer.fc` 为例，代码如下：
-
-  ..  code-block:: python
-
-      fc = paddle.layer.fc(input=input, layer_attr=paddle.attr.ExtraLayerAttribute(drop_rate=0.5))
-
-  * 使用 :code:`paddle.layer.dropout`，以 :code:`paddle.layer.fc` 为例，代码如下：
-
-  ..  code-block:: python
-
-      fc = paddle.layer.fc(input=input)
-      drop_fc = paddle.layer.dropout(input=fc, dropout_rate=0.5)
-
-* :code:`paddle.layer.dropout` 实际上使用了 :code:`paddle.layer.add_to`，并在该layer里采用第一种方式设置 :code:`drop_rate` 来使用dropout的。这种方式对内存消耗较大。
-
-* PaddlePaddle在激活函数里实现dropout，而不是在layer里实现。
-
-* :code:`paddle.layer.lstmemory`、:code:`paddle.layer.grumemory`、:code:`paddle.layer.recurrent` 不是通过一般的方式来实现对输出的激活，所以不能采用第一种方式在这几个layer里设置 :code:`drop_rate` 来使用dropout。若要对这几个layer使用dropout，可采用第二种方式，即使用 :code:`paddle.layer.dropout`。
-
-22. 如何设置学习率退火（learning rate annealing）
-------------------------------------------------
-
-在相应的优化算法里设置learning_rate_schedule及相关参数，以使用Adam算法为例，代码如下：
-
-..  code-block:: python
-
-    optimizer = paddle.optimizer.Adam(
-        learning_rate=1e-3,
-        learning_rate_decay_a=0.5,
-        learning_rate_decay_b=0.75,
-        learning_rate_schedule="poly",)
-
-PaddlePaddle目前支持8种learning_rate_schedule，这8种learning_rate_schedule及其对应学习率计算方式如下：
-
-* "constant"
-
-  lr = learning_rate
-
-* "poly"
-
-  lr = learning_rate * pow(1 + learning_rate_decay_a * num_samples_processed, -learning_rate_decay_b)
-
-  其中，num_samples_processed为已训练样本数，下同。
-
-* "caffe_poly"
-
-  lr = learning_rate * pow(1.0 - num_samples_processed / learning_rate_decay_a, learning_rate_decay_b)
-
-* "exp"
-
-  lr = learning_rate * pow(learning_rate_decay_a, num_samples_processed / learning_rate_decay_b)
-
-* "discexp"
-
-  lr = learning_rate * pow(learning_rate_decay_a, floor(num_samples_processed / learning_rate_decay_b))
-
-* "linear"
-
-  lr = max(learning_rate - learning_rate_decay_a * num_samples_processed, learning_rate_decay_b)
-
-* "manual"
-
-  这是一种按已训练样本数分段取值的学习率退火方法。使用该learning_rate_schedule时，用户通过参数 :code:`learning_rate_args` 设置学习率衰减因子分段函数，当前的学习率为所设置 :code:`learning_rate` 与当前的衰减因子的乘积。以使用Adam算法为例，代码如下：
-
-  ..  code-block:: python
-
-      optimizer = paddle.optimizer.Adam(
-          learning_rate=1e-3,
-          learning_rate_schedule="manual",
-          learning_rate_args="1000:1.0,2000:0.9,3000:0.8",)
-
-  在该示例中，当已训练样本数小于等于1000时，学习率为 :code:`1e-3 * 1.0`；当已训练样本数大于1000小于等于2000时，学习率为 :code:`1e-3 * 0.9`；当已训练样本数大于2000时，学习率为 :code:`1e-3 * 0.8`。
-
-* "pass_manual"
-
-  这是一种按已训练pass数分段取值的学习率退火方法。使用该learning_rate_schedule时，用户通过参数 :code:`learning_rate_args` 设置学习率衰减因子分段函数，当前的学习率为所设置 :code:`learning_rate` 与当前的衰减因子的乘积。以使用Adam算法为例，代码如下：
-
-  ..  code-block:: python
-
-      optimizer = paddle.optimizer.Adam(
-          learning_rate=1e-3,
-          learning_rate_schedule="manual",
-          learning_rate_args="1:1.0,2:0.9,3:0.8",)
-
-  在该示例中，当已训练pass数小于等于1时，学习率为 :code:`1e-3 * 1.0`；当已训练pass数大于1小于等于2时，学习率为 :code:`1e-3 * 0.9`；当已训练pass数大于2时，学习率为 :code:`1e-3 * 0.8`。
-
-23. 出现 :code:`Duplicated layer name` 错误怎么办
---------------------------------------------------
-
-出现该错误的原因一般是用户对不同layer的参数 :code:`name` 设置了相同的取值。遇到该错误时，先找出参数 :code:`name` 取值相同的layer，然后将这些layer的参数 :code:`name` 设置为不同的值。
-
-24. PaddlePaddle 中不同的 recurrent layer 的区别
---------------------------------------------------
-以LSTM为例，在PaddlePaddle中包含以下 recurrent layer：
-
-* :code:`paddle.layer.lstmemory`
-* :code:`paddle.networks.simple_lstm`
-* :code:`paddle.networks.lstmemory_group`
-* :code:`paddle.networks.bidirectional_lstm`
-
-按照具体实现方式可以归纳为2类：
-
-1. 由 recurrent_group 实现的 recurrent layer：
-
-  * 用户在使用这一类recurrent layer时，可以访问由recurrent unit在一个时间步内计算得到的中间值（例如：hidden states, memory cells等）；
-  * 上述的 :code:`paddle.networks.lstmemory_group` 是这一类的 recurrent layer ；
-
-2. 将recurrent layer作为一个整体来实现：
-
-  * 用户在使用这一类recurrent layer，只能访问它们的输出值；
-  * 上述的 :code:`paddle.networks.lstmemory_group` 、 :code:`paddle.networks.simple_lstm` 和 :code:`paddle.networks.bidirectional_lstm` 属于这一类的实现；
-
-将recurrent layer作为一个整体来实现， 能够针对CPU和GPU的计算做更多优化， 所以相比于recurrent group的实现方式， 第二类 recurrent layer 计算效率更高。 在实际应用中，如果用户不需要访问LSTM的中间变量，而只需要获得recurrent layer计算的输出，我们建议使用第二类实现。
-
-此外，关于LSTM, PaddlePaddle中还包含 :code:`paddle.networks.lstmemory_unit` 这一计算单元：
-
-  * 不同于上述介绍的recurrent layer , :code:`paddle.networks.lstmemory_unit` 定义了LSTM单元在一个时间步内的计算过程，它并不是一个完整的recurrent layer，也不能接收序列数据作为输入；
-  * :code:`paddle.networks.lstmemory_unit` 只能在recurrent_group中作为step function使用；
+  build_and_install/index_cn.rst
+  model/index_cn.rst
+  parameter/index_cn.rst
+  local/index_cn.rst
+  cluster/index_cn.rst
diff --git a/develop/doc_cn/_sources/faq/local/index_cn.rst.txt b/develop/doc_cn/_sources/faq/local/index_cn.rst.txt
new file mode 100644
index 0000000000000000000000000000000000000000..75c4ba028e497e29e9030a86514348726d9c0a80
--- /dev/null
+++ b/develop/doc_cn/_sources/faq/local/index_cn.rst.txt
@@ -0,0 +1,213 @@
+###############
+本地训练与预测
+###############
+
+..  contents::
+
+1. 如何减少内存占用
+-------------------
+
+神经网络的训练本身是一个非常消耗内存和显存的工作，经常会消耗数10GB的内存和数GB的显存。
+PaddlePaddle的内存占用主要分为如下几个方面\:
+
+* DataProvider缓冲池内存（只针对内存）
+* 神经元激活内存（针对内存和显存）
+* 参数内存 （针对内存和显存）
+* 其他内存杂项
+
+其中，其他内存杂项是指PaddlePaddle本身所用的一些内存，包括字符串分配，临时变量等等，暂不考虑在内。
+
+减少DataProvider缓冲池内存
+++++++++++++++++++++++++++
+
+PyDataProvider使用的是异步加载，同时在内存里直接随即选取数据来做Shuffle。即
+
+..  graphviz::
+
+    digraph {
+        rankdir=LR;
+        数据文件 -> 内存池 -> PaddlePaddle训练
+    }
+
+所以，减小这个内存池即可减小内存占用，同时也可以加速开始训练前数据载入的过程。但是，这
+个内存池实际上决定了shuffle的粒度。所以，如果将这个内存池减小，又要保证数据是随机的，
+那么最好将数据文件在每次读取之前做一次shuffle。可能的代码为
+
+..  literalinclude:: src/reduce_min_pool_size.py
+
+这样做可以极大的减少内存占用，并且可能会加速训练过程，详细文档参考 :ref:`api_pydataprovider2` 。
+
+神经元激活内存
+++++++++++++++
+
+神经网络在训练的时候，会对每一个激活暂存一些数据，如神经元激活值等。
+在反向传递的时候，这些数据会被用来更新参数。这些数据使用的内存主要和两个参数有关系，
+一是batch size，另一个是每条序列(Sequence)长度。所以，其实也是和每个mini-batch中包含
+的时间步信息成正比。
+
+所以做法可以有两种：
+
+* 减小batch size。 即在网络配置中 :code:`settings(batch_size=1000)` 设置成一个小一些的值。但是batch size本身是神经网络的超参数，减小batch size可能会对训练结果产生影响。
+* 减小序列的长度，或者直接扔掉非常长的序列。比如，一个数据集大部分序列长度是100-200,
+  但是突然有一个10000长的序列，就很容易导致内存超限，特别是在LSTM等RNN中。
+
+参数内存
+++++++++
+
+PaddlePaddle支持非常多的优化算法(Optimizer)，不同的优化算法需要使用不同大小的内存。
+例如使用 :code:`adadelta` 算法，则需要使用等于权重参数规模大约5倍的内存。举例，如果参数保存下来的模型目录
+文件为 :code:`100M`， 那么该优化算法至少需要 :code:`500M` 的内存。
+
+可以考虑使用一些优化算法，例如 :code:`momentum`。
+
+2. 如何加速训练速度
+-------------------
+
+加速PaddlePaddle训练可以考虑从以下几个方面\：
+
+* 减少数据载入的耗时
+* 加速训练速度
+* 利用分布式训练驾驭更多的计算资源
+
+减少数据载入的耗时
+++++++++++++++++++
+
+使用\ :code:`pydataprovider`\ 时，可以减少缓存池的大小，同时设置内存缓存功能，即可以极大的加速数据载入流程。
+:code:`DataProvider` 缓存池的减小，和之前减小通过减小缓存池来减小内存占用的原理一致。
+
+..  literalinclude:: src/reduce_min_pool_size.py
+
+同时 :code:`@provider` 接口有一个 :code:`cache` 参数来控制缓存方法，将其设置成 :code:`CacheType.CACHE_PASS_IN_MEM` 的话，会将第一个 :code:`pass` (过完所有训练数据即为一个pass)生成的数据缓存在内存里，在之后的 :code:`pass` 中，不会再从 :code:`python` 端读取数据，而是直接从内存的缓存里读取数据。这也会极大减少数据读入的耗时。
+
+
+加速训练速度
+++++++++++++
+
+PaddlePaddle支持Sparse的训练，sparse训练需要训练特征是 :code:`sparse_binary_vector` 、 :code:`sparse_vector` 、或者 :code:`integer_value` 的任一一种。同时，与这个训练数据交互的Layer，需要将其Parameter设置成 sparse 更新模式，即设置 :code:`sparse_update=True`
+
+这里使用简单的 :code:`word2vec` 训练语言模型距离，具体使用方法为\:
+
+使用一个词前两个词和后两个词，来预测这个中间的词。这个任务的DataProvider为\:
+
+..  literalinclude:: src/word2vec_dataprovider.py
+
+这个任务的配置为\:
+
+..  literalinclude:: src/word2vec_config.py
+
+
+利用更多的计算资源
+++++++++++++++++++
+
+利用更多的计算资源可以分为一下几个方式来进行\:
+
+* 单机CPU训练
+
+  * 使用多线程训练。设置命令行参数 :code:`trainer_count`。
+
+* 单机GPU训练
+
+  * 使用显卡训练。设置命令行参数 :code:`use_gpu`。
+  * 使用多块显卡训练。设置命令行参数 :code:`use_gpu` 和 :code:`trainer_count` 。
+
+* 多机训练
+
+  * 请参考 :ref:`cluster_train` 。
+
+3. 如何指定GPU设备
+------------------
+
+例如机器上有4块GPU，编号从0开始，指定使用2、3号GPU：
+
+* 方式1：通过 `CUDA_VISIBLE_DEVICES <http://www.acceleware.com/blog/cudavisibledevices-masking-gpus>`_ 环境变量来指定特定的GPU。
+
+..      code-block:: bash
+
+        env CUDA_VISIBLE_DEVICES=2,3 paddle train --use_gpu=true --trainer_count=2
+
+* 方式2：通过命令行参数 ``--gpu_id`` 指定。
+
+..      code-block:: bash
+
+        paddle train --use_gpu=true --trainer_count=2 --gpu_id=2
+
+
+4. 训练过程中出现 :code:`Floating point exception`, 训练因此退出怎么办?
+------------------------------------------------------------------------
+
+Paddle二进制在运行时捕获了浮点数异常，只要出现浮点数异常(即训练过程中出现NaN或者Inf)，立刻退出。浮点异常通常的原因是浮点数溢出、除零等问题。
+主要原因包括两个方面:
+
+* 训练过程中参数或者训练过程中的梯度尺度过大，导致参数累加，乘除等时候，导致了浮点数溢出。
+* 模型一直不收敛，发散到了一个数值特别大的地方。
+* 训练数据有问题，导致参数收敛到了一些奇异的情况。或者输入数据尺度过大，有些特征的取值达到数百万，这时进行矩阵乘法运算就可能导致浮点数溢出。
+
+这里有两种有效的解决方法：
+
+1. 设置 :code:`gradient_clipping_threshold` 参数，示例代码如下：
+
+..  code-block:: python
+
+optimizer = paddle.optimizer.RMSProp(
+    learning_rate=1e-3,
+    gradient_clipping_threshold=10.0,
+    regularization=paddle.optimizer.L2Regularization(rate=8e-4))
+
+具体可以参考  `nmt_without_attention  <https://github.com/PaddlePaddle/models/blob/develop/nmt_without_attention/train.py#L35>`_ 示例。
+
+2. 设置 :code:`error_clipping_threshold` 参数，示例代码如下：
+
+..  code-block:: python
+
+decoder_inputs = paddle.layer.fc(
+    act=paddle.activation.Linear(),
+    size=decoder_size * 3,
+    bias_attr=False,
+    input=[context, current_word],
+    layer_attr=paddle.attr.ExtraLayerAttribute(
+        error_clipping_threshold=100.0))
+
+完整代码可以参考示例 `machine translation <https://github.com/PaddlePaddle/book/blob/develop/08.machine_translation/train.py#L66>`_ 。
+
+两种方法的区别：
+
+1. 两者都是对梯度的截断，但截断时机不同，前者在 :code:`optimzier` 更新网络参数时应用；后者在激活函数反向计算时被调用；
+2. 截断对象不同：前者截断可学习参数的梯度，后者截断回传给前层的梯度;
+
+除此之外，还可以通过减小学习律或者对数据进行归一化处理来解决这类问题。
+
+5.  如何调用 infer 接口输出多个layer的预测结果
+-----------------------------------------------
+
+* 将需要输出的层作为 :code:`paddle.inference.Inference()` 接口的 :code:`output_layer` 参数输入，代码如下：
+
+..  code-block:: python
+
+    inferer = paddle.inference.Inference(output_layer=[layer1, layer2], parameters=parameters)
+
+* 指定要输出的字段进行输出。以输出 :code:`value` 字段为例，代码如下：
+
+..  code-block:: python
+
+    out = inferer.infer(input=data_batch, field=["value"])
+
+需要注意的是：
+
+* 如果指定了2个layer作为输出层，实际上需要的输出结果是两个矩阵；
+* 假设第一个layer的输出A是一个 N1 * M1 的矩阵，第二个 Layer 的输出B是一个 N2 * M2 的矩阵；
+* paddle.v2 默认会将A和B 横向拼接，当N1 和 N2 大小不一样时，会报如下的错误：
+
+..      code-block:: python
+
+    ValueError: all the input array dimensions except for the concatenation axis must match exactly
+
+多个层的输出矩阵的高度不一致导致拼接失败，这种情况常常发生在：
+
+* 同时输出序列层和非序列层；
+* 多个输出层处理多个不同长度的序列;
+
+此时可以在调用infer接口时通过设置 :code:`flatten_result=False` , 跳过“拼接”步骤，来解决上面的问题。这时，infer接口的返回值是一个python list:
+
+* list 中元素的个数等于网络中输出层的个数；
+* list 中每个元素是一个layer的输出结果矩阵，类型是numpy的ndarray；
+* 每一个layer输出矩阵的高度，在非序列输入时：等于样本数；序列输入时等于：输入序列中元素的总数；宽度等于配置中layer的size；
diff --git a/develop/doc_cn/_sources/faq/model/index_cn.rst.txt b/develop/doc_cn/_sources/faq/model/index_cn.rst.txt
new file mode 100644
index 0000000000000000000000000000000000000000..b47bbe05bdb39d1ade9434a7e54bf6ca88a91cc9
--- /dev/null
+++ b/develop/doc_cn/_sources/faq/model/index_cn.rst.txt
@@ -0,0 +1,69 @@
+#########
+模型配置
+#########
+
+..  contents::
+
+1. 出现 :code:`Duplicated layer name` 错误怎么办
+--------------------------------------------------
+
+出现该错误的原因一般是用户对不同layer的参数 :code:`name` 设置了相同的取值。遇到该错误时，先找出参数 :code:`name` 取值相同的layer，然后将这些layer的参数 :code:`name` 设置为不同的值。
+
+2. :code:`paddle.layer.memory` 的参数 :code:`name` 如何使用
+-------------------------------------------------------------
+
+* :code:`paddle.layer.memory` 用于获取特定layer上一时间步的输出，该layer是通过参数 :code:`name` 指定，即，:code:`paddle.layer.memory` 会关联参数 :code:`name` 取值相同的layer，并将该layer上一时间步的输出作为自身当前时间步的输出。
+
+* PaddlePaddle的所有layer都有唯一的name，用户通过参数 :code:`name` 设定，当用户没有显式设定时，PaddlePaddle会自动设定。而 :code:`paddle.layer.memory` 不是真正的layer，其name由参数 :code:`memory_name` 设定，当用户没有显式设定时，PaddlePaddle会自动设定。:code:`paddle.layer.memory` 的参数 :code:`name` 用于指定其要关联的layer，需要用户显式设定。
+
+3. 两种使用 drop_out 的方法有何区别
+------------------------------------
+
+* 在PaddlePaddle中使用dropout有两种方式
+
+  * 在相应layer的 :code:`layer_atter` 设置 :code:`drop_rate`，以 :code:`paddle.layer.fc` 为例，代码如下：
+
+  ..  code-block:: python
+
+      fc = paddle.layer.fc(input=input, layer_attr=paddle.attr.ExtraLayerAttribute(drop_rate=0.5))
+
+  * 使用 :code:`paddle.layer.dropout`，以 :code:`paddle.layer.fc` 为例，代码如下：
+
+  ..  code-block:: python
+
+      fc = paddle.layer.fc(input=input)
+      drop_fc = paddle.layer.dropout(input=fc, dropout_rate=0.5)
+
+* :code:`paddle.layer.dropout` 实际上使用了 :code:`paddle.layer.add_to`，并在该layer里采用第一种方式设置 :code:`drop_rate` 来使用dropout的。这种方式对内存消耗较大。
+
+* PaddlePaddle在激活函数里实现dropout，而不是在layer里实现。
+
+* :code:`paddle.layer.lstmemory`、:code:`paddle.layer.grumemory`、:code:`paddle.layer.recurrent` 不是通过一般的方式来实现对输出的激活，所以不能采用第一种方式在这几个layer里设置 :code:`drop_rate` 来使用dropout。若要对这几个layer使用dropout，可采用第二种方式，即使用 :code:`paddle.layer.dropout`。
+
+4. 不同的 recurrent layer 的区别
+----------------------------------
+以LSTM为例，在PaddlePaddle中包含以下 recurrent layer：
+
+* :code:`paddle.layer.lstmemory`
+* :code:`paddle.networks.simple_lstm`
+* :code:`paddle.networks.lstmemory_group`
+* :code:`paddle.networks.bidirectional_lstm`
+
+按照具体实现方式可以归纳为2类：
+
+1. 由 recurrent_group 实现的 recurrent layer：
+
+  * 用户在使用这一类recurrent layer时，可以访问由recurrent unit在一个时间步内计算得到的中间值（例如：hidden states, memory cells等）；
+  * 上述的 :code:`paddle.networks.lstmemory_group` 是这一类的 recurrent layer ；
+
+2. 将recurrent layer作为一个整体来实现：
+
+  * 用户在使用这一类recurrent layer，只能访问它们的输出值；
+  * 上述的 :code:`paddle.networks.lstmemory_group` 、 :code:`paddle.networks.simple_lstm` 和 :code:`paddle.networks.bidirectional_lstm` 属于这一类的实现；
+
+将recurrent layer作为一个整体来实现， 能够针对CPU和GPU的计算做更多优化， 所以相比于recurrent group的实现方式， 第二类 recurrent layer 计算效率更高。 在实际应用中，如果用户不需要访问LSTM的中间变量，而只需要获得recurrent layer计算的输出，我们建议使用第二类实现。
+
+此外，关于LSTM, PaddlePaddle中还包含 :code:`paddle.networks.lstmemory_unit` 这一计算单元：
+
+  * 不同于上述介绍的recurrent layer , :code:`paddle.networks.lstmemory_unit` 定义了LSTM单元在一个时间步内的计算过程，它并不是一个完整的recurrent layer，也不能接收序列数据作为输入；
+  * :code:`paddle.networks.lstmemory_unit` 只能在recurrent_group中作为step function使用；
diff --git a/develop/doc_cn/_sources/faq/parameter/index_cn.rst.txt b/develop/doc_cn/_sources/faq/parameter/index_cn.rst.txt
new file mode 100644
index 0000000000000000000000000000000000000000..c721b623183cc7d8d17e2c9fb1635ea07b8970cc
--- /dev/null
+++ b/develop/doc_cn/_sources/faq/parameter/index_cn.rst.txt
@@ -0,0 +1,201 @@
+#########
+参数设置
+#########
+
+..  contents::
+
+1. 如何选择SGD算法的学习率
+--------------------------
+
+在采用sgd/async_sgd进行训练时，一个重要的问题是选择正确的learning_rate。如果learning_rate太大，那么训练有可能不收敛，如果learning_rate太小，那么收敛可能很慢，导致训练时间过长。
+
+通常做法是从一个比较大的learning_rate开始试，如果不收敛，那减少学习率10倍继续试验，直到训练收敛为止。那么如何判断训练不收敛呢？可以估计出如果模型采用不变的输出最小的cost0是多少。
+
+如果训练过程的的cost明显高于这个常数输出的cost，那么我们可以判断为训练不收敛。举一个例子，假如我们是三分类问题，采用multi-class-cross-entropy作为cost，数据中0,1,2三类的比例为 :code:`0.2, 0.5, 0.3` , 那么常数输出所能达到的最小cost是 :code:`-(0.2*log(0.2)+0.5*log(0.5)+0.3*log(0.3))=1.03` 。如果训练一个pass（或者更早）后，cost还大于这个数，那么可以认为训练不收敛，应该降低学习率。
+
+2. 如何设置学习率退火（learning rate annealing）
+------------------------------------------------
+
+在相应的优化算法里设置learning_rate_schedule及相关参数，以使用Adam算法为例，代码如下：
+
+..  code-block:: python
+
+    optimizer = paddle.optimizer.Adam(
+        learning_rate=1e-3,
+        learning_rate_decay_a=0.5,
+        learning_rate_decay_b=0.75,
+        learning_rate_schedule="poly",)
+
+PaddlePaddle目前支持8种learning_rate_schedule，这8种learning_rate_schedule及其对应学习率计算方式如下：
+
+* "constant"
+  
+  lr = learning_rate
+
+* "poly"
+
+  lr = learning_rate * pow(1 + learning_rate_decay_a * num_samples_processed, -learning_rate_decay_b)
+
+  其中，num_samples_processed为已训练样本数，下同。
+
+* "caffe_poly"
+
+  lr = learning_rate * pow(1.0 - num_samples_processed / learning_rate_decay_a, learning_rate_decay_b)
+
+* "exp"
+
+  lr = learning_rate * pow(learning_rate_decay_a, num_samples_processed / learning_rate_decay_b)
+
+* "discexp"
+
+  lr = learning_rate * pow(learning_rate_decay_a, floor(num_samples_processed / learning_rate_decay_b))
+
+* "linear"
+
+  lr = max(learning_rate - learning_rate_decay_a * num_samples_processed, learning_rate_decay_b)
+
+* "manual"
+
+  这是一种按已训练样本数分段取值的学习率退火方法。使用该learning_rate_schedule时，用户通过参数 :code:`learning_rate_args` 设置学习率衰减因子分段函数，当前的学习率为所设置 :code:`learning_rate` 与当前的衰减因子的乘积。以使用Adam算法为例，代码如下：
+
+  ..  code-block:: python
+
+      optimizer = paddle.optimizer.Adam(
+          learning_rate=1e-3,
+          learning_rate_schedule="manual",
+          learning_rate_args="1000:1.0,2000:0.9,3000:0.8",)
+
+  在该示例中，当已训练样本数小于等于1000时，学习率为 :code:`1e-3 * 1.0`；当已训练样本数大于1000小于等于2000时，学习率为 :code:`1e-3 * 0.9`；当已训练样本数大于2000时，学习率为 :code:`1e-3 * 0.8`。
+
+* "pass_manual"
+
+  这是一种按已训练pass数分段取值的学习率退火方法。使用该learning_rate_schedule时，用户通过参数 :code:`learning_rate_args` 设置学习率衰减因子分段函数，当前的学习率为所设置 :code:`learning_rate` 与当前的衰减因子的乘积。以使用Adam算法为例，代码如下：
+
+  ..  code-block:: python
+
+      optimizer = paddle.optimizer.Adam(
+          learning_rate=1e-3,
+          learning_rate_schedule="manual",
+          learning_rate_args="1:1.0,2:0.9,3:0.8",)
+
+  在该示例中，当已训练pass数小于等于1时，学习率为 :code:`1e-3 * 1.0`；当已训练pass数大于1小于等于2时，学习率为 :code:`1e-3 * 0.9`；当已训练pass数大于2时，学习率为 :code:`1e-3 * 0.8`。
+
+3. 如何初始化参数
+-----------------
+
+默认情况下，PaddlePaddle使用均值0，标准差为 :math:`\frac{1}{\sqrt{d}}` 来初始化参数。其中 :math:`d` 为参数矩阵的宽度。这种初始化方式在一般情况下不会产生很差的结果。如果用户想要自定义初始化方式，PaddlePaddle目前提供两种参数初始化的方式\:
+
+* 高斯分布。将 :code:`param_attr` 设置成 :code:`param_attr=ParamAttr(initial_mean=0.0, initial_std=1.0)`
+* 均匀分布。将 :code:`param_attr` 设置成 :code:`param_attr=ParamAttr(initial_max=1.0, initial_min=-1.0)`
+
+比如设置一个全连接层的参数初始化方式和bias初始化方式，可以使用如下代码。
+
+..  code-block:: python
+
+    hidden = fc_layer(input=ipt, param_attr=ParamAttr(initial_max=1.0, initial_min=-1.0),
+                      bias_attr=ParamAttr(initial_mean=1.0, initial_std=0.0))
+
+上述代码将bias全部初始化为1.0, 同时将参数初始化为 :code:`[1.0, -1.0]` 的均匀分布。
+
+4. 如何共享参数
+---------------
+
+PaddlePaddle的参数使用名字 :code:`name` 作为参数的ID，相同名字的参数，会共享参数。设置参数的名字，可以使用 :code:`ParamAttr(name="YOUR_PARAM_NAME")` 来设置。更方便的设置方式，是使得要共享的参数使用同样的 :code:`ParamAttr` 对象。
+
+简单的全连接网络，参数共享的配置示例为\:
+
+..  literalinclude:: ../../python/paddle/trainer_config_helpers/tests/configs/shared_fc.py
+
+这里 :code:`hidden_a` 和 :code:`hidden_b` 使用了同样的parameter和bias。并且softmax层的两个输入也使用了同样的参数 :code:`softmax_param`。
+
+5. 如何加载预训练参数
+------------------------
+
+* 对加载预训练参数的层，设置其参数属性 :code:`is_static=True`，使该层的参数在训练过程中保持不变。以embedding层为例，代码如下：
+
+..  code-block:: python
+
+    emb_para = paddle.attr.Param(name='emb', is_static=True)
+    paddle.layer.embedding(size=word_dim, input=x, param_attr=emb_para)
+
+
+* 从模型文件将预训练参数载入 :code:`numpy.array`，在创建parameters后，使用 :code:`parameters.set()` 加载预训练参数。PaddlePaddle保存的模型参数文件前16字节为头信息，用户将参数载入 :code:`numpy.array` 时须从第17字节开始。以embedding层为例，代码如下：
+
+..  code-block:: python
+
+    def load_parameter(file_name, h, w):
+        with open(file_name, 'rb') as f:
+            f.read(16)  # skip header.
+            return np.fromfile(f, dtype=np.float32).reshape(h, w)
+
+    parameters = paddle.parameters.create(my_cost)
+    parameters.set('emb', load_parameter(emb_param_file, 30000, 256))
+
+6. 存储的参数格式是什么，如何和明文进行相互转化
+--------------------------------------------------
+
+PaddlePaddle保存的模型参数文件内容由16字节头信息和网络参数两部分组成。头信息中，1~4字节表示PaddlePaddle版本信息，请直接填充0；5~8字节表示每个参数占用的字节数，当保存的网络参数为float类型时为4，double类型时为8；9~16字节表示保存的参数总个数。
+
+将PaddlePaddle保存的模型参数还原回明文时，可以使用相应数据类型的 :code:`numpy.array` 加载具体网络参数，此时可以跳过PaddlePaddle模型参数文件的头信息。若在PaddlePaddle编译时，未指定按照double精度编译，默认情况下按照float精度计算，保存的参数也是float类型。这时在使用 :code:`numpy.array` 时，一般设置 :code:`dtype=float32` 。示例如下：
+
+..  code-block:: python
+
+    def read_parameter(fname, width):
+        s = open(fname).read()
+        # skip header
+        vec = np.fromstring(s[16:], dtype=np.float32)
+        # width is the size of the corresponding layer
+        np.savetxt(fname + ".csv", vec.reshape(width, -1),
+                fmt="%.6f", delimiter=",")
+
+
+将明文参数转化为PaddlePaddle可加载的模型参数时，首先构造头信息，再写入网络参数。下面的代码将随机生成的矩阵转化为可以被PaddlePaddle加载的模型参数。
+
+..  code-block:: python
+
+    def gen_rand_param(param_file, width, height, need_trans):
+        np.random.seed()
+        header = struct.pack("iil", 0, 4, height * width)
+        param = np.float32(np.random.rand(height, width))
+        with open(param_file, "w") as fparam:
+            fparam.write(header + param.tostring())
+
+7. A protocol message was rejected because it was too big
+------------------------------------------------------------
+
+如果在训练NLP相关模型时，出现以下错误：
+
+..  code-block:: bash
+
+    [libprotobuf ERROR google/protobuf/io/coded_stream.cc:171] A protocol message was rejected because it was too big (more than 67108864 bytes).  To increase the limit (or to disable these warnings), see CodedInputStream::SetTotalBytesLimit() in google/protobuf/io/coded_stream.h.
+    F1205 14:59:50.295174 14703 TrainerConfigHelper.cpp:59] Check failed: m->conf.ParseFromString(configProtoStr)
+
+可能的原因是：传给dataprovider的某一个args过大，一般是由于直接传递大字典导致的。错误的define_py_data_sources2类似：
+
+..  code-block:: python
+
+     src_dict = dict()
+     for line_count, line in enumerate(open(src_dict_path, "r")):
+        src_dict[line.strip()] = line_count
+
+     define_py_data_sources2(
+        train_list,
+        test_list,
+        module="dataprovider",
+        obj="process",
+        args={"src_dict": src_dict})
+
+解决方案是：将字典的地址作为args传给dataprovider，然后在dataprovider里面根据该地址加载字典。即define_py_data_sources2应改为：
+
+..  code-block:: python
+
+     define_py_data_sources2(
+        train_list,
+        test_list,
+        module="dataprovider",
+        obj="process",
+        args={"src_dict_path": src_dict_path})
+
+完整源码可参考 `seqToseq <https://github.com/PaddlePaddle/Paddle/tree/develop/demo/seqToseq>`_ 示例。
+
+
diff --git a/develop/doc_cn/api/index_cn.html b/develop/doc_cn/api/index_cn.html
index 04bd5ba33e878e1e7bf6d28a4c32483d0097b36e..95ed3ed45e222b82a47559129a9d493993ddd788 100644
--- a/develop/doc_cn/api/index_cn.html
+++ b/develop/doc_cn/api/index_cn.html
@@ -156,7 +156,14 @@
 <li class="toctree-l2"><a class="reference internal" href="v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v1/data_provider/dataprovider_cn.html b/develop/doc_cn/api/v1/data_provider/dataprovider_cn.html
index 35b94a107fb7d8d65f146ee64ae5e8069d015fde..115142fbe70cc15db22f2a1bb93e6a307b2f070f 100644
--- a/develop/doc_cn/api/v1/data_provider/dataprovider_cn.html
+++ b/develop/doc_cn/api/v1/data_provider/dataprovider_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v1/data_provider/pydataprovider2_cn.html b/develop/doc_cn/api/v1/data_provider/pydataprovider2_cn.html
index 310dfc781aa097707929f2468e9a4a70c9e5759f..8f529b15c483b370e386a7637cabc8070df92a7c 100644
--- a/develop/doc_cn/api/v1/data_provider/pydataprovider2_cn.html
+++ b/develop/doc_cn/api/v1/data_provider/pydataprovider2_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v1/index_cn.html b/develop/doc_cn/api/v1/index_cn.html
index 53fbf68088ddfad0691d6073976f27379265af53..39943ec979b0e9632a645cd303e8b636b7930ed2 100644
--- a/develop/doc_cn/api/v1/index_cn.html
+++ b/develop/doc_cn/api/v1/index_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v1/predict/swig_py_paddle_cn.html b/develop/doc_cn/api/v1/predict/swig_py_paddle_cn.html
index 03e33f2e0653b4651f36709dc85c394d71d87290..ea0ac5b1ea2997d9597022d544375c0c85b50cf0 100644
--- a/develop/doc_cn/api/v1/predict/swig_py_paddle_cn.html
+++ b/develop/doc_cn/api/v1/predict/swig_py_paddle_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v2/config/activation.html b/develop/doc_cn/api/v2/config/activation.html
index e38d5583a5b01aee15cbc1c422c9c9dd51a76435..68546fda773921d64e13b32b2a7a9037d522eda3 100644
--- a/develop/doc_cn/api/v2/config/activation.html
+++ b/develop/doc_cn/api/v2/config/activation.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v2/config/attr.html b/develop/doc_cn/api/v2/config/attr.html
index b7394464301b6e382dbe727cd16bc74853e85302..c7efed0864feef18c886a87bb43cb62ca6281d60 100644
--- a/develop/doc_cn/api/v2/config/attr.html
+++ b/develop/doc_cn/api/v2/config/attr.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v2/config/evaluators.html b/develop/doc_cn/api/v2/config/evaluators.html
index d397f66a3dde9fb558129ead5661917bdfdceb20..5ebc6018c3d1e183d06f069b55186a6bac5b9bf3 100644
--- a/develop/doc_cn/api/v2/config/evaluators.html
+++ b/develop/doc_cn/api/v2/config/evaluators.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v2/config/layer.html b/develop/doc_cn/api/v2/config/layer.html
index 1a33aae0f5dc736002ac51a3cd96b4665c219e49..c472d1e9cc3f1a01e9d171fd1688027f7858cebe 100644
--- a/develop/doc_cn/api/v2/config/layer.html
+++ b/develop/doc_cn/api/v2/config/layer.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v2/config/networks.html b/develop/doc_cn/api/v2/config/networks.html
index f4840621cc374f68edf297307167183b6d672d2e..0f9cc772eb69d67e9fc1be370415ad3c01a10a28 100644
--- a/develop/doc_cn/api/v2/config/networks.html
+++ b/develop/doc_cn/api/v2/config/networks.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v2/config/optimizer.html b/develop/doc_cn/api/v2/config/optimizer.html
index 343d2c8ac6eecb0af858b75ea88fecb9fd756166..b5ee1d7d7020faf8e338c54e054fa4824f2dd8f8 100644
--- a/develop/doc_cn/api/v2/config/optimizer.html
+++ b/develop/doc_cn/api/v2/config/optimizer.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v2/config/pooling.html b/develop/doc_cn/api/v2/config/pooling.html
index f24c693579dc16fff54c7efb28bdca780403b9b1..4cbc2f5a6c26ea18f320865dd3d56f417eecf0c1 100644
--- a/develop/doc_cn/api/v2/config/pooling.html
+++ b/develop/doc_cn/api/v2/config/pooling.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v2/data.html b/develop/doc_cn/api/v2/data.html
index 5c0de416e979d2c88c8b920b435a81c07ce86eae..5e7653c2a790d831d1ec38d68fe56de364f0f644 100644
--- a/develop/doc_cn/api/v2/data.html
+++ b/develop/doc_cn/api/v2/data.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v2/model_configs.html b/develop/doc_cn/api/v2/model_configs.html
index d5d2e60978744bfa7ece0902b483177609fcdabc..9ea7de5144ca3f3036126b924d2d8fbd3b78b0f9 100644
--- a/develop/doc_cn/api/v2/model_configs.html
+++ b/develop/doc_cn/api/v2/model_configs.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/api/v2/run_logic.html b/develop/doc_cn/api/v2/run_logic.html
index 9d678618e01f258db8eb737c9c5f445d74fae4fa..f8221806dc0fd6fd863743f332c5d7d957754f29 100644
--- a/develop/doc_cn/api/v2/run_logic.html
+++ b/develop/doc_cn/api/v2/run_logic.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2 current"><a class="current reference internal" href="#">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/api.html b/develop/doc_cn/design/api.html
index 300159f2421e12cc69e1962f014325da564612f7..d66f3efea1926d3cac43a0ac89bfa947e3d1fc7e 100644
--- a/develop/doc_cn/design/api.html
+++ b/develop/doc_cn/design/api.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/auto_gradient_check.html b/develop/doc_cn/design/auto_gradient_check.html
index c35c9efdfd4d63cb0204f6bfaa121b5f34ee08ec..24a9adcea9ab0e104f2833e7902864b3d5b73cab 100644
--- a/develop/doc_cn/design/auto_gradient_check.html
+++ b/develop/doc_cn/design/auto_gradient_check.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/block.html b/develop/doc_cn/design/block.html
index d02471ac45d6e20b8ffea853ff5a00e1af04c2c1..3e51f210be935e5031d443c0028e2506f2eeab92 100644
--- a/develop/doc_cn/design/block.html
+++ b/develop/doc_cn/design/block.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/build_system/README.html b/develop/doc_cn/design/build_system/README.html
index a845c9953e1d0bdb5ef4ce371dd9ac18392d26c0..d3a291dd304d37e8e9b87fb4a6daaed987bfe643 100644
--- a/develop/doc_cn/design/build_system/README.html
+++ b/develop/doc_cn/design/build_system/README.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/cluster_train/README.html b/develop/doc_cn/design/cluster_train/README.html
index 6fa31f9969471603da6ea14a7951f7fae661162a..a3ee8b924fb8bd55e257395864fcb5078da88f44 100644
--- a/develop/doc_cn/design/cluster_train/README.html
+++ b/develop/doc_cn/design/cluster_train/README.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/cluster_train/checkpointing.html b/develop/doc_cn/design/cluster_train/checkpointing.html
index 0f01f170e421c29fdeebff72254d04a0516b16b9..b8647cb424479f178767e54c0f6a6049b2ad3e96 100644
--- a/develop/doc_cn/design/cluster_train/checkpointing.html
+++ b/develop/doc_cn/design/cluster_train/checkpointing.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/cluster_train/data_dispatch.html b/develop/doc_cn/design/cluster_train/data_dispatch.html
index 4a442d5ce2ce7edbb1100da0004330307babe2cd..608dcd89e1871c702836b360455f2b48d001c10f 100644
--- a/develop/doc_cn/design/cluster_train/data_dispatch.html
+++ b/develop/doc_cn/design/cluster_train/data_dispatch.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/cluster_train/large_model_dist_train.html b/develop/doc_cn/design/cluster_train/large_model_dist_train.html
index e4e6e41341ef24ce7124351a9ace2c31ae97ff57..6e6ef377ffe433104be5327449ae8d4f33005d51 100644
--- a/develop/doc_cn/design/cluster_train/large_model_dist_train.html
+++ b/develop/doc_cn/design/cluster_train/large_model_dist_train.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/cluster_train/master_server.html b/develop/doc_cn/design/cluster_train/master_server.html
index d663bed86debb38f442c1f9e773a7a89ba73be05..fefc2905cde614ad389387b411a6f3ea24699010 100644
--- a/develop/doc_cn/design/cluster_train/master_server.html
+++ b/develop/doc_cn/design/cluster_train/master_server.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/cluster_train/pserver_client.html b/develop/doc_cn/design/cluster_train/pserver_client.html
index ac478d37fc134c29a309c72e2db2f98dd9bef2d4..e83b2d82f789d9d683941256b4473e1213de5313 100644
--- a/develop/doc_cn/design/cluster_train/pserver_client.html
+++ b/develop/doc_cn/design/cluster_train/pserver_client.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/cluster_train/remote_parameter_updater.html b/develop/doc_cn/design/cluster_train/remote_parameter_updater.html
index e67718afab110b695d6a2b8e44c6c8ba1d7a4e3a..856b236096a4c97431b3da8e8ebc139abd62385f 100644
--- a/develop/doc_cn/design/cluster_train/remote_parameter_updater.html
+++ b/develop/doc_cn/design/cluster_train/remote_parameter_updater.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/cluster_train/save_model.html b/develop/doc_cn/design/cluster_train/save_model.html
index be9dbd7aacfc7c7346a85d9d98e8d5a1d44ee318..53eed8b52970655c0fc544998fc5a7816819dbeb 100644
--- a/develop/doc_cn/design/cluster_train/save_model.html
+++ b/develop/doc_cn/design/cluster_train/save_model.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/cluster_train/submit-job.html b/develop/doc_cn/design/cluster_train/submit-job.html
index 67e6297587a6e7631b6b2e053d5018a47e94cbff..eccb0bb86f66d58a02ff05b9896d1d513e317dfd 100644
--- a/develop/doc_cn/design/cluster_train/submit-job.html
+++ b/develop/doc_cn/design/cluster_train/submit-job.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/file_manager/README.html b/develop/doc_cn/design/file_manager/README.html
index 990adfd2e58906a134dd7b296d028fd3265d609e..ef4fdd26034b38a4c352a458a949daa8e8079c8d 100644
--- a/develop/doc_cn/design/file_manager/README.html
+++ b/develop/doc_cn/design/file_manager/README.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/file_manager/pfs/pfsclient.html b/develop/doc_cn/design/file_manager/pfs/pfsclient.html
index 9e461a2303f7c881bf41a8012ce2f13229336e83..df81272220e811bcc6c51da7b43ec8e5569f42e5 100644
--- a/develop/doc_cn/design/file_manager/pfs/pfsclient.html
+++ b/develop/doc_cn/design/file_manager/pfs/pfsclient.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/functions_operators_layers.html b/develop/doc_cn/design/functions_operators_layers.html
index 18af552ac45eca2895350e64266a56f342b0fdd7..054199c723d7027c73f5bfd5c79327e1551c9c21 100644
--- a/develop/doc_cn/design/functions_operators_layers.html
+++ b/develop/doc_cn/design/functions_operators_layers.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/graph.html b/develop/doc_cn/design/graph.html
index d6f8a3a45b872b73469ffd04ea752de12b35c61a..5a0b45df427526b7544b0e4b06e0be76a97eaac9 100644
--- a/develop/doc_cn/design/graph.html
+++ b/develop/doc_cn/design/graph.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/if_else_op.html b/develop/doc_cn/design/if_else_op.html
index f9d4e7f78833143fee5400b77791279d273ff91f..31f91bc58e9b10ca76741a4a92fbf19f0a02fd10 100644
--- a/develop/doc_cn/design/if_else_op.html
+++ b/develop/doc_cn/design/if_else_op.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/multi_language_interface/00.why_plain_c.html b/develop/doc_cn/design/multi_language_interface/00.why_plain_c.html
index c5f9c74d07a648b3b858b1f7c2d888359bc56782..6c98aad738c695ae4c4a3942769b154a82b7dff7 100644
--- a/develop/doc_cn/design/multi_language_interface/00.why_plain_c.html
+++ b/develop/doc_cn/design/multi_language_interface/00.why_plain_c.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/multi_language_interface/01.inference_implementation.html b/develop/doc_cn/design/multi_language_interface/01.inference_implementation.html
index c8adada796fd438cf459f48227ea374881db7d28..5e2b78f9e6674b037eda8b5e6ac8598c54f5365d 100644
--- a/develop/doc_cn/design/multi_language_interface/01.inference_implementation.html
+++ b/develop/doc_cn/design/multi_language_interface/01.inference_implementation.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/ops/rnn.html b/develop/doc_cn/design/ops/rnn.html
index a9a7236b53d8a2d4f804e8002af86f658a787293..32dd34b558e4109f9ae3aac3aa258f9292c29738 100644
--- a/develop/doc_cn/design/ops/rnn.html
+++ b/develop/doc_cn/design/ops/rnn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/parameters_in_cpp.html b/develop/doc_cn/design/parameters_in_cpp.html
index e97deb2372bcde7b5abfc0c51cab491a74892a6b..b365c9f402e389f8cc726d027d0f2fb0864035e5 100644
--- a/develop/doc_cn/design/parameters_in_cpp.html
+++ b/develop/doc_cn/design/parameters_in_cpp.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/program.html b/develop/doc_cn/design/program.html
index aafac5763f031cc7d0aa7251f108f9327f25be98..8cb4cc7878c1f6b2600a3d4b6cd52570d52dbd8d 100644
--- a/develop/doc_cn/design/program.html
+++ b/develop/doc_cn/design/program.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/reader/README.html b/develop/doc_cn/design/reader/README.html
index eeb9e2b1f43052360d9b145be352e6224f05987b..76f677e6f9c8c258de51d707efc2ef15f5bda454 100644
--- a/develop/doc_cn/design/reader/README.html
+++ b/develop/doc_cn/design/reader/README.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/refactor/distributed_architecture.html b/develop/doc_cn/design/refactor/distributed_architecture.html
index c9b4e0ec4ae7e4eea239601b7a39cd80959bd5ac..f2835550e0a308f0169d4bea1d8a0105c185f1e9 100644
--- a/develop/doc_cn/design/refactor/distributed_architecture.html
+++ b/develop/doc_cn/design/refactor/distributed_architecture.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/refactor/parameter_server.html b/develop/doc_cn/design/refactor/parameter_server.html
index 7adfce38275d0b4d8b4d61f9bd5c24a77a1bd085..ae04fc19278f35ed3089eadd85c9b6b02d9b9162 100644
--- a/develop/doc_cn/design/refactor/parameter_server.html
+++ b/develop/doc_cn/design/refactor/parameter_server.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/refactorization.html b/develop/doc_cn/design/refactorization.html
index b88906f71d0e35698e5e7914acf669d4d4c5d2f0..01977e218196159c5262992cad43c72fb0df853b 100644
--- a/develop/doc_cn/design/refactorization.html
+++ b/develop/doc_cn/design/refactorization.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/releasing_process.html b/develop/doc_cn/design/releasing_process.html
index f5372a8427b41462a664cea24c9008a0e7e7942b..3b0777715c0a6f1ccd81a0e89e1e5dbd8466d1eb 100644
--- a/develop/doc_cn/design/releasing_process.html
+++ b/develop/doc_cn/design/releasing_process.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/scope.html b/develop/doc_cn/design/scope.html
index 9382c51f32d59f7896c64a431209333ff643cfaa..9ebf1d969c3fa4458835f6a4ae2205ed39358496 100644
--- a/develop/doc_cn/design/scope.html
+++ b/develop/doc_cn/design/scope.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/simple_op_design.html b/develop/doc_cn/design/simple_op_design.html
index e044de551d7f3f1f2e526d6c2aaac021f8621ae8..f264f10a7ca1a751dfb84ee7125f4e234c8c9cb3 100644
--- a/develop/doc_cn/design/simple_op_design.html
+++ b/develop/doc_cn/design/simple_op_design.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/design/var_desc.html b/develop/doc_cn/design/var_desc.html
index 10f0243127e56a3876fb1639a8a687e13c074c13..3a0cb2bf203ef843f2a8e4c124f10a5f71f98f58 100644
--- a/develop/doc_cn/design/var_desc.html
+++ b/develop/doc_cn/design/var_desc.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/faq/build_and_install/index_cn.html b/develop/doc_cn/faq/build_and_install/index_cn.html
new file mode 100644
index 0000000000000000000000000000000000000000..a1b50a1fde1f33d43075a68b6dd8e85326ad66d8
--- /dev/null
+++ b/develop/doc_cn/faq/build_and_install/index_cn.html
@@ -0,0 +1,369 @@
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>编译安装与单元测试 &mdash; PaddlePaddle  文档</title>
+  
+
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  
+  
+    <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  
+
+  
+  
+        <link rel="index" title="索引"
+              href="../../genindex.html"/>
+        <link rel="search" title="搜索" href="../../search.html"/>
+    <link rel="top" title="PaddlePaddle  文档" href="../../index.html"/>
+        <link rel="up" title="FAQ" href="../index_cn.html"/>
+        <link rel="next" title="模型配置" href="../model/index_cn.html"/>
+        <link rel="prev" title="FAQ" href="../index_cn.html"/> 
+
+  <link rel="stylesheet" href="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/css/perfect-scrollbar.min.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/override.css" type="text/css" />
+  <script>
+  var _hmt = _hmt || [];
+  (function() {
+    var hm = document.createElement("script");
+    hm.src = "//hm.baidu.com/hm.js?b9a314ab40d04d805655aab1deee08ba";
+    var s = document.getElementsByTagName("script")[0]; 
+    s.parentNode.insertBefore(hm, s);
+  })();
+  </script>
+
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+</head>
+
+<body class="wy-body-for-nav" role="document">
+
+  
+  <header class="site-header">
+    <div class="site-logo">
+      <a href="/"><img src="../../_static/images/PP_w.png"></a>
+    </div>
+    <div class="site-nav-links">
+      <div class="site-menu">
+        <a class="fork-on-github" href="https://github.com/PaddlePaddle/Paddle" target="_blank"><i class="fa fa-github"></i>Fork me on Github</a>
+        <div class="language-switcher dropdown">
+          <a type="button" data-toggle="dropdown">
+            <span>English</span>
+            <i class="fa fa-angle-up"></i>
+            <i class="fa fa-angle-down"></i>
+          </a>
+          <ul class="dropdown-menu">
+            <li><a href="/doc_cn">中文</a></li>
+            <li><a href="/doc">English</a></li>
+          </ul>
+        </div>
+        <ul class="site-page-links">
+          <li><a href="/">Home</a></li>
+        </ul>
+      </div>
+      <div class="doc-module">
+        
+        <ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">FAQ</a></li>
+</ul>
+
+        
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>        
+      </div>
+    </div>
+  </header>
+  
+  <div class="main-content-wrap">
+
+    
+    <nav class="doc-menu-vertical" role="navigation">
+        
+          
+          <ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../getstarted/build_and_install/index_cn.html">安装与编译</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/docker_install_cn.html">PaddlePaddle的Docker容器使用方式</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/cmake/build_from_source_cn.html">PaddlePaddle的编译选项</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../getstarted/concepts/use_concepts_cn.html">基本使用概念</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cmd_parameter/index_cn.html">设置命令行参数</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/use_case_cn.html">使用案例</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/arguments_cn.html">参数概述</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/detail_introduction_cn.html">细节描述</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cluster/cluster_train_cn.html">运行分布式训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_basis_cn.html">Kubernetes 简介</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_cn.html">Kubernetes单机训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_distributed_cn.html">Kubernetes分布式训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/build_cn.html">编译PaddlePaddle和运行单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/write_docs_cn.html">如何贡献/修改文档</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/contribute_to_paddle_cn.html">如何贡献代码</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/deep_model/rnn/index_cn.html">RNN相关模型</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/rnn_config_cn.html">RNN配置</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/recurrent_group_cn.html">Recurrent Group教程</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hierarchical_layer_cn.html">支持双层序列作为输入的Layer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hrnn_rnn_api_compare_cn.html">单双层RNN API对比介绍</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/optimization/gpu_profiling_cn.html">GPU性能分析与调优</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/model_configs.html">模型配置</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/activation.html">Activation</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/layer.html">Layers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/evaluators.html">Evaluators</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/optimizer.html">Optimizer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/pooling.html">Pooling</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/networks.html">Networks</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/attr.html">Parameter Attribute</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/data.html">数据访问</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
+</ul>
+</li>
+<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">FAQ</a><ul class="current">
+<li class="toctree-l2 current"><a class="current reference internal" href="#">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
+</ul>
+
+        
+    </nav>
+    
+    <section class="doc-content-wrap">
+
+      
+
+ 
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+  <ul class="wy-breadcrumbs">
+      
+        <li><a href="../index_cn.html">FAQ</a> > </li>
+      
+    <li>编译安装与单元测试</li>
+  </ul>
+</div>
+      
+      <div class="wy-nav-content" id="doc-content">
+        <div class="rst-content">
+          <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
+           <div itemprop="articleBody">
+            
+  <div class="section" id="id1">
+<h1><a class="toc-backref" href="#id2">编译安装与单元测试</a><a class="headerlink" href="#id1" title="永久链接至标题">¶</a></h1>
+<div class="contents topic" id="contents">
+<p class="topic-title first">Contents</p>
+<ul class="simple">
+<li><a class="reference internal" href="#id1" id="id2">编译安装与单元测试</a><ul>
+<li><a class="reference internal" href="#docker-gpu-cuda-driver-version-is-insufficient" id="id3">1. 运行Docker GPU镜像出现 &#8220;CUDA driver version is insufficient&#8221;</a></li>
+<li><a class="reference internal" href="#cmake-pythonlibspythoninterp" id="id4">2. CMake源码编译, 找到的PythonLibs和PythonInterp版本不一致</a></li>
+<li><a class="reference internal" href="#cmake-paddle0-0-0" id="id5">3. CMake源码编译，Paddle版本号为0.0.0</a></li>
+<li><a class="reference internal" href="#paddlepaddle-whl-is-not-a-supported-wheel-on-this-platform" id="id6">4. paddlepaddle*.whl is not a supported wheel on this platform.</a></li>
+<li><a class="reference internal" href="#import-paddle-v2-as-paddle-importerror-no-module-named-v2" id="id7">5. 编译安装后执行 import paddle.v2 as paddle 报ImportError: No module named v2</a></li>
+<li><a class="reference internal" href="#illegal-instruction" id="id8">6. 遇到“非法指令”或者是“illegal instruction”</a></li>
+<li><a class="reference internal" href="#python" id="id9">7.  python相关的单元测试都过不了</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<div class="section" id="docker-gpu-cuda-driver-version-is-insufficient">
+<h2><a class="toc-backref" href="#id3">1. 运行Docker GPU镜像出现 &#8220;CUDA driver version is insufficient&#8221;</a><a class="headerlink" href="#docker-gpu-cuda-driver-version-is-insufficient" title="永久链接至标题">¶</a></h2>
+<p>用户在使用PaddlePaddle GPU的Docker镜像的时候，常常出现 <cite>Cuda Error: CUDA driver version is insufficient for CUDA runtime version</cite>, 原因在于没有把机器上CUDA相关的驱动和库映射到容器内部。
+具体的解决方法是：</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">export</span> <span class="nv">CUDA_SO</span><span class="o">=</span><span class="s2">&quot;</span><span class="k">$(</span><span class="se">\l</span>s usr/lib64/libcuda* <span class="p">|</span> xargs -I<span class="o">{}</span> <span class="nb">echo</span> <span class="s1">&#39;-v {}:{}&#39;</span><span class="k">)</span><span class="s2"> </span><span class="k">$(</span><span class="se">\l</span>s /usr/lib64/libnvidia* <span class="p">|</span> xargs -I<span class="o">{}</span> <span class="nb">echo</span> <span class="s1">&#39;-v {}:{}&#39;</span><span class="k">)</span><span class="s2">&quot;</span>
+$ <span class="nb">export</span> <span class="nv">DEVICES</span><span class="o">=</span><span class="k">$(</span><span class="se">\l</span>s /dev/nvidia* <span class="p">|</span> xargs -I<span class="o">{}</span> <span class="nb">echo</span> <span class="s1">&#39;--device {}:{}&#39;</span><span class="k">)</span>
+$ docker run <span class="si">${</span><span class="nv">CUDA_SO</span><span class="si">}</span> <span class="si">${</span><span class="nv">DEVICES</span><span class="si">}</span> -it paddledev/paddlepaddle:latest-gpu
+</pre></div>
+</div>
+<p>更多关于Docker的安装与使用, 请参考 <a class="reference external" href="http://www.paddlepaddle.org/doc_cn/build_and_install/install/docker_install.html">PaddlePaddle Docker 文档</a> 。</p>
+</div>
+<div class="section" id="cmake-pythonlibspythoninterp">
+<h2><a class="toc-backref" href="#id4">2. CMake源码编译, 找到的PythonLibs和PythonInterp版本不一致</a><a class="headerlink" href="#cmake-pythonlibspythoninterp" title="永久链接至标题">¶</a></h2>
+<p>这是目前CMake寻找Python的逻辑存在缺陷，如果系统安装了多个Python版本，CMake找到的Python库和Python解释器版本可能有不一致现象，导致编译PaddlePaddle失败。正确的解决方法是，
+用户强制指定特定的Python版本，具体操作如下：</p>
+<blockquote>
+<div><div class="highlight-bash"><div class="highlight"><pre><span></span>cmake .. -DPYTHON_EXECUTABLE<span class="o">=</span>&lt;exc_path&gt; -DPYTHON_LIBRARY<span class="o">=</span>&lt;lib_path&gt;  -DPYTHON_INCLUDE_DIR<span class="o">=</span>&lt;inc_path&gt;
+</pre></div>
+</div>
+</div></blockquote>
+<p>用户需要指定本机上Python的路径：<code class="docutils literal"><span class="pre">&lt;exc_path&gt;</span></code>, <code class="docutils literal"><span class="pre">&lt;lib_path&gt;</span></code>, <code class="docutils literal"><span class="pre">&lt;inc_path&gt;</span></code></p>
+</div>
+<div class="section" id="cmake-paddle0-0-0">
+<h2><a class="toc-backref" href="#id5">3. CMake源码编译，Paddle版本号为0.0.0</a><a class="headerlink" href="#cmake-paddle0-0-0" title="永久链接至标题">¶</a></h2>
+<p>如果运行 <code class="code docutils literal"><span class="pre">paddle</span> <span class="pre">version</span></code>, 出现 <code class="code docutils literal"><span class="pre">PaddlePaddle</span> <span class="pre">0.0.0</span></code>；或者运行 <code class="code docutils literal"><span class="pre">cmake</span> <span class="pre">..</span></code>，出现</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>CMake Warning at cmake/version.cmake:20 <span class="o">(</span>message<span class="o">)</span>:
+  Cannot add paddle version from git tag
+</pre></div>
+</div>
+<p>那么用户需要拉取所有的远程分支到本机，命令为 <code class="code docutils literal"><span class="pre">git</span> <span class="pre">fetch</span> <span class="pre">upstream</span></code>，然后重新cmake即可。</p>
+</div>
+<div class="section" id="paddlepaddle-whl-is-not-a-supported-wheel-on-this-platform">
+<h2><a class="toc-backref" href="#id6">4. paddlepaddle*.whl is not a supported wheel on this platform.</a><a class="headerlink" href="#paddlepaddle-whl-is-not-a-supported-wheel-on-this-platform" title="永久链接至标题">¶</a></h2>
+<p>出现这个问题的主要原因是，没有找到和当前系统匹配的paddlepaddle安装包。最新的paddlepaddle python安装包支持Linux x86_64和MacOS 10.12操作系统，并安装了python 2.7和pip 9.0.1。</p>
+<p>更新 <code class="code docutils literal"><span class="pre">pip</span></code> 包的方法是:</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>pip install --upgrade pip
+</pre></div>
+</div>
+<p>如果还不行，可以执行 <code class="code docutils literal"><span class="pre">python</span> <span class="pre">-c</span> <span class="pre">&quot;import</span> <span class="pre">pip;</span> <span class="pre">print(pip.pep425tags.get_supported())&quot;</span></code> 获取当前系统支持的python包的后缀，
+并对比是否和正在安装的后缀一致。</p>
+<p>如果系统支持的是 <code class="code docutils literal"><span class="pre">linux_x86_64</span></code> 而安装包是 <code class="code docutils literal"><span class="pre">manylinux1_x86_64</span></code> ，需要升级pip版本到最新；
+如果系统支持 <code class="code docutils literal"><span class="pre">manylinux1_x86_64</span></code> 而安装包（本地）是 <code class="code docutils literal"><span class="pre">linux_x86_64</span></code> ，可以重命名这个whl包为 <code class="code docutils literal"><span class="pre">manylinux1_x86_64</span></code> 再安装。</p>
+</div>
+<div class="section" id="import-paddle-v2-as-paddle-importerror-no-module-named-v2">
+<h2><a class="toc-backref" href="#id7">5. 编译安装后执行 import paddle.v2 as paddle 报ImportError: No module named v2</a><a class="headerlink" href="#import-paddle-v2-as-paddle-importerror-no-module-named-v2" title="永久链接至标题">¶</a></h2>
+<p>先查看一下是否曾经安装过paddle v1版本，有的话需要先卸载：</p>
+<p>pip uninstall py_paddle paddle</p>
+<p>然后安装paddle的python环境, 在build目录下执行</p>
+<p>pip install python/dist/paddle*.whl &amp;&amp; pip install ../paddle/dist/py_paddle*.whl</p>
+</div>
+<div class="section" id="illegal-instruction">
+<h2><a class="toc-backref" href="#id8">6. 遇到“非法指令”或者是“illegal instruction”</a><a class="headerlink" href="#illegal-instruction" title="永久链接至标题">¶</a></h2>
+<p>PaddlePaddle使用avx SIMD指令提高cpu执行效率，因此错误的使用二进制发行版可能会导致这种错误，请选择正确的版本。</p>
+</div>
+<div class="section" id="python">
+<h2><a class="toc-backref" href="#id9">7.  python相关的单元测试都过不了</a><a class="headerlink" href="#python" title="永久链接至标题">¶</a></h2>
+<p>如果出现以下python相关的单元测试都过不了的情况：</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="m">24</span> - test_PyDataProvider <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">26</span> - test_RecurrentGradientMachine <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">27</span> - test_NetworkCompare <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">28</span> - test_PyDataProvider2 <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">32</span> - test_Prediction <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">33</span> - test_Compare <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">34</span> - test_Trainer <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">35</span> - test_TrainerOnePass <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">36</span> - test_CompareTwoNets <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">37</span> - test_CompareTwoOpts <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">38</span> - test_CompareSparse <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">39</span> - test_recurrent_machine_generation <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">40</span> - test_PyDataProviderWrapper <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">41</span> - test_config_parser <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">42</span> - test_swig_api <span class="o">(</span>Failed<span class="o">)</span>
+<span class="m">43</span> - layers_test <span class="o">(</span>Failed<span class="o">)</span>
+</pre></div>
+</div>
+<p>并且查询PaddlePaddle单元测试的日志，提示：</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>paddle package is already in your PYTHONPATH. But unittest need a clean environment.
+Please uninstall paddle package before start unittest. Try to <span class="s1">&#39;pip uninstall paddle&#39;</span>.
+</pre></div>
+</div>
+<p>解决办法是：</p>
+<ul class="simple">
+<li>卸载PaddlePaddle包 <code class="code docutils literal"><span class="pre">pip</span> <span class="pre">uninstall</span> <span class="pre">paddle</span></code>, 清理掉老旧的PaddlePaddle安装包，使得单元测试有一个干净的环境。如果PaddlePaddle包已经在python的site-packages里面，单元测试会引用site-packages里面的python包，而不是源码目录里 <code class="code docutils literal"><span class="pre">/python</span></code> 目录下的python包。同时，即便设置 <code class="code docutils literal"><span class="pre">PYTHONPATH</span></code> 到 <code class="code docutils literal"><span class="pre">/python</span></code> 也没用，因为python的搜索路径是优先已经安装的python包。</li>
+</ul>
+</div>
+</div>
+
+
+           </div>
+          </div>
+          <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../model/index_cn.html" class="btn btn-neutral float-right" title="模型配置" accesskey="n">Next <span class="fa fa-arrow-circle-right"></span></a>
+      
+      
+        <a href="../index_cn.html" class="btn btn-neutral" title="FAQ" accesskey="p"><span class="fa fa-arrow-circle-left"></span> Previous</a>
+      
+    </div>
+  
+
+  <hr/>
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2016, PaddlePaddle developers.
+
+    </p>
+  </div>
+  Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>. 
+
+</footer>
+
+        </div>
+      </div>
+
+    </section>
+
+  </div>
+  
+
+
+  
+
+    <script type="text/javascript">
+        var DOCUMENTATION_OPTIONS = {
+            URL_ROOT:'../../',
+            VERSION:'',
+            COLLAPSE_INDEX:false,
+            FILE_SUFFIX:'.html',
+            HAS_SOURCE:  true,
+            SOURCELINK_SUFFIX: ".txt",
+        };
+    </script>
+      <script type="text/javascript" src="../../_static/jquery.js"></script>
+      <script type="text/javascript" src="../../_static/underscore.js"></script>
+      <script type="text/javascript" src="../../_static/doctools.js"></script>
+      <script type="text/javascript" src="../../_static/translations.js"></script>
+      <script type="text/javascript" src="https://cdn.bootcss.com/mathjax/2.7.0/MathJax.js"></script>
+       
+  
+
+  
+  
+    <script type="text/javascript" src="../../_static/js/theme.js"></script>
+  
+  
+  <script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/js/bootstrap.min.js" integrity="sha384-Tc5IQib027qvyjSMfHjOMaLkfuWVxZxUPnCJA7l2mCWNIpG9mGCD8wGNIcPD7Txa" crossorigin="anonymous"></script>
+  <script src="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/js/perfect-scrollbar.jquery.min.js"></script>
+  <script src="../../_static/js/paddle_doc_init.js"></script> 
+
+</body>
+</html>
\ No newline at end of file
diff --git a/develop/doc_cn/faq/cluster/index_cn.html b/develop/doc_cn/faq/cluster/index_cn.html
new file mode 100644
index 0000000000000000000000000000000000000000..86b97479846e2692e8bf238d1d6d9c2dacf9f425
--- /dev/null
+++ b/develop/doc_cn/faq/cluster/index_cn.html
@@ -0,0 +1,285 @@
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>集群训练与预测 &mdash; PaddlePaddle  文档</title>
+  
+
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  
+  
+    <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  
+
+  
+  
+        <link rel="index" title="索引"
+              href="../../genindex.html"/>
+        <link rel="search" title="搜索" href="../../search.html"/>
+    <link rel="top" title="PaddlePaddle  文档" href="../../index.html"/>
+        <link rel="up" title="FAQ" href="../index_cn.html"/>
+        <link rel="prev" title="本地训练与预测" href="../local/index_cn.html"/> 
+
+  <link rel="stylesheet" href="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/css/perfect-scrollbar.min.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/override.css" type="text/css" />
+  <script>
+  var _hmt = _hmt || [];
+  (function() {
+    var hm = document.createElement("script");
+    hm.src = "//hm.baidu.com/hm.js?b9a314ab40d04d805655aab1deee08ba";
+    var s = document.getElementsByTagName("script")[0]; 
+    s.parentNode.insertBefore(hm, s);
+  })();
+  </script>
+
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+</head>
+
+<body class="wy-body-for-nav" role="document">
+
+  
+  <header class="site-header">
+    <div class="site-logo">
+      <a href="/"><img src="../../_static/images/PP_w.png"></a>
+    </div>
+    <div class="site-nav-links">
+      <div class="site-menu">
+        <a class="fork-on-github" href="https://github.com/PaddlePaddle/Paddle" target="_blank"><i class="fa fa-github"></i>Fork me on Github</a>
+        <div class="language-switcher dropdown">
+          <a type="button" data-toggle="dropdown">
+            <span>English</span>
+            <i class="fa fa-angle-up"></i>
+            <i class="fa fa-angle-down"></i>
+          </a>
+          <ul class="dropdown-menu">
+            <li><a href="/doc_cn">中文</a></li>
+            <li><a href="/doc">English</a></li>
+          </ul>
+        </div>
+        <ul class="site-page-links">
+          <li><a href="/">Home</a></li>
+        </ul>
+      </div>
+      <div class="doc-module">
+        
+        <ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">FAQ</a></li>
+</ul>
+
+        
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>        
+      </div>
+    </div>
+  </header>
+  
+  <div class="main-content-wrap">
+
+    
+    <nav class="doc-menu-vertical" role="navigation">
+        
+          
+          <ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../getstarted/build_and_install/index_cn.html">安装与编译</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/docker_install_cn.html">PaddlePaddle的Docker容器使用方式</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/cmake/build_from_source_cn.html">PaddlePaddle的编译选项</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../getstarted/concepts/use_concepts_cn.html">基本使用概念</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cmd_parameter/index_cn.html">设置命令行参数</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/use_case_cn.html">使用案例</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/arguments_cn.html">参数概述</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/detail_introduction_cn.html">细节描述</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cluster/cluster_train_cn.html">运行分布式训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_basis_cn.html">Kubernetes 简介</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_cn.html">Kubernetes单机训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_distributed_cn.html">Kubernetes分布式训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/build_cn.html">编译PaddlePaddle和运行单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/write_docs_cn.html">如何贡献/修改文档</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/contribute_to_paddle_cn.html">如何贡献代码</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/deep_model/rnn/index_cn.html">RNN相关模型</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/rnn_config_cn.html">RNN配置</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/recurrent_group_cn.html">Recurrent Group教程</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hierarchical_layer_cn.html">支持双层序列作为输入的Layer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hrnn_rnn_api_compare_cn.html">单双层RNN API对比介绍</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/optimization/gpu_profiling_cn.html">GPU性能分析与调优</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/model_configs.html">模型配置</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/activation.html">Activation</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/layer.html">Layers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/evaluators.html">Evaluators</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/optimizer.html">Optimizer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/pooling.html">Pooling</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/networks.html">Networks</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/attr.html">Parameter Attribute</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/data.html">数据访问</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
+</ul>
+</li>
+<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">FAQ</a><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2 current"><a class="current reference internal" href="#">集群训练与预测</a></li>
+</ul>
+</li>
+</ul>
+
+        
+    </nav>
+    
+    <section class="doc-content-wrap">
+
+      
+
+ 
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+  <ul class="wy-breadcrumbs">
+      
+        <li><a href="../index_cn.html">FAQ</a> > </li>
+      
+    <li>集群训练与预测</li>
+  </ul>
+</div>
+      
+      <div class="wy-nav-content" id="doc-content">
+        <div class="rst-content">
+          <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
+           <div itemprop="articleBody">
+            
+  <div class="section" id="id1">
+<h1><a class="toc-backref" href="#id3">集群训练与预测</a><a class="headerlink" href="#id1" title="永久链接至标题">¶</a></h1>
+<div class="contents topic" id="contents">
+<p class="topic-title first">Contents</p>
+<ul class="simple">
+<li><a class="reference internal" href="#id1" id="id3">集群训练与预测</a><ul>
+<li><a class="reference internal" href="#id2" id="id4">1. 集群多节点训练，日志中保存均为网络通信类错误</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<div class="section" id="id2">
+<h2><a class="toc-backref" href="#id4">1. 集群多节点训练，日志中保存均为网络通信类错误</a><a class="headerlink" href="#id2" title="永久链接至标题">¶</a></h2>
+<p>集群多节点训练，日志报错为网络通信类错误，比如 <code class="code docutils literal"><span class="pre">Connection</span> <span class="pre">reset</span> <span class="pre">by</span> <span class="pre">peer</span></code> 等。
+此类报错通常是由于某一个节点的错误导致这个节点的训练进程退出，从而引发其他节点无法连接导致，可以参考下面的步骤排查：</p>
+<ul class="simple">
+<li>从 <code class="code docutils literal"><span class="pre">train.log</span></code> ， <code class="code docutils literal"><span class="pre">server.log</span></code> 找到最早报错的地方，查看是否是其他错误引发的报错（比如FPE，内存不足，磁盘空间不足等）。</li>
+<li>如果发现最早的报错就是网络通信的问题，很有可能是非独占方式执行导致的端口冲突，可以联系OP，看当前MPI集群是否支持resource=full参数提交，如果支持增加此参数提交，并更换job 端口。</li>
+<li>如果当前MPI集群并不支持任务独占模式，可以联系OP是否可以更换集群或升级当前集群。</li>
+</ul>
+</div>
+</div>
+
+
+           </div>
+          </div>
+          <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+      
+        <a href="../local/index_cn.html" class="btn btn-neutral" title="本地训练与预测" accesskey="p"><span class="fa fa-arrow-circle-left"></span> Previous</a>
+      
+    </div>
+  
+
+  <hr/>
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2016, PaddlePaddle developers.
+
+    </p>
+  </div>
+  Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>. 
+
+</footer>
+
+        </div>
+      </div>
+
+    </section>
+
+  </div>
+  
+
+
+  
+
+    <script type="text/javascript">
+        var DOCUMENTATION_OPTIONS = {
+            URL_ROOT:'../../',
+            VERSION:'',
+            COLLAPSE_INDEX:false,
+            FILE_SUFFIX:'.html',
+            HAS_SOURCE:  true,
+            SOURCELINK_SUFFIX: ".txt",
+        };
+    </script>
+      <script type="text/javascript" src="../../_static/jquery.js"></script>
+      <script type="text/javascript" src="../../_static/underscore.js"></script>
+      <script type="text/javascript" src="../../_static/doctools.js"></script>
+      <script type="text/javascript" src="../../_static/translations.js"></script>
+      <script type="text/javascript" src="https://cdn.bootcss.com/mathjax/2.7.0/MathJax.js"></script>
+       
+  
+
+  
+  
+    <script type="text/javascript" src="../../_static/js/theme.js"></script>
+  
+  
+  <script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/js/bootstrap.min.js" integrity="sha384-Tc5IQib027qvyjSMfHjOMaLkfuWVxZxUPnCJA7l2mCWNIpG9mGCD8wGNIcPD7Txa" crossorigin="anonymous"></script>
+  <script src="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/js/perfect-scrollbar.jquery.min.js"></script>
+  <script src="../../_static/js/paddle_doc_init.js"></script> 
+
+</body>
+</html>
\ No newline at end of file
diff --git a/develop/doc_cn/faq/index_cn.html b/develop/doc_cn/faq/index_cn.html
index db008d610dcdebddf622734bf3b0b4621a3276f6..e319d6c1c7f150a4647b99743e617a215cd98099 100644
--- a/develop/doc_cn/faq/index_cn.html
+++ b/develop/doc_cn/faq/index_cn.html
@@ -33,6 +33,7 @@
               href="../genindex.html"/>
         <link rel="search" title="搜索" href="../search.html"/>
     <link rel="top" title="PaddlePaddle  文档" href="../index.html"/>
+        <link rel="next" title="编译安装与单元测试" href="build_and_install/index_cn.html"/>
         <link rel="prev" title="Training and Inference" href="../api/v2/run_logic.html"/> 
 
   <link rel="stylesheet" href="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/css/perfect-scrollbar.min.css" type="text/css" />
@@ -155,7 +156,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1 current"><a class="current reference internal" href="#">FAQ</a></li>
+<li class="toctree-l1 current"><a class="current reference internal" href="#">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
@@ -186,629 +194,15 @@
            <div itemprop="articleBody">
             
   <div class="section" id="faq">
-<h1><a class="toc-backref" href="#id12">FAQ</a><a class="headerlink" href="#faq" title="永久链接至标题">¶</a></h1>
-<div class="contents topic" id="contents">
-<p class="topic-title first">Contents</p>
-<ul class="simple">
-<li><a class="reference internal" href="#faq" id="id12">FAQ</a><ul>
-<li><a class="reference internal" href="#id1" id="id13">1. 如何减少内存占用</a><ul>
-<li><a class="reference internal" href="#dataprovider" id="id14">减少DataProvider缓冲池内存</a></li>
-<li><a class="reference internal" href="#id2" id="id15">神经元激活内存</a></li>
-<li><a class="reference internal" href="#id3" id="id16">参数内存</a></li>
-</ul>
-</li>
-<li><a class="reference internal" href="#paddlepaddle" id="id17">2. 如何加速PaddlePaddle的训练速度</a><ul>
-<li><a class="reference internal" href="#id4" id="id18">减少数据载入的耗时</a></li>
-<li><a class="reference internal" href="#id5" id="id19">加速训练速度</a></li>
-<li><a class="reference internal" href="#id6" id="id20">利用更多的计算资源</a></li>
-</ul>
-</li>
-<li><a class="reference internal" href="#illegal-instruction" id="id21">3. 遇到“非法指令”或者是“illegal instruction”</a></li>
-<li><a class="reference internal" href="#sgd" id="id22">4. 如何选择SGD算法的学习率</a></li>
-<li><a class="reference internal" href="#id7" id="id23">5. 如何初始化参数</a></li>
-<li><a class="reference internal" href="#id8" id="id24">6. 如何共享参数</a></li>
-<li><a class="reference internal" href="#paddlepaddle-whl-is-not-a-supported-wheel-on-this-platform" id="id25">7. paddlepaddle*.whl is not a supported wheel on this platform.</a></li>
-<li><a class="reference internal" href="#python" id="id26">8.  python相关的单元测试都过不了</a></li>
-<li><a class="reference internal" href="#docker-gpu-cuda-driver-version-is-insufficient" id="id27">9. 运行Docker GPU镜像出现 &#8220;CUDA driver version is insufficient&#8221;</a></li>
-<li><a class="reference internal" href="#cmake-pythonlibspythoninterp" id="id28">10. CMake源码编译, 找到的PythonLibs和PythonInterp版本不一致</a></li>
-<li><a class="reference internal" href="#cmake-paddle0-0-0" id="id29">11. CMake源码编译，Paddle版本号为0.0.0</a></li>
-<li><a class="reference internal" href="#a-protocol-message-was-rejected-because-it-was-too-big" id="id30">12. A protocol message was rejected because it was too big</a></li>
-<li><a class="reference internal" href="#gpu" id="id31">13. 如何指定GPU设备</a></li>
-<li><a class="reference internal" href="#floating-point-exception" id="id32">14. 训练过程中出现 <code class="code docutils literal"><span class="pre">Floating</span> <span class="pre">point</span> <span class="pre">exception</span></code>, 训练因此退出怎么办?</a></li>
-<li><a class="reference internal" href="#import-paddle-v2-as-paddle-importerror-no-module-named-v2" id="id33">15. 编译安装后执行 import paddle.v2 as paddle 报ImportError: No module named v2</a></li>
-<li><a class="reference internal" href="#id9" id="id34">16. PaddlePaddle存储的参数格式是什么，如何和明文进行相互转化</a></li>
-<li><a class="reference internal" href="#id10" id="id35">17. 如何加载预训练参数</a></li>
-<li><a class="reference internal" href="#id11" id="id36">18. 集群多节点训练，日志中保存均为网络通信类错误</a></li>
-<li><a class="reference internal" href="#infer-layer" id="id37">19.  如何调用 infer 接口输出多个layer的预测结果</a></li>
-<li><a class="reference internal" href="#paddle-layer-memory-name" id="id38">20. <code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 的参数 <code class="code docutils literal"><span class="pre">name</span></code> 如何使用</a></li>
-<li><a class="reference internal" href="#drop-out" id="id39">21. 两种使用 drop_out 的方法有何区别？</a></li>
-<li><a class="reference internal" href="#learning-rate-annealing" id="id40">22. 如何设置学习率退火（learning rate annealing）</a></li>
-<li><a class="reference internal" href="#duplicated-layer-name" id="id41">23. 出现 <code class="code docutils literal"><span class="pre">Duplicated</span> <span class="pre">layer</span> <span class="pre">name</span></code> 错误怎么办</a></li>
-<li><a class="reference internal" href="#paddlepaddle-recurrent-layer" id="id42">24. PaddlePaddle 中不同的 recurrent layer 的区别</a></li>
-</ul>
-</li>
-</ul>
-</div>
-<div class="section" id="id1">
-<h2><a class="toc-backref" href="#id13">1. 如何减少内存占用</a><a class="headerlink" href="#id1" title="永久链接至标题">¶</a></h2>
-<p>神经网络的训练本身是一个非常消耗内存和显存的工作，经常会消耗数10GB的内存和数GB的显存。
-PaddlePaddle的内存占用主要分为如下几个方面:</p>
-<ul class="simple">
-<li>DataProvider缓冲池内存（只针对内存）</li>
-<li>神经元激活内存（针对内存和显存）</li>
-<li>参数内存 （针对内存和显存）</li>
-<li>其他内存杂项</li>
-</ul>
-<p>其中，其他内存杂项是指PaddlePaddle本身所用的一些内存，包括字符串分配，临时变量等等，暂不考虑在内。</p>
-<div class="section" id="dataprovider">
-<h3><a class="toc-backref" href="#id14">减少DataProvider缓冲池内存</a><a class="headerlink" href="#dataprovider" title="永久链接至标题">¶</a></h3>
-<p>PyDataProvider使用的是异步加载，同时在内存里直接随即选取数据来做Shuffle。即</p>
-<img src="../_images/graphviz-9be6aad37f57c60f4b971dde0ef44ce27179cf9a.png" alt="digraph {
-    rankdir=LR;
-    数据文件 -&gt; 内存池 -&gt; PaddlePaddle训练
-}" />
-<p>所以，减小这个内存池即可减小内存占用，同时也可以加速开始训练前数据载入的过程。但是，这
-个内存池实际上决定了shuffle的粒度。所以，如果将这个内存池减小，又要保证数据是随机的，
-那么最好将数据文件在每次读取之前做一次shuffle。可能的代码为</p>
-<div class="highlight-default"><div class="highlight"><pre><span></span><span class="nd">@provider</span><span class="p">(</span><span class="n">min_pool_size</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="o">...</span><span class="p">)</span>
-<span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">settings</span><span class="p">,</span> <span class="n">filename</span><span class="p">):</span>
-    <span class="n">os</span><span class="o">.</span><span class="n">system</span><span class="p">(</span><span class="s1">&#39;shuf </span><span class="si">%s</span><span class="s1"> &gt; </span><span class="si">%s</span><span class="s1">.shuf&#39;</span> <span class="o">%</span> <span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">filename</span><span class="p">))</span>  <span class="c1"># shuffle before.</span>
-    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s1">&#39;</span><span class="si">%s</span><span class="s1">.shuf&#39;</span> <span class="o">%</span> <span class="n">filename</span><span class="p">,</span> <span class="s1">&#39;r&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
-        <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">f</span><span class="p">:</span>
-            <span class="k">yield</span> <span class="n">get_sample_from_line</span><span class="p">(</span><span class="n">line</span><span class="p">)</span>
-</pre></div>
-</div>
-<p>这样做可以极大的减少内存占用，并且可能会加速训练过程，详细文档参考 <a class="reference internal" href="../api/v1/data_provider/pydataprovider2_cn.html#api-pydataprovider2"><span class="std std-ref">PyDataProvider2的使用</span></a> 。</p>
-</div>
-<div class="section" id="id2">
-<h3><a class="toc-backref" href="#id15">神经元激活内存</a><a class="headerlink" href="#id2" title="永久链接至标题">¶</a></h3>
-<p>神经网络在训练的时候，会对每一个激活暂存一些数据，如神经元激活值等。
-在反向传递的时候，这些数据会被用来更新参数。这些数据使用的内存主要和两个参数有关系，
-一是batch size，另一个是每条序列(Sequence)长度。所以，其实也是和每个mini-batch中包含
-的时间步信息成正比。</p>
-<p>所以做法可以有两种：</p>
-<ul class="simple">
-<li>减小batch size。 即在网络配置中 <code class="code docutils literal"><span class="pre">settings(batch_size=1000)</span></code> 设置成一个小一些的值。但是batch size本身是神经网络的超参数，减小batch size可能会对训练结果产生影响。</li>
-<li>减小序列的长度，或者直接扔掉非常长的序列。比如，一个数据集大部分序列长度是100-200,
-但是突然有一个10000长的序列，就很容易导致内存超限，特别是在LSTM等RNN中。</li>
-</ul>
-</div>
-<div class="section" id="id3">
-<h3><a class="toc-backref" href="#id16">参数内存</a><a class="headerlink" href="#id3" title="永久链接至标题">¶</a></h3>
-<p>PaddlePaddle支持非常多的优化算法(Optimizer)，不同的优化算法需要使用不同大小的内存。
-例如使用 <code class="code docutils literal"><span class="pre">adadelta</span></code> 算法，则需要使用等于权重参数规模大约5倍的内存。举例，如果参数保存下来的模型目录
-文件为 <code class="code docutils literal"><span class="pre">100M</span></code>， 那么该优化算法至少需要 <code class="code docutils literal"><span class="pre">500M</span></code> 的内存。</p>
-<p>可以考虑使用一些优化算法，例如 <code class="code docutils literal"><span class="pre">momentum</span></code>。</p>
-</div>
-</div>
-<div class="section" id="paddlepaddle">
-<h2><a class="toc-backref" href="#id17">2. 如何加速PaddlePaddle的训练速度</a><a class="headerlink" href="#paddlepaddle" title="永久链接至标题">¶</a></h2>
-<p>加速PaddlePaddle训练可以考虑从以下几个方面：</p>
-<ul class="simple">
-<li>减少数据载入的耗时</li>
-<li>加速训练速度</li>
-<li>利用分布式训练驾驭更多的计算资源</li>
-</ul>
-<div class="section" id="id4">
-<h3><a class="toc-backref" href="#id18">减少数据载入的耗时</a><a class="headerlink" href="#id4" title="永久链接至标题">¶</a></h3>
-<p>使用<code class="code docutils literal"><span class="pre">pydataprovider</span></code>时，可以减少缓存池的大小，同时设置内存缓存功能，即可以极大的加速数据载入流程。
-<code class="code docutils literal"><span class="pre">DataProvider</span></code> 缓存池的减小，和之前减小通过减小缓存池来减小内存占用的原理一致。</p>
-<div class="highlight-default"><div class="highlight"><pre><span></span><span class="nd">@provider</span><span class="p">(</span><span class="n">min_pool_size</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="o">...</span><span class="p">)</span>
-<span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">settings</span><span class="p">,</span> <span class="n">filename</span><span class="p">):</span>
-    <span class="n">os</span><span class="o">.</span><span class="n">system</span><span class="p">(</span><span class="s1">&#39;shuf </span><span class="si">%s</span><span class="s1"> &gt; </span><span class="si">%s</span><span class="s1">.shuf&#39;</span> <span class="o">%</span> <span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">filename</span><span class="p">))</span>  <span class="c1"># shuffle before.</span>
-    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s1">&#39;</span><span class="si">%s</span><span class="s1">.shuf&#39;</span> <span class="o">%</span> <span class="n">filename</span><span class="p">,</span> <span class="s1">&#39;r&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
-        <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">f</span><span class="p">:</span>
-            <span class="k">yield</span> <span class="n">get_sample_from_line</span><span class="p">(</span><span class="n">line</span><span class="p">)</span>
-</pre></div>
-</div>
-<p>同时 <code class="code docutils literal"><span class="pre">&#64;provider</span></code> 接口有一个 <code class="code docutils literal"><span class="pre">cache</span></code> 参数来控制缓存方法，将其设置成 <code class="code docutils literal"><span class="pre">CacheType.CACHE_PASS_IN_MEM</span></code> 的话，会将第一个 <code class="code docutils literal"><span class="pre">pass</span></code> (过完所有训练数据即为一个pass)生成的数据缓存在内存里，在之后的 <code class="code docutils literal"><span class="pre">pass</span></code> 中，不会再从 <code class="code docutils literal"><span class="pre">python</span></code> 端读取数据，而是直接从内存的缓存里读取数据。这也会极大减少数据读入的耗时。</p>
-</div>
-<div class="section" id="id5">
-<h3><a class="toc-backref" href="#id19">加速训练速度</a><a class="headerlink" href="#id5" title="永久链接至标题">¶</a></h3>
-<p>PaddlePaddle支持Sparse的训练，sparse训练需要训练特征是 <code class="code docutils literal"><span class="pre">sparse_binary_vector</span></code> 、 <code class="code docutils literal"><span class="pre">sparse_vector</span></code> 、或者 <code class="code docutils literal"><span class="pre">integer_value</span></code> 的任一一种。同时，与这个训练数据交互的Layer，需要将其Parameter设置成 sparse 更新模式，即设置 <code class="code docutils literal"><span class="pre">sparse_update=True</span></code></p>
-<p>这里使用简单的 <code class="code docutils literal"><span class="pre">word2vec</span></code> 训练语言模型距离，具体使用方法为:</p>
-<p>使用一个词前两个词和后两个词，来预测这个中间的词。这个任务的DataProvider为:</p>
-<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">DICT_DIM</span> <span class="o">=</span> <span class="mi">3000</span>
-
-
-<span class="nd">@provider</span><span class="p">(</span><span class="n">input_types</span><span class="o">=</span><span class="p">[</span><span class="n">integer_sequence</span><span class="p">(</span><span class="n">DICT_DIM</span><span class="p">),</span> <span class="n">integer_value</span><span class="p">(</span><span class="n">DICT_DIM</span><span class="p">)])</span>
-<span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">settings</span><span class="p">,</span> <span class="n">filename</span><span class="p">):</span>
-    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">filename</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
-        <span class="c1"># yield word ids to predict inner word id</span>
-        <span class="c1"># such as [28, 29, 10, 4], 4</span>
-        <span class="c1"># It means the sentance is  28, 29, 4, 10, 4.</span>
-        <span class="k">yield</span> <span class="n">read_next_from_file</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>
-</pre></div>
-</div>
-<p>这个任务的配置为:</p>
-<div class="highlight-default"><div class="highlight"><pre><span></span><span class="o">...</span>  <span class="c1"># the settings and define data provider is omitted.</span>
-<span class="n">DICT_DIM</span> <span class="o">=</span> <span class="mi">3000</span>  <span class="c1"># dictionary dimension.</span>
-<span class="n">word_ids</span> <span class="o">=</span> <span class="n">data_layer</span><span class="p">(</span><span class="s1">&#39;word_ids&#39;</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">DICT_DIM</span><span class="p">)</span>
-
-<span class="n">emb</span> <span class="o">=</span> <span class="n">embedding_layer</span><span class="p">(</span>
-    <span class="nb">input</span><span class="o">=</span><span class="n">word_ids</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span> <span class="n">param_attr</span><span class="o">=</span><span class="n">ParamAttr</span><span class="p">(</span><span class="n">sparse_update</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
-<span class="n">emb_sum</span> <span class="o">=</span> <span class="n">pooling_layer</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">emb</span><span class="p">,</span> <span class="n">pooling_type</span><span class="o">=</span><span class="n">SumPooling</span><span class="p">())</span>
-<span class="n">predict</span> <span class="o">=</span> <span class="n">fc_layer</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">emb_sum</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">DICT_DIM</span><span class="p">,</span> <span class="n">act</span><span class="o">=</span><span class="n">Softmax</span><span class="p">())</span>
-<span class="n">outputs</span><span class="p">(</span>
-    <span class="n">classification_cost</span><span class="p">(</span>
-        <span class="nb">input</span><span class="o">=</span><span class="n">predict</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="n">data_layer</span><span class="p">(</span>
-            <span class="s1">&#39;label&#39;</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">DICT_DIM</span><span class="p">)))</span>
-</pre></div>
-</div>
-</div>
-<div class="section" id="id6">
-<h3><a class="toc-backref" href="#id20">利用更多的计算资源</a><a class="headerlink" href="#id6" title="永久链接至标题">¶</a></h3>
-<p>利用更多的计算资源可以分为一下几个方式来进行:</p>
-<ul class="simple">
-<li>单机CPU训练<ul>
-<li>使用多线程训练。设置命令行参数 <code class="code docutils literal"><span class="pre">trainer_count</span></code>。</li>
-</ul>
-</li>
-<li>单机GPU训练<ul>
-<li>使用显卡训练。设置命令行参数 <code class="code docutils literal"><span class="pre">use_gpu</span></code>。</li>
-<li>使用多块显卡训练。设置命令行参数 <code class="code docutils literal"><span class="pre">use_gpu</span></code> 和 <code class="code docutils literal"><span class="pre">trainer_count</span></code> 。</li>
-</ul>
-</li>
-<li>多机训练<ul>
-<li>请参考 <a class="reference internal" href="../howto/usage/cluster/cluster_train_cn.html#cluster-train"><span class="std std-ref">运行分布式训练</span></a> 。</li>
-</ul>
-</li>
-</ul>
-</div>
-</div>
-<div class="section" id="illegal-instruction">
-<h2><a class="toc-backref" href="#id21">3. 遇到“非法指令”或者是“illegal instruction”</a><a class="headerlink" href="#illegal-instruction" title="永久链接至标题">¶</a></h2>
-<p>PaddlePaddle使用avx SIMD指令提高cpu执行效率，因此错误的使用二进制发行版可能会导致这种错误，请选择正确的版本。</p>
-</div>
-<div class="section" id="sgd">
-<h2><a class="toc-backref" href="#id22">4. 如何选择SGD算法的学习率</a><a class="headerlink" href="#sgd" title="永久链接至标题">¶</a></h2>
-<p>在采用sgd/async_sgd进行训练时，一个重要的问题是选择正确的learning_rate。如果learning_rate太大，那么训练有可能不收敛，如果learning_rate太小，那么收敛可能很慢，导致训练时间过长。</p>
-<p>通常做法是从一个比较大的learning_rate开始试，如果不收敛，那减少学习率10倍继续试验，直到训练收敛为止。那么如何判断训练不收敛呢？可以估计出如果模型采用不变的输出最小的cost0是多少。</p>
-<p>如果训练过程的的cost明显高于这个常数输出的cost，那么我们可以判断为训练不收敛。举一个例子，假如我们是三分类问题，采用multi-class-cross-entropy作为cost，数据中0,1,2三类的比例为 <code class="code docutils literal"><span class="pre">0.2,</span> <span class="pre">0.5,</span> <span class="pre">0.3</span></code> , 那么常数输出所能达到的最小cost是 <code class="code docutils literal"><span class="pre">-(0.2*log(0.2)+0.5*log(0.5)+0.3*log(0.3))=1.03</span></code> 。如果训练一个pass（或者更早）后，cost还大于这个数，那么可以认为训练不收敛，应该降低学习率。</p>
-</div>
-<div class="section" id="id7">
-<h2><a class="toc-backref" href="#id23">5. 如何初始化参数</a><a class="headerlink" href="#id7" title="永久链接至标题">¶</a></h2>
-<p>默认情况下，PaddlePaddle使用均值0，标准差为 <span class="math">\(\frac{1}{\sqrt{d}}\)</span> 来初始化参数。其中 <span class="math">\(d\)</span> 为参数矩阵的宽度。这种初始化方式在一般情况下不会产生很差的结果。如果用户想要自定义初始化方式，PaddlePaddle目前提供两种参数初始化的方式:</p>
-<ul class="simple">
-<li>高斯分布。将 <code class="code docutils literal"><span class="pre">param_attr</span></code> 设置成 <code class="code docutils literal"><span class="pre">param_attr=ParamAttr(initial_mean=0.0,</span> <span class="pre">initial_std=1.0)</span></code></li>
-<li>均匀分布。将 <code class="code docutils literal"><span class="pre">param_attr</span></code> 设置成 <code class="code docutils literal"><span class="pre">param_attr=ParamAttr(initial_max=1.0,</span> <span class="pre">initial_min=-1.0)</span></code></li>
-</ul>
-<p>比如设置一个全连接层的参数初始化方式和bias初始化方式，可以使用如下代码。</p>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">hidden</span> <span class="o">=</span> <span class="n">fc_layer</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">ipt</span><span class="p">,</span> <span class="n">param_attr</span><span class="o">=</span><span class="n">ParamAttr</span><span class="p">(</span><span class="n">initial_max</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">initial_min</span><span class="o">=-</span><span class="mf">1.0</span><span class="p">),</span>
-                  <span class="n">bias_attr</span><span class="o">=</span><span class="n">ParamAttr</span><span class="p">(</span><span class="n">initial_mean</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">initial_std</span><span class="o">=</span><span class="mf">0.0</span><span class="p">))</span>
-</pre></div>
-</div>
-<p>上述代码将bias全部初始化为1.0, 同时将参数初始化为 <code class="code docutils literal"><span class="pre">[1.0,</span> <span class="pre">-1.0]</span></code> 的均匀分布。</p>
-</div>
-<div class="section" id="id8">
-<h2><a class="toc-backref" href="#id24">6. 如何共享参数</a><a class="headerlink" href="#id8" title="永久链接至标题">¶</a></h2>
-<p>PaddlePaddle的参数使用名字 <code class="code docutils literal"><span class="pre">name</span></code> 作为参数的ID，相同名字的参数，会共享参数。设置参数的名字，可以使用 <code class="code docutils literal"><span class="pre">ParamAttr(name=&quot;YOUR_PARAM_NAME&quot;)</span></code> 来设置。更方便的设置方式，是使得要共享的参数使用同样的 <code class="code docutils literal"><span class="pre">ParamAttr</span></code> 对象。</p>
-<p>简单的全连接网络，参数共享的配置示例为:</p>
-<div class="highlight-default"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">paddle.trainer_config_helpers</span> <span class="k">import</span> <span class="o">*</span>
-
-<span class="n">settings</span><span class="p">(</span><span class="n">learning_rate</span><span class="o">=</span><span class="mf">1e-4</span><span class="p">,</span> <span class="n">batch_size</span><span class="o">=</span><span class="mi">1000</span><span class="p">)</span>
-
-<span class="n">a</span> <span class="o">=</span> <span class="n">data_layer</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;feature_a&#39;</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">200</span><span class="p">)</span>
-<span class="n">b</span> <span class="o">=</span> <span class="n">data_layer</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;feature_b&#39;</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">200</span><span class="p">)</span>
-
-<span class="n">fc_param</span> <span class="o">=</span> <span class="n">ParamAttr</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;fc_param&#39;</span><span class="p">,</span> <span class="n">initial_max</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">initial_min</span><span class="o">=-</span><span class="mf">1.0</span><span class="p">)</span>
-<span class="n">bias_param</span> <span class="o">=</span> <span class="n">ParamAttr</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;bias_param&#39;</span><span class="p">,</span> <span class="n">initial_mean</span><span class="o">=</span><span class="mf">0.0</span><span class="p">,</span> <span class="n">initial_std</span><span class="o">=</span><span class="mf">0.0</span><span class="p">)</span>
-
-<span class="n">softmax_param</span> <span class="o">=</span> <span class="n">ParamAttr</span><span class="p">(</span>
-    <span class="n">name</span><span class="o">=</span><span class="s1">&#39;softmax_param&#39;</span><span class="p">,</span> <span class="n">initial_max</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">initial_min</span><span class="o">=-</span><span class="mf">1.0</span><span class="p">)</span>
-
-<span class="n">hidden_a</span> <span class="o">=</span> <span class="n">fc_layer</span><span class="p">(</span>
-    <span class="nb">input</span><span class="o">=</span><span class="n">a</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">200</span><span class="p">,</span> <span class="n">param_attr</span><span class="o">=</span><span class="n">fc_param</span><span class="p">,</span> <span class="n">bias_attr</span><span class="o">=</span><span class="n">bias_param</span><span class="p">)</span>
-<span class="n">hidden_b</span> <span class="o">=</span> <span class="n">fc_layer</span><span class="p">(</span>
-    <span class="nb">input</span><span class="o">=</span><span class="n">b</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">200</span><span class="p">,</span> <span class="n">param_attr</span><span class="o">=</span><span class="n">fc_param</span><span class="p">,</span> <span class="n">bias_attr</span><span class="o">=</span><span class="n">bias_param</span><span class="p">)</span>
-
-<span class="n">predict</span> <span class="o">=</span> <span class="n">fc_layer</span><span class="p">(</span>
-    <span class="nb">input</span><span class="o">=</span><span class="p">[</span><span class="n">hidden_a</span><span class="p">,</span> <span class="n">hidden_b</span><span class="p">],</span>
-    <span class="n">param_attr</span><span class="o">=</span><span class="p">[</span><span class="n">softmax_param</span><span class="p">,</span> <span class="n">softmax_param</span><span class="p">],</span>
-    <span class="n">bias_attr</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
-    <span class="n">size</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span>
-    <span class="n">act</span><span class="o">=</span><span class="n">SoftmaxActivation</span><span class="p">())</span>
-
-<span class="n">outputs</span><span class="p">(</span>
-    <span class="n">classification_cost</span><span class="p">(</span>
-        <span class="nb">input</span><span class="o">=</span><span class="n">predict</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="n">data_layer</span><span class="p">(</span>
-            <span class="n">name</span><span class="o">=</span><span class="s1">&#39;label&#39;</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">10</span><span class="p">)))</span>
-</pre></div>
-</div>
-<p>这里 <code class="code docutils literal"><span class="pre">hidden_a</span></code> 和 <code class="code docutils literal"><span class="pre">hidden_b</span></code> 使用了同样的parameter和bias。并且softmax层的两个输入也使用了同样的参数 <code class="code docutils literal"><span class="pre">softmax_param</span></code>。</p>
-</div>
-<div class="section" id="paddlepaddle-whl-is-not-a-supported-wheel-on-this-platform">
-<h2><a class="toc-backref" href="#id25">7. paddlepaddle*.whl is not a supported wheel on this platform.</a><a class="headerlink" href="#paddlepaddle-whl-is-not-a-supported-wheel-on-this-platform" title="永久链接至标题">¶</a></h2>
-<p>出现这个问题的主要原因是，没有找到和当前系统匹配的paddlepaddle安装包。最新的paddlepaddle python安装包支持Linux x86_64和MacOS 10.12操作系统，并安装了python 2.7和pip 9.0.1。</p>
-<p>更新 <code class="code docutils literal"><span class="pre">pip</span></code> 包的方法是:</p>
-<div class="highlight-bash"><div class="highlight"><pre><span></span>pip install --upgrade pip
-</pre></div>
-</div>
-<p>如果还不行，可以执行 <code class="code docutils literal"><span class="pre">python</span> <span class="pre">-c</span> <span class="pre">&quot;import</span> <span class="pre">pip;</span> <span class="pre">print(pip.pep425tags.get_supported())&quot;</span></code> 获取当前系统支持的python包的后缀，
-并对比是否和正在安装的后缀一致。</p>
-<p>如果系统支持的是 <code class="code docutils literal"><span class="pre">linux_x86_64</span></code> 而安装包是 <code class="code docutils literal"><span class="pre">manylinux1_x86_64</span></code> ，需要升级pip版本到最新；
-如果系统支持 <code class="code docutils literal"><span class="pre">manylinux1_x86_64</span></code> 而安装包（本地）是 <code class="code docutils literal"><span class="pre">linux_x86_64</span></code> ，可以重命名这个whl包为 <code class="code docutils literal"><span class="pre">manylinux1_x86_64</span></code> 再安装。</p>
-</div>
-<div class="section" id="python">
-<h2><a class="toc-backref" href="#id26">8.  python相关的单元测试都过不了</a><a class="headerlink" href="#python" title="永久链接至标题">¶</a></h2>
-<p>如果出现以下python相关的单元测试都过不了的情况：</p>
-<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="m">24</span> - test_PyDataProvider <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">26</span> - test_RecurrentGradientMachine <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">27</span> - test_NetworkCompare <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">28</span> - test_PyDataProvider2 <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">32</span> - test_Prediction <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">33</span> - test_Compare <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">34</span> - test_Trainer <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">35</span> - test_TrainerOnePass <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">36</span> - test_CompareTwoNets <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">37</span> - test_CompareTwoOpts <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">38</span> - test_CompareSparse <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">39</span> - test_recurrent_machine_generation <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">40</span> - test_PyDataProviderWrapper <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">41</span> - test_config_parser <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">42</span> - test_swig_api <span class="o">(</span>Failed<span class="o">)</span>
-<span class="m">43</span> - layers_test <span class="o">(</span>Failed<span class="o">)</span>
-</pre></div>
-</div>
-<p>并且查询PaddlePaddle单元测试的日志，提示：</p>
-<div class="highlight-bash"><div class="highlight"><pre><span></span>paddle package is already in your PYTHONPATH. But unittest need a clean environment.
-Please uninstall paddle package before start unittest. Try to <span class="s1">&#39;pip uninstall paddle&#39;</span>.
-</pre></div>
-</div>
-<p>解决办法是：</p>
-<ul class="simple">
-<li>卸载PaddlePaddle包 <code class="code docutils literal"><span class="pre">pip</span> <span class="pre">uninstall</span> <span class="pre">paddle</span></code>, 清理掉老旧的PaddlePaddle安装包，使得单元测试有一个干净的环境。如果PaddlePaddle包已经在python的site-packages里面，单元测试会引用site-packages里面的python包，而不是源码目录里 <code class="code docutils literal"><span class="pre">/python</span></code> 目录下的python包。同时，即便设置 <code class="code docutils literal"><span class="pre">PYTHONPATH</span></code> 到 <code class="code docutils literal"><span class="pre">/python</span></code> 也没用，因为python的搜索路径是优先已经安装的python包。</li>
-</ul>
-</div>
-<div class="section" id="docker-gpu-cuda-driver-version-is-insufficient">
-<h2><a class="toc-backref" href="#id27">9. 运行Docker GPU镜像出现 &#8220;CUDA driver version is insufficient&#8221;</a><a class="headerlink" href="#docker-gpu-cuda-driver-version-is-insufficient" title="永久链接至标题">¶</a></h2>
-<p>用户在使用PaddlePaddle GPU的Docker镜像的时候，常常出现 <cite>Cuda Error: CUDA driver version is insufficient for CUDA runtime version</cite>, 原因在于没有把机器上CUDA相关的驱动和库映射到容器内部。
-具体的解决方法是：</p>
-<div class="highlight-bash"><div class="highlight"><pre><span></span>$ <span class="nb">export</span> <span class="nv">CUDA_SO</span><span class="o">=</span><span class="s2">&quot;</span><span class="k">$(</span><span class="se">\l</span>s usr/lib64/libcuda* <span class="p">|</span> xargs -I<span class="o">{}</span> <span class="nb">echo</span> <span class="s1">&#39;-v {}:{}&#39;</span><span class="k">)</span><span class="s2"> </span><span class="k">$(</span><span class="se">\l</span>s /usr/lib64/libnvidia* <span class="p">|</span> xargs -I<span class="o">{}</span> <span class="nb">echo</span> <span class="s1">&#39;-v {}:{}&#39;</span><span class="k">)</span><span class="s2">&quot;</span>
-$ <span class="nb">export</span> <span class="nv">DEVICES</span><span class="o">=</span><span class="k">$(</span><span class="se">\l</span>s /dev/nvidia* <span class="p">|</span> xargs -I<span class="o">{}</span> <span class="nb">echo</span> <span class="s1">&#39;--device {}:{}&#39;</span><span class="k">)</span>
-$ docker run <span class="si">${</span><span class="nv">CUDA_SO</span><span class="si">}</span> <span class="si">${</span><span class="nv">DEVICES</span><span class="si">}</span> -it paddledev/paddlepaddle:latest-gpu
-</pre></div>
-</div>
-<p>更多关于Docker的安装与使用, 请参考 <a class="reference external" href="http://www.paddlepaddle.org/doc_cn/build_and_install/install/docker_install.html">PaddlePaddle Docker 文档</a> 。</p>
-</div>
-<div class="section" id="cmake-pythonlibspythoninterp">
-<h2><a class="toc-backref" href="#id28">10. CMake源码编译, 找到的PythonLibs和PythonInterp版本不一致</a><a class="headerlink" href="#cmake-pythonlibspythoninterp" title="永久链接至标题">¶</a></h2>
-<p>这是目前CMake寻找Python的逻辑存在缺陷，如果系统安装了多个Python版本，CMake找到的Python库和Python解释器版本可能有不一致现象，导致编译PaddlePaddle失败。正确的解决方法是，
-用户强制指定特定的Python版本，具体操作如下：</p>
-<blockquote>
-<div><div class="highlight-bash"><div class="highlight"><pre><span></span>cmake .. -DPYTHON_EXECUTABLE<span class="o">=</span>&lt;exc_path&gt; -DPYTHON_LIBRARY<span class="o">=</span>&lt;lib_path&gt;  -DPYTHON_INCLUDE_DIR<span class="o">=</span>&lt;inc_path&gt;
-</pre></div>
-</div>
-</div></blockquote>
-<p>用户需要指定本机上Python的路径：<code class="docutils literal"><span class="pre">&lt;exc_path&gt;</span></code>, <code class="docutils literal"><span class="pre">&lt;lib_path&gt;</span></code>, <code class="docutils literal"><span class="pre">&lt;inc_path&gt;</span></code></p>
-</div>
-<div class="section" id="cmake-paddle0-0-0">
-<h2><a class="toc-backref" href="#id29">11. CMake源码编译，Paddle版本号为0.0.0</a><a class="headerlink" href="#cmake-paddle0-0-0" title="永久链接至标题">¶</a></h2>
-<p>如果运行 <code class="code docutils literal"><span class="pre">paddle</span> <span class="pre">version</span></code>, 出现 <code class="code docutils literal"><span class="pre">PaddlePaddle</span> <span class="pre">0.0.0</span></code>；或者运行 <code class="code docutils literal"><span class="pre">cmake</span> <span class="pre">..</span></code>，出现</p>
-<div class="highlight-bash"><div class="highlight"><pre><span></span>CMake Warning at cmake/version.cmake:20 <span class="o">(</span>message<span class="o">)</span>:
-  Cannot add paddle version from git tag
-</pre></div>
-</div>
-<p>那么用户需要拉取所有的远程分支到本机，命令为 <code class="code docutils literal"><span class="pre">git</span> <span class="pre">fetch</span> <span class="pre">upstream</span></code>，然后重新cmake即可。</p>
-</div>
-<div class="section" id="a-protocol-message-was-rejected-because-it-was-too-big">
-<h2><a class="toc-backref" href="#id30">12. A protocol message was rejected because it was too big</a><a class="headerlink" href="#a-protocol-message-was-rejected-because-it-was-too-big" title="永久链接至标题">¶</a></h2>
-<p>如果在训练NLP相关模型时，出现以下错误：</p>
-<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">[</span>libprotobuf ERROR google/protobuf/io/coded_stream.cc:171<span class="o">]</span> A protocol message was rejected because it was too big <span class="o">(</span>more than <span class="m">67108864</span> bytes<span class="o">)</span>.  To increase the limit <span class="o">(</span>or to disable these warnings<span class="o">)</span>, see CodedInputStream::SetTotalBytesLimit<span class="o">()</span> in google/protobuf/io/coded_stream.h.
-F1205 <span class="m">14</span>:59:50.295174 <span class="m">14703</span> TrainerConfigHelper.cpp:59<span class="o">]</span> Check failed: m-&gt;conf.ParseFromString<span class="o">(</span>configProtoStr<span class="o">)</span>
-</pre></div>
-</div>
-<p>可能的原因是：传给dataprovider的某一个args过大，一般是由于直接传递大字典导致的。错误的define_py_data_sources2类似：</p>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">src_dict</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">()</span>
-<span class="k">for</span> <span class="n">line_count</span><span class="p">,</span> <span class="n">line</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="nb">open</span><span class="p">(</span><span class="n">src_dict_path</span><span class="p">,</span> <span class="s2">&quot;r&quot;</span><span class="p">)):</span>
-   <span class="n">src_dict</span><span class="p">[</span><span class="n">line</span><span class="o">.</span><span class="n">strip</span><span class="p">()]</span> <span class="o">=</span> <span class="n">line_count</span>
-
-<span class="n">define_py_data_sources2</span><span class="p">(</span>
-   <span class="n">train_list</span><span class="p">,</span>
-   <span class="n">test_list</span><span class="p">,</span>
-   <span class="n">module</span><span class="o">=</span><span class="s2">&quot;dataprovider&quot;</span><span class="p">,</span>
-   <span class="n">obj</span><span class="o">=</span><span class="s2">&quot;process&quot;</span><span class="p">,</span>
-   <span class="n">args</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;src_dict&quot;</span><span class="p">:</span> <span class="n">src_dict</span><span class="p">})</span>
-</pre></div>
-</div>
-<p>解决方案是：将字典的地址作为args传给dataprovider，然后在dataprovider里面根据该地址加载字典。即define_py_data_sources2应改为：</p>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">define_py_data_sources2</span><span class="p">(</span>
-   <span class="n">train_list</span><span class="p">,</span>
-   <span class="n">test_list</span><span class="p">,</span>
-   <span class="n">module</span><span class="o">=</span><span class="s2">&quot;dataprovider&quot;</span><span class="p">,</span>
-   <span class="n">obj</span><span class="o">=</span><span class="s2">&quot;process&quot;</span><span class="p">,</span>
-   <span class="n">args</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;src_dict_path&quot;</span><span class="p">:</span> <span class="n">src_dict_path</span><span class="p">})</span>
-</pre></div>
-</div>
-<p>完整源码可参考 <a class="reference external" href="https://github.com/PaddlePaddle/Paddle/tree/develop/demo/seqToseq">seqToseq</a> 示例。</p>
-</div>
-<div class="section" id="gpu">
-<h2><a class="toc-backref" href="#id31">13. 如何指定GPU设备</a><a class="headerlink" href="#gpu" title="永久链接至标题">¶</a></h2>
-<p>例如机器上有4块GPU，编号从0开始，指定使用2、3号GPU：</p>
-<ul class="simple">
-<li>方式1：通过 <a class="reference external" href="http://www.acceleware.com/blog/cudavisibledevices-masking-gpus">CUDA_VISIBLE_DEVICES</a> 环境变量来指定特定的GPU。</li>
-</ul>
-<div class="highlight-bash"><div class="highlight"><pre><span></span>env <span class="nv">CUDA_VISIBLE_DEVICES</span><span class="o">=</span><span class="m">2</span>,3 paddle train --use_gpu<span class="o">=</span><span class="nb">true</span> --trainer_count<span class="o">=</span><span class="m">2</span>
-</pre></div>
-</div>
-<ul class="simple">
-<li>方式2：通过命令行参数 <code class="docutils literal"><span class="pre">--gpu_id</span></code> 指定。</li>
-</ul>
-<div class="highlight-bash"><div class="highlight"><pre><span></span>paddle train --use_gpu<span class="o">=</span><span class="nb">true</span> --trainer_count<span class="o">=</span><span class="m">2</span> --gpu_id<span class="o">=</span><span class="m">2</span>
-</pre></div>
-</div>
-</div>
-<div class="section" id="floating-point-exception">
-<h2><a class="toc-backref" href="#id32">14. 训练过程中出现 <code class="code docutils literal"><span class="pre">Floating</span> <span class="pre">point</span> <span class="pre">exception</span></code>, 训练因此退出怎么办?</a><a class="headerlink" href="#floating-point-exception" title="永久链接至标题">¶</a></h2>
-<p>Paddle二进制在运行时捕获了浮点数异常，只要出现浮点数异常(即训练过程中出现NaN或者Inf)，立刻退出。浮点异常通常的原因是浮点数溢出、除零等问题。
-主要原因包括两个方面:</p>
-<ul class="simple">
-<li>训练过程中参数或者训练过程中的梯度尺度过大，导致参数累加，乘除等时候，导致了浮点数溢出。</li>
-<li>模型一直不收敛，发散到了一个数值特别大的地方。</li>
-<li>训练数据有问题，导致参数收敛到了一些奇异的情况。或者输入数据尺度过大，有些特征的取值达到数百万，这时进行矩阵乘法运算就可能导致浮点数溢出。</li>
-</ul>
-<p>这里有两种有效的解决方法：</p>
-<ol class="arabic simple">
-<li>设置 <code class="code docutils literal"><span class="pre">gradient_clipping_threshold</span></code> 参数，示例代码如下：</li>
-</ol>
-<div class="highlight-python"><div class="highlight"><pre><span></span>
-</pre></div>
-</div>
-<dl class="docutils">
-<dt>optimizer = paddle.optimizer.RMSProp(</dt>
-<dd>learning_rate=1e-3,
-gradient_clipping_threshold=10.0,
-regularization=paddle.optimizer.L2Regularization(rate=8e-4))</dd>
-</dl>
-<p>具体可以参考  <a class="reference external" href="https://github.com/PaddlePaddle/models/blob/develop/nmt_without_attention/train.py#L35">nmt_without_attention</a> 示例。</p>
-<ol class="arabic simple" start="2">
-<li>设置 <code class="code docutils literal"><span class="pre">error_clipping_threshold</span></code> 参数，示例代码如下：</li>
-</ol>
-<div class="highlight-python"><div class="highlight"><pre><span></span>
-</pre></div>
-</div>
-<dl class="docutils">
-<dt>decoder_inputs = paddle.layer.fc(</dt>
-<dd><p class="first">act=paddle.activation.Linear(),
-size=decoder_size * 3,
-bias_attr=False,
-input=[context, current_word],
-layer_attr=paddle.attr.ExtraLayerAttribute(</p>
-<blockquote class="last">
-<div>error_clipping_threshold=100.0))</div></blockquote>
-</dd>
-</dl>
-<p>完整代码可以参考示例 <a class="reference external" href="https://github.com/PaddlePaddle/book/blob/develop/08.machine_translation/train.py#L66">machine translation</a> 。</p>
-<p>两种方法的区别：</p>
-<ol class="arabic simple">
-<li>两者都是对梯度的截断，但截断时机不同，前者在 <code class="code docutils literal"><span class="pre">optimzier</span></code> 更新网络参数时应用；后者在激活函数反向计算时被调用；</li>
-<li>截断对象不同：前者截断可学习参数的梯度，后者截断回传给前层的梯度;</li>
-</ol>
-<p>除此之外，还可以通过减小学习律或者对数据进行归一化处理来解决这类问题。</p>
-</div>
-<div class="section" id="import-paddle-v2-as-paddle-importerror-no-module-named-v2">
-<h2><a class="toc-backref" href="#id33">15. 编译安装后执行 import paddle.v2 as paddle 报ImportError: No module named v2</a><a class="headerlink" href="#import-paddle-v2-as-paddle-importerror-no-module-named-v2" title="永久链接至标题">¶</a></h2>
-<p>先查看一下是否曾经安装过paddle v1版本，有的话需要先卸载：</p>
-<p>pip uninstall py_paddle paddle</p>
-<p>然后安装paddle的python环境, 在build目录下执行</p>
-<p>pip install python/dist/paddle*.whl &amp;&amp; pip install ../paddle/dist/py_paddle*.whl</p>
-</div>
-<div class="section" id="id9">
-<h2><a class="toc-backref" href="#id34">16. PaddlePaddle存储的参数格式是什么，如何和明文进行相互转化</a><a class="headerlink" href="#id9" title="永久链接至标题">¶</a></h2>
-<p>PaddlePaddle保存的模型参数文件内容由16字节头信息和网络参数两部分组成。头信息中，1~4字节表示PaddlePaddle版本信息，请直接填充0；5~8字节表示每个参数占用的字节数，当保存的网络参数为float类型时为4，double类型时为8；9~16字节表示保存的参数总个数。</p>
-<p>将PaddlePaddle保存的模型参数还原回明文时，可以使用相应数据类型的 <code class="code docutils literal"><span class="pre">numpy.array</span></code> 加载具体网络参数，此时可以跳过PaddlePaddle模型参数文件的头信息。若在PaddlePaddle编译时，未指定按照double精度编译，默认情况下按照float精度计算，保存的参数也是float类型。这时在使用 <code class="code docutils literal"><span class="pre">numpy.array</span></code> 时，一般设置 <code class="code docutils literal"><span class="pre">dtype=float32</span></code> 。示例如下：</p>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">read_parameter</span><span class="p">(</span><span class="n">fname</span><span class="p">,</span> <span class="n">width</span><span class="p">):</span>
-    <span class="n">s</span> <span class="o">=</span> <span class="nb">open</span><span class="p">(</span><span class="n">fname</span><span class="p">)</span><span class="o">.</span><span class="n">read</span><span class="p">()</span>
-    <span class="c1"># skip header</span>
-    <span class="n">vec</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">fromstring</span><span class="p">(</span><span class="n">s</span><span class="p">[</span><span class="mi">16</span><span class="p">:],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
-    <span class="c1"># width is the size of the corresponding layer</span>
-    <span class="n">np</span><span class="o">.</span><span class="n">savetxt</span><span class="p">(</span><span class="n">fname</span> <span class="o">+</span> <span class="s2">&quot;.csv&quot;</span><span class="p">,</span> <span class="n">vec</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">width</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">),</span>
-            <span class="n">fmt</span><span class="o">=</span><span class="s2">&quot;</span><span class="si">%.6f</span><span class="s2">&quot;</span><span class="p">,</span> <span class="n">delimiter</span><span class="o">=</span><span class="s2">&quot;,&quot;</span><span class="p">)</span>
-</pre></div>
-</div>
-<p>将明文参数转化为PaddlePaddle可加载的模型参数时，首先构造头信息，再写入网络参数。下面的代码将随机生成的矩阵转化为可以被PaddlePaddle加载的模型参数。</p>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">gen_rand_param</span><span class="p">(</span><span class="n">param_file</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">need_trans</span><span class="p">):</span>
-    <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">seed</span><span class="p">()</span>
-    <span class="n">header</span> <span class="o">=</span> <span class="n">struct</span><span class="o">.</span><span class="n">pack</span><span class="p">(</span><span class="s2">&quot;iil&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="n">height</span> <span class="o">*</span> <span class="n">width</span><span class="p">)</span>
-    <span class="n">param</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">))</span>
-    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">param_file</span><span class="p">,</span> <span class="s2">&quot;w&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">fparam</span><span class="p">:</span>
-        <span class="n">fparam</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="n">header</span> <span class="o">+</span> <span class="n">param</span><span class="o">.</span><span class="n">tostring</span><span class="p">())</span>
-</pre></div>
-</div>
-</div>
-<div class="section" id="id10">
-<h2><a class="toc-backref" href="#id35">17. 如何加载预训练参数</a><a class="headerlink" href="#id10" title="永久链接至标题">¶</a></h2>
-<ul class="simple">
-<li>对加载预训练参数的层，设置其参数属性 <code class="code docutils literal"><span class="pre">is_static=True</span></code>，使该层的参数在训练过程中保持不变。以embedding层为例，代码如下：</li>
-</ul>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">emb_para</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">attr</span><span class="o">.</span><span class="n">Param</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="n">is_static</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
-<span class="n">paddle</span><span class="o">.</span><span class="n">layer</span><span class="o">.</span><span class="n">embedding</span><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="n">word_dim</span><span class="p">,</span> <span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">param_attr</span><span class="o">=</span><span class="n">emb_para</span><span class="p">)</span>
-</pre></div>
-</div>
-<ul class="simple">
-<li>从模型文件将预训练参数载入 <code class="code docutils literal"><span class="pre">numpy.array</span></code>，在创建parameters后，使用 <code class="code docutils literal"><span class="pre">parameters.set()</span></code> 加载预训练参数。PaddlePaddle保存的模型参数文件前16字节为头信息，用户将参数载入 <code class="code docutils literal"><span class="pre">numpy.array</span></code> 时须从第17字节开始。以embedding层为例，代码如下：</li>
-</ul>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">load_parameter</span><span class="p">(</span><span class="n">file_name</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">):</span>
-    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">file_name</span><span class="p">,</span> <span class="s1">&#39;rb&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
-        <span class="n">f</span><span class="o">.</span><span class="n">read</span><span class="p">(</span><span class="mi">16</span><span class="p">)</span>  <span class="c1"># skip header.</span>
-        <span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">fromfile</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">)</span>
-
-<span class="n">parameters</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">parameters</span><span class="o">.</span><span class="n">create</span><span class="p">(</span><span class="n">my_cost</span><span class="p">)</span>
-<span class="n">parameters</span><span class="o">.</span><span class="n">set</span><span class="p">(</span><span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="n">load_parameter</span><span class="p">(</span><span class="n">emb_param_file</span><span class="p">,</span> <span class="mi">30000</span><span class="p">,</span> <span class="mi">256</span><span class="p">))</span>
-</pre></div>
-</div>
-</div>
-<div class="section" id="id11">
-<h2><a class="toc-backref" href="#id36">18. 集群多节点训练，日志中保存均为网络通信类错误</a><a class="headerlink" href="#id11" title="永久链接至标题">¶</a></h2>
-<p>集群多节点训练，日志报错为网络通信类错误，比如 <code class="code docutils literal"><span class="pre">Connection</span> <span class="pre">reset</span> <span class="pre">by</span> <span class="pre">peer</span></code> 等。
-此类报错通常是由于某一个节点的错误导致这个节点的训练进程退出，从而引发其他节点无法连接导致，可以参考下面的步骤排查：</p>
-<ul class="simple">
-<li>从 <code class="code docutils literal"><span class="pre">train.log</span></code> ， <code class="code docutils literal"><span class="pre">server.log</span></code> 找到最早报错的地方，查看是否是其他错误引发的报错（比如FPE，内存不足，磁盘空间不足等）。</li>
-<li>如果发现最早的报错就是网络通信的问题，很有可能是非独占方式执行导致的端口冲突，可以联系OP，看当前MPI集群是否支持resource=full参数提交，如果支持增加此参数提交，并更换job 端口。</li>
-<li>如果当前MPI集群并不支持任务独占模式，可以联系OP是否可以更换集群或升级当前集群。</li>
-</ul>
-</div>
-<div class="section" id="infer-layer">
-<h2><a class="toc-backref" href="#id37">19.  如何调用 infer 接口输出多个layer的预测结果</a><a class="headerlink" href="#infer-layer" title="永久链接至标题">¶</a></h2>
-<ul class="simple">
-<li>将需要输出的层作为 <code class="code docutils literal"><span class="pre">paddle.inference.Inference()</span></code> 接口的 <code class="code docutils literal"><span class="pre">output_layer</span></code> 参数输入，代码如下：</li>
-</ul>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">inferer</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">inference</span><span class="o">.</span><span class="n">Inference</span><span class="p">(</span><span class="n">output_layer</span><span class="o">=</span><span class="p">[</span><span class="n">layer1</span><span class="p">,</span> <span class="n">layer2</span><span class="p">],</span> <span class="n">parameters</span><span class="o">=</span><span class="n">parameters</span><span class="p">)</span>
-</pre></div>
-</div>
-<ul class="simple">
-<li>指定要输出的字段进行输出。以输出 <code class="code docutils literal"><span class="pre">value</span></code> 字段为例，代码如下：</li>
-</ul>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">out</span> <span class="o">=</span> <span class="n">inferer</span><span class="o">.</span><span class="n">infer</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">data_batch</span><span class="p">,</span> <span class="n">field</span><span class="o">=</span><span class="p">[</span><span class="s2">&quot;value&quot;</span><span class="p">])</span>
-</pre></div>
-</div>
-<p>需要注意的是：</p>
-<ul class="simple">
-<li>如果指定了2个layer作为输出层，实际上需要的输出结果是两个矩阵；</li>
-<li>假设第一个layer的输出A是一个 N1 * M1 的矩阵，第二个 Layer 的输出B是一个 N2 * M2 的矩阵；</li>
-<li>paddle.v2 默认会将A和B 横向拼接，当N1 和 N2 大小不一样时，会报如下的错误：</li>
-</ul>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="ne">ValueError</span><span class="p">:</span> <span class="nb">all</span> <span class="n">the</span> <span class="nb">input</span> <span class="n">array</span> <span class="n">dimensions</span> <span class="k">except</span> <span class="k">for</span> <span class="n">the</span> <span class="n">concatenation</span> <span class="n">axis</span> <span class="n">must</span> <span class="n">match</span> <span class="n">exactly</span>
-</pre></div>
-</div>
-<p>多个层的输出矩阵的高度不一致导致拼接失败，这种情况常常发生在：</p>
-<ul class="simple">
-<li>同时输出序列层和非序列层；</li>
-<li>多个输出层处理多个不同长度的序列;</li>
-</ul>
-<p>此时可以在调用infer接口时通过设置 <code class="code docutils literal"><span class="pre">flatten_result=False</span></code> , 跳过“拼接”步骤，来解决上面的问题。这时，infer接口的返回值是一个python list:</p>
-<ul class="simple">
-<li>list 中元素的个数等于网络中输出层的个数；</li>
-<li>list 中每个元素是一个layer的输出结果矩阵，类型是numpy的ndarray；</li>
-<li>每一个layer输出矩阵的高度，在非序列输入时：等于样本数；序列输入时等于：输入序列中元素的总数；宽度等于配置中layer的size；</li>
-</ul>
-</div>
-<div class="section" id="paddle-layer-memory-name">
-<h2><a class="toc-backref" href="#id38">20. <code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 的参数 <code class="code docutils literal"><span class="pre">name</span></code> 如何使用</a><a class="headerlink" href="#paddle-layer-memory-name" title="永久链接至标题">¶</a></h2>
-<ul class="simple">
-<li><code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 用于获取特定layer上一时间步的输出，该layer是通过参数 <code class="code docutils literal"><span class="pre">name</span></code> 指定，即，<code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 会关联参数 <code class="code docutils literal"><span class="pre">name</span></code> 取值相同的layer，并将该layer上一时间步的输出作为自身当前时间步的输出。</li>
-<li>PaddlePaddle的所有layer都有唯一的name，用户通过参数 <code class="code docutils literal"><span class="pre">name</span></code> 设定，当用户没有显式设定时，PaddlePaddle会自动设定。而 <code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 不是真正的layer，其name由参数 <code class="code docutils literal"><span class="pre">memory_name</span></code> 设定，当用户没有显式设定时，PaddlePaddle会自动设定。<code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 的参数 <code class="code docutils literal"><span class="pre">name</span></code> 用于指定其要关联的layer，需要用户显式设定。</li>
-</ul>
-</div>
-<div class="section" id="drop-out">
-<h2><a class="toc-backref" href="#id39">21. 两种使用 drop_out 的方法有何区别？</a><a class="headerlink" href="#drop-out" title="永久链接至标题">¶</a></h2>
+<h1>FAQ<a class="headerlink" href="#faq" title="永久链接至标题">¶</a></h1>
+<div class="toctree-wrapper compound">
 <ul>
-<li><p class="first">在PaddlePaddle中使用dropout有两种方式</p>
-<ul class="simple">
-<li>在相应layer的 <code class="code docutils literal"><span class="pre">layer_atter</span></code> 设置 <code class="code docutils literal"><span class="pre">drop_rate</span></code>，以 <code class="code docutils literal"><span class="pre">paddle.layer.fc</span></code> 为例，代码如下：</li>
+<li class="toctree-l1"><a class="reference internal" href="build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l1"><a class="reference internal" href="model/index_cn.html">模型配置</a></li>
+<li class="toctree-l1"><a class="reference internal" href="parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l1"><a class="reference internal" href="local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l1"><a class="reference internal" href="cluster/index_cn.html">集群训练与预测</a></li>
 </ul>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">fc</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">layer</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="n">layer_attr</span><span class="o">=</span><span class="n">paddle</span><span class="o">.</span><span class="n">attr</span><span class="o">.</span><span class="n">ExtraLayerAttribute</span><span class="p">(</span><span class="n">drop_rate</span><span class="o">=</span><span class="mf">0.5</span><span class="p">))</span>
-</pre></div>
-</div>
-<ul class="simple">
-<li>使用 <code class="code docutils literal"><span class="pre">paddle.layer.dropout</span></code>，以 <code class="code docutils literal"><span class="pre">paddle.layer.fc</span></code> 为例，代码如下：</li>
-</ul>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">fc</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">layer</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">)</span>
-<span class="n">drop_fc</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">layer</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">fc</span><span class="p">,</span> <span class="n">dropout_rate</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span>
-</pre></div>
-</div>
-</li>
-<li><p class="first"><code class="code docutils literal"><span class="pre">paddle.layer.dropout</span></code> 实际上使用了 <code class="code docutils literal"><span class="pre">paddle.layer.add_to</span></code>，并在该layer里采用第一种方式设置 <code class="code docutils literal"><span class="pre">drop_rate</span></code> 来使用dropout的。这种方式对内存消耗较大。</p>
-</li>
-<li><p class="first">PaddlePaddle在激活函数里实现dropout，而不是在layer里实现。</p>
-</li>
-<li><p class="first"><code class="code docutils literal"><span class="pre">paddle.layer.lstmemory</span></code>、<code class="code docutils literal"><span class="pre">paddle.layer.grumemory</span></code>、<code class="code docutils literal"><span class="pre">paddle.layer.recurrent</span></code> 不是通过一般的方式来实现对输出的激活，所以不能采用第一种方式在这几个layer里设置 <code class="code docutils literal"><span class="pre">drop_rate</span></code> 来使用dropout。若要对这几个layer使用dropout，可采用第二种方式，即使用 <code class="code docutils literal"><span class="pre">paddle.layer.dropout</span></code>。</p>
-</li>
-</ul>
-</div>
-<div class="section" id="learning-rate-annealing">
-<h2><a class="toc-backref" href="#id40">22. 如何设置学习率退火（learning rate annealing）</a><a class="headerlink" href="#learning-rate-annealing" title="永久链接至标题">¶</a></h2>
-<p>在相应的优化算法里设置learning_rate_schedule及相关参数，以使用Adam算法为例，代码如下：</p>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span>
-    <span class="n">learning_rate</span><span class="o">=</span><span class="mf">1e-3</span><span class="p">,</span>
-    <span class="n">learning_rate_decay_a</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
-    <span class="n">learning_rate_decay_b</span><span class="o">=</span><span class="mf">0.75</span><span class="p">,</span>
-    <span class="n">learning_rate_schedule</span><span class="o">=</span><span class="s2">&quot;poly&quot;</span><span class="p">,)</span>
-</pre></div>
-</div>
-<p>PaddlePaddle目前支持8种learning_rate_schedule，这8种learning_rate_schedule及其对应学习率计算方式如下：</p>
-<ul>
-<li><p class="first">&#8220;constant&#8221;</p>
-<p>lr = learning_rate</p>
-</li>
-<li><p class="first">&#8220;poly&#8221;</p>
-<p>lr = learning_rate * pow(1 + learning_rate_decay_a * num_samples_processed, -learning_rate_decay_b)</p>
-<p>其中，num_samples_processed为已训练样本数，下同。</p>
-</li>
-<li><p class="first">&#8220;caffe_poly&#8221;</p>
-<p>lr = learning_rate * pow(1.0 - num_samples_processed / learning_rate_decay_a, learning_rate_decay_b)</p>
-</li>
-<li><p class="first">&#8220;exp&#8221;</p>
-<p>lr = learning_rate * pow(learning_rate_decay_a, num_samples_processed / learning_rate_decay_b)</p>
-</li>
-<li><p class="first">&#8220;discexp&#8221;</p>
-<p>lr = learning_rate * pow(learning_rate_decay_a, floor(num_samples_processed / learning_rate_decay_b))</p>
-</li>
-<li><p class="first">&#8220;linear&#8221;</p>
-<p>lr = max(learning_rate - learning_rate_decay_a * num_samples_processed, learning_rate_decay_b)</p>
-</li>
-<li><p class="first">&#8220;manual&#8221;</p>
-<p>这是一种按已训练样本数分段取值的学习率退火方法。使用该learning_rate_schedule时，用户通过参数 <code class="code docutils literal"><span class="pre">learning_rate_args</span></code> 设置学习率衰减因子分段函数，当前的学习率为所设置 <code class="code docutils literal"><span class="pre">learning_rate</span></code> 与当前的衰减因子的乘积。以使用Adam算法为例，代码如下：</p>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span>
-    <span class="n">learning_rate</span><span class="o">=</span><span class="mf">1e-3</span><span class="p">,</span>
-    <span class="n">learning_rate_schedule</span><span class="o">=</span><span class="s2">&quot;manual&quot;</span><span class="p">,</span>
-    <span class="n">learning_rate_args</span><span class="o">=</span><span class="s2">&quot;1000:1.0,2000:0.9,3000:0.8&quot;</span><span class="p">,)</span>
-</pre></div>
-</div>
-<p>在该示例中，当已训练样本数小于等于1000时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">1.0</span></code>；当已训练样本数大于1000小于等于2000时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">0.9</span></code>；当已训练样本数大于2000时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">0.8</span></code>。</p>
-</li>
-<li><p class="first">&#8220;pass_manual&#8221;</p>
-<p>这是一种按已训练pass数分段取值的学习率退火方法。使用该learning_rate_schedule时，用户通过参数 <code class="code docutils literal"><span class="pre">learning_rate_args</span></code> 设置学习率衰减因子分段函数，当前的学习率为所设置 <code class="code docutils literal"><span class="pre">learning_rate</span></code> 与当前的衰减因子的乘积。以使用Adam算法为例，代码如下：</p>
-<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span>
-    <span class="n">learning_rate</span><span class="o">=</span><span class="mf">1e-3</span><span class="p">,</span>
-    <span class="n">learning_rate_schedule</span><span class="o">=</span><span class="s2">&quot;manual&quot;</span><span class="p">,</span>
-    <span class="n">learning_rate_args</span><span class="o">=</span><span class="s2">&quot;1:1.0,2:0.9,3:0.8&quot;</span><span class="p">,)</span>
-</pre></div>
-</div>
-<p>在该示例中，当已训练pass数小于等于1时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">1.0</span></code>；当已训练pass数大于1小于等于2时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">0.9</span></code>；当已训练pass数大于2时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">0.8</span></code>。</p>
-</li>
-</ul>
-</div>
-<div class="section" id="duplicated-layer-name">
-<h2><a class="toc-backref" href="#id41">23. 出现 <code class="code docutils literal"><span class="pre">Duplicated</span> <span class="pre">layer</span> <span class="pre">name</span></code> 错误怎么办</a><a class="headerlink" href="#duplicated-layer-name" title="永久链接至标题">¶</a></h2>
-<p>出现该错误的原因一般是用户对不同layer的参数 <code class="code docutils literal"><span class="pre">name</span></code> 设置了相同的取值。遇到该错误时，先找出参数 <code class="code docutils literal"><span class="pre">name</span></code> 取值相同的layer，然后将这些layer的参数 <code class="code docutils literal"><span class="pre">name</span></code> 设置为不同的值。</p>
-</div>
-<div class="section" id="paddlepaddle-recurrent-layer">
-<h2><a class="toc-backref" href="#id42">24. PaddlePaddle 中不同的 recurrent layer 的区别</a><a class="headerlink" href="#paddlepaddle-recurrent-layer" title="永久链接至标题">¶</a></h2>
-<p>以LSTM为例，在PaddlePaddle中包含以下 recurrent layer：</p>
-<ul class="simple">
-<li><code class="code docutils literal"><span class="pre">paddle.layer.lstmemory</span></code></li>
-<li><code class="code docutils literal"><span class="pre">paddle.networks.simple_lstm</span></code></li>
-<li><code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_group</span></code></li>
-<li><code class="code docutils literal"><span class="pre">paddle.networks.bidirectional_lstm</span></code></li>
-</ul>
-<p>按照具体实现方式可以归纳为2类：</p>
-<ol class="arabic simple">
-<li>由 recurrent_group 实现的 recurrent layer：</li>
-</ol>
-<blockquote>
-<div><ul class="simple">
-<li>用户在使用这一类recurrent layer时，可以访问由recurrent unit在一个时间步内计算得到的中间值（例如：hidden states, memory cells等）；</li>
-<li>上述的 <code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_group</span></code> 是这一类的 recurrent layer ；</li>
-</ul>
-</div></blockquote>
-<ol class="arabic simple" start="2">
-<li>将recurrent layer作为一个整体来实现：</li>
-</ol>
-<blockquote>
-<div><ul class="simple">
-<li>用户在使用这一类recurrent layer，只能访问它们的输出值；</li>
-<li>上述的 <code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_group</span></code> 、 <code class="code docutils literal"><span class="pre">paddle.networks.simple_lstm</span></code> 和 <code class="code docutils literal"><span class="pre">paddle.networks.bidirectional_lstm</span></code> 属于这一类的实现；</li>
-</ul>
-</div></blockquote>
-<p>将recurrent layer作为一个整体来实现， 能够针对CPU和GPU的计算做更多优化， 所以相比于recurrent group的实现方式， 第二类 recurrent layer 计算效率更高。 在实际应用中，如果用户不需要访问LSTM的中间变量，而只需要获得recurrent layer计算的输出，我们建议使用第二类实现。</p>
-<p>此外，关于LSTM, PaddlePaddle中还包含 <code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_unit</span></code> 这一计算单元：</p>
-<blockquote>
-<div><ul class="simple">
-<li>不同于上述介绍的recurrent layer , <code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_unit</span></code> 定义了LSTM单元在一个时间步内的计算过程，它并不是一个完整的recurrent layer，也不能接收序列数据作为输入；</li>
-<li><code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_unit</span></code> 只能在recurrent_group中作为step function使用；</li>
-</ul>
-</div></blockquote>
 </div>
 </div>
 
@@ -819,6 +213,8 @@ layer_attr=paddle.attr.ExtraLayerAttribute(</p>
   
     <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
       
+        <a href="build_and_install/index_cn.html" class="btn btn-neutral float-right" title="编译安装与单元测试" accesskey="n">Next <span class="fa fa-arrow-circle-right"></span></a>
+      
       
         <a href="../api/v2/run_logic.html" class="btn btn-neutral" title="Training and Inference" accesskey="p"><span class="fa fa-arrow-circle-left"></span> Previous</a>
       
diff --git a/develop/doc_cn/faq/local/index_cn.html b/develop/doc_cn/faq/local/index_cn.html
new file mode 100644
index 0000000000000000000000000000000000000000..92781c7812206af0fb3e03811d52f3894f73aa3f
--- /dev/null
+++ b/develop/doc_cn/faq/local/index_cn.html
@@ -0,0 +1,518 @@
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>本地训练与预测 &mdash; PaddlePaddle  文档</title>
+  
+
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  
+  
+    <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  
+
+  
+  
+        <link rel="index" title="索引"
+              href="../../genindex.html"/>
+        <link rel="search" title="搜索" href="../../search.html"/>
+    <link rel="top" title="PaddlePaddle  文档" href="../../index.html"/>
+        <link rel="up" title="FAQ" href="../index_cn.html"/>
+        <link rel="next" title="集群训练与预测" href="../cluster/index_cn.html"/>
+        <link rel="prev" title="参数设置" href="../parameter/index_cn.html"/> 
+
+  <link rel="stylesheet" href="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/css/perfect-scrollbar.min.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/override.css" type="text/css" />
+  <script>
+  var _hmt = _hmt || [];
+  (function() {
+    var hm = document.createElement("script");
+    hm.src = "//hm.baidu.com/hm.js?b9a314ab40d04d805655aab1deee08ba";
+    var s = document.getElementsByTagName("script")[0]; 
+    s.parentNode.insertBefore(hm, s);
+  })();
+  </script>
+
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+</head>
+
+<body class="wy-body-for-nav" role="document">
+
+  
+  <header class="site-header">
+    <div class="site-logo">
+      <a href="/"><img src="../../_static/images/PP_w.png"></a>
+    </div>
+    <div class="site-nav-links">
+      <div class="site-menu">
+        <a class="fork-on-github" href="https://github.com/PaddlePaddle/Paddle" target="_blank"><i class="fa fa-github"></i>Fork me on Github</a>
+        <div class="language-switcher dropdown">
+          <a type="button" data-toggle="dropdown">
+            <span>English</span>
+            <i class="fa fa-angle-up"></i>
+            <i class="fa fa-angle-down"></i>
+          </a>
+          <ul class="dropdown-menu">
+            <li><a href="/doc_cn">中文</a></li>
+            <li><a href="/doc">English</a></li>
+          </ul>
+        </div>
+        <ul class="site-page-links">
+          <li><a href="/">Home</a></li>
+        </ul>
+      </div>
+      <div class="doc-module">
+        
+        <ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">FAQ</a></li>
+</ul>
+
+        
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>        
+      </div>
+    </div>
+  </header>
+  
+  <div class="main-content-wrap">
+
+    
+    <nav class="doc-menu-vertical" role="navigation">
+        
+          
+          <ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../getstarted/build_and_install/index_cn.html">安装与编译</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/docker_install_cn.html">PaddlePaddle的Docker容器使用方式</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/cmake/build_from_source_cn.html">PaddlePaddle的编译选项</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../getstarted/concepts/use_concepts_cn.html">基本使用概念</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cmd_parameter/index_cn.html">设置命令行参数</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/use_case_cn.html">使用案例</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/arguments_cn.html">参数概述</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/detail_introduction_cn.html">细节描述</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cluster/cluster_train_cn.html">运行分布式训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_basis_cn.html">Kubernetes 简介</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_cn.html">Kubernetes单机训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_distributed_cn.html">Kubernetes分布式训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/build_cn.html">编译PaddlePaddle和运行单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/write_docs_cn.html">如何贡献/修改文档</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/contribute_to_paddle_cn.html">如何贡献代码</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/deep_model/rnn/index_cn.html">RNN相关模型</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/rnn_config_cn.html">RNN配置</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/recurrent_group_cn.html">Recurrent Group教程</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hierarchical_layer_cn.html">支持双层序列作为输入的Layer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hrnn_rnn_api_compare_cn.html">单双层RNN API对比介绍</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/optimization/gpu_profiling_cn.html">GPU性能分析与调优</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/model_configs.html">模型配置</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/activation.html">Activation</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/layer.html">Layers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/evaluators.html">Evaluators</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/optimizer.html">Optimizer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/pooling.html">Pooling</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/networks.html">Networks</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/attr.html">Parameter Attribute</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/data.html">数据访问</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
+</ul>
+</li>
+<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">FAQ</a><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2 current"><a class="current reference internal" href="#">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
+</ul>
+
+        
+    </nav>
+    
+    <section class="doc-content-wrap">
+
+      
+
+ 
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+  <ul class="wy-breadcrumbs">
+      
+        <li><a href="../index_cn.html">FAQ</a> > </li>
+      
+    <li>本地训练与预测</li>
+  </ul>
+</div>
+      
+      <div class="wy-nav-content" id="doc-content">
+        <div class="rst-content">
+          <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
+           <div itemprop="articleBody">
+            
+  <div class="section" id="id1">
+<h1><a class="toc-backref" href="#id9">本地训练与预测</a><a class="headerlink" href="#id1" title="永久链接至标题">¶</a></h1>
+<div class="contents topic" id="contents">
+<p class="topic-title first">Contents</p>
+<ul class="simple">
+<li><a class="reference internal" href="#id1" id="id9">本地训练与预测</a><ul>
+<li><a class="reference internal" href="#id2" id="id10">1. 如何减少内存占用</a><ul>
+<li><a class="reference internal" href="#dataprovider" id="id11">减少DataProvider缓冲池内存</a></li>
+<li><a class="reference internal" href="#id3" id="id12">神经元激活内存</a></li>
+<li><a class="reference internal" href="#id4" id="id13">参数内存</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#id5" id="id14">2. 如何加速训练速度</a><ul>
+<li><a class="reference internal" href="#id6" id="id15">减少数据载入的耗时</a></li>
+<li><a class="reference internal" href="#id7" id="id16">加速训练速度</a></li>
+<li><a class="reference internal" href="#id8" id="id17">利用更多的计算资源</a></li>
+</ul>
+</li>
+<li><a class="reference internal" href="#gpu" id="id18">3. 如何指定GPU设备</a></li>
+<li><a class="reference internal" href="#floating-point-exception" id="id19">4. 训练过程中出现 <code class="code docutils literal"><span class="pre">Floating</span> <span class="pre">point</span> <span class="pre">exception</span></code>, 训练因此退出怎么办?</a></li>
+<li><a class="reference internal" href="#infer-layer" id="id20">5.  如何调用 infer 接口输出多个layer的预测结果</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<div class="section" id="id2">
+<h2><a class="toc-backref" href="#id10">1. 如何减少内存占用</a><a class="headerlink" href="#id2" title="永久链接至标题">¶</a></h2>
+<p>神经网络的训练本身是一个非常消耗内存和显存的工作，经常会消耗数10GB的内存和数GB的显存。
+PaddlePaddle的内存占用主要分为如下几个方面:</p>
+<ul class="simple">
+<li>DataProvider缓冲池内存（只针对内存）</li>
+<li>神经元激活内存（针对内存和显存）</li>
+<li>参数内存 （针对内存和显存）</li>
+<li>其他内存杂项</li>
+</ul>
+<p>其中，其他内存杂项是指PaddlePaddle本身所用的一些内存，包括字符串分配，临时变量等等，暂不考虑在内。</p>
+<div class="section" id="dataprovider">
+<h3><a class="toc-backref" href="#id11">减少DataProvider缓冲池内存</a><a class="headerlink" href="#dataprovider" title="永久链接至标题">¶</a></h3>
+<p>PyDataProvider使用的是异步加载，同时在内存里直接随即选取数据来做Shuffle。即</p>
+<img src="../../_images/graphviz-9be6aad37f57c60f4b971dde0ef44ce27179cf9a.png" alt="digraph {
+    rankdir=LR;
+    数据文件 -&gt; 内存池 -&gt; PaddlePaddle训练
+}" />
+<p>所以，减小这个内存池即可减小内存占用，同时也可以加速开始训练前数据载入的过程。但是，这
+个内存池实际上决定了shuffle的粒度。所以，如果将这个内存池减小，又要保证数据是随机的，
+那么最好将数据文件在每次读取之前做一次shuffle。可能的代码为</p>
+<div class="highlight-default"><div class="highlight"><pre><span></span><span class="nd">@provider</span><span class="p">(</span><span class="n">min_pool_size</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="o">...</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">settings</span><span class="p">,</span> <span class="n">filename</span><span class="p">):</span>
+    <span class="n">os</span><span class="o">.</span><span class="n">system</span><span class="p">(</span><span class="s1">&#39;shuf </span><span class="si">%s</span><span class="s1"> &gt; </span><span class="si">%s</span><span class="s1">.shuf&#39;</span> <span class="o">%</span> <span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">filename</span><span class="p">))</span>  <span class="c1"># shuffle before.</span>
+    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s1">&#39;</span><span class="si">%s</span><span class="s1">.shuf&#39;</span> <span class="o">%</span> <span class="n">filename</span><span class="p">,</span> <span class="s1">&#39;r&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">f</span><span class="p">:</span>
+            <span class="k">yield</span> <span class="n">get_sample_from_line</span><span class="p">(</span><span class="n">line</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>这样做可以极大的减少内存占用，并且可能会加速训练过程，详细文档参考 <a class="reference internal" href="../../api/v1/data_provider/pydataprovider2_cn.html#api-pydataprovider2"><span class="std std-ref">PyDataProvider2的使用</span></a> 。</p>
+</div>
+<div class="section" id="id3">
+<h3><a class="toc-backref" href="#id12">神经元激活内存</a><a class="headerlink" href="#id3" title="永久链接至标题">¶</a></h3>
+<p>神经网络在训练的时候，会对每一个激活暂存一些数据，如神经元激活值等。
+在反向传递的时候，这些数据会被用来更新参数。这些数据使用的内存主要和两个参数有关系，
+一是batch size，另一个是每条序列(Sequence)长度。所以，其实也是和每个mini-batch中包含
+的时间步信息成正比。</p>
+<p>所以做法可以有两种：</p>
+<ul class="simple">
+<li>减小batch size。 即在网络配置中 <code class="code docutils literal"><span class="pre">settings(batch_size=1000)</span></code> 设置成一个小一些的值。但是batch size本身是神经网络的超参数，减小batch size可能会对训练结果产生影响。</li>
+<li>减小序列的长度，或者直接扔掉非常长的序列。比如，一个数据集大部分序列长度是100-200,
+但是突然有一个10000长的序列，就很容易导致内存超限，特别是在LSTM等RNN中。</li>
+</ul>
+</div>
+<div class="section" id="id4">
+<h3><a class="toc-backref" href="#id13">参数内存</a><a class="headerlink" href="#id4" title="永久链接至标题">¶</a></h3>
+<p>PaddlePaddle支持非常多的优化算法(Optimizer)，不同的优化算法需要使用不同大小的内存。
+例如使用 <code class="code docutils literal"><span class="pre">adadelta</span></code> 算法，则需要使用等于权重参数规模大约5倍的内存。举例，如果参数保存下来的模型目录
+文件为 <code class="code docutils literal"><span class="pre">100M</span></code>， 那么该优化算法至少需要 <code class="code docutils literal"><span class="pre">500M</span></code> 的内存。</p>
+<p>可以考虑使用一些优化算法，例如 <code class="code docutils literal"><span class="pre">momentum</span></code>。</p>
+</div>
+</div>
+<div class="section" id="id5">
+<h2><a class="toc-backref" href="#id14">2. 如何加速训练速度</a><a class="headerlink" href="#id5" title="永久链接至标题">¶</a></h2>
+<p>加速PaddlePaddle训练可以考虑从以下几个方面：</p>
+<ul class="simple">
+<li>减少数据载入的耗时</li>
+<li>加速训练速度</li>
+<li>利用分布式训练驾驭更多的计算资源</li>
+</ul>
+<div class="section" id="id6">
+<h3><a class="toc-backref" href="#id15">减少数据载入的耗时</a><a class="headerlink" href="#id6" title="永久链接至标题">¶</a></h3>
+<p>使用<code class="code docutils literal"><span class="pre">pydataprovider</span></code>时，可以减少缓存池的大小，同时设置内存缓存功能，即可以极大的加速数据载入流程。
+<code class="code docutils literal"><span class="pre">DataProvider</span></code> 缓存池的减小，和之前减小通过减小缓存池来减小内存占用的原理一致。</p>
+<div class="highlight-default"><div class="highlight"><pre><span></span><span class="nd">@provider</span><span class="p">(</span><span class="n">min_pool_size</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="o">...</span><span class="p">)</span>
+<span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">settings</span><span class="p">,</span> <span class="n">filename</span><span class="p">):</span>
+    <span class="n">os</span><span class="o">.</span><span class="n">system</span><span class="p">(</span><span class="s1">&#39;shuf </span><span class="si">%s</span><span class="s1"> &gt; </span><span class="si">%s</span><span class="s1">.shuf&#39;</span> <span class="o">%</span> <span class="p">(</span><span class="n">filename</span><span class="p">,</span> <span class="n">filename</span><span class="p">))</span>  <span class="c1"># shuffle before.</span>
+    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s1">&#39;</span><span class="si">%s</span><span class="s1">.shuf&#39;</span> <span class="o">%</span> <span class="n">filename</span><span class="p">,</span> <span class="s1">&#39;r&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+        <span class="k">for</span> <span class="n">line</span> <span class="ow">in</span> <span class="n">f</span><span class="p">:</span>
+            <span class="k">yield</span> <span class="n">get_sample_from_line</span><span class="p">(</span><span class="n">line</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>同时 <code class="code docutils literal"><span class="pre">&#64;provider</span></code> 接口有一个 <code class="code docutils literal"><span class="pre">cache</span></code> 参数来控制缓存方法，将其设置成 <code class="code docutils literal"><span class="pre">CacheType.CACHE_PASS_IN_MEM</span></code> 的话，会将第一个 <code class="code docutils literal"><span class="pre">pass</span></code> (过完所有训练数据即为一个pass)生成的数据缓存在内存里，在之后的 <code class="code docutils literal"><span class="pre">pass</span></code> 中，不会再从 <code class="code docutils literal"><span class="pre">python</span></code> 端读取数据，而是直接从内存的缓存里读取数据。这也会极大减少数据读入的耗时。</p>
+</div>
+<div class="section" id="id7">
+<h3><a class="toc-backref" href="#id16">加速训练速度</a><a class="headerlink" href="#id7" title="永久链接至标题">¶</a></h3>
+<p>PaddlePaddle支持Sparse的训练，sparse训练需要训练特征是 <code class="code docutils literal"><span class="pre">sparse_binary_vector</span></code> 、 <code class="code docutils literal"><span class="pre">sparse_vector</span></code> 、或者 <code class="code docutils literal"><span class="pre">integer_value</span></code> 的任一一种。同时，与这个训练数据交互的Layer，需要将其Parameter设置成 sparse 更新模式，即设置 <code class="code docutils literal"><span class="pre">sparse_update=True</span></code></p>
+<p>这里使用简单的 <code class="code docutils literal"><span class="pre">word2vec</span></code> 训练语言模型距离，具体使用方法为:</p>
+<p>使用一个词前两个词和后两个词，来预测这个中间的词。这个任务的DataProvider为:</p>
+<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">DICT_DIM</span> <span class="o">=</span> <span class="mi">3000</span>
+
+
+<span class="nd">@provider</span><span class="p">(</span><span class="n">input_types</span><span class="o">=</span><span class="p">[</span><span class="n">integer_sequence</span><span class="p">(</span><span class="n">DICT_DIM</span><span class="p">),</span> <span class="n">integer_value</span><span class="p">(</span><span class="n">DICT_DIM</span><span class="p">)])</span>
+<span class="k">def</span> <span class="nf">process</span><span class="p">(</span><span class="n">settings</span><span class="p">,</span> <span class="n">filename</span><span class="p">):</span>
+    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">filename</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+        <span class="c1"># yield word ids to predict inner word id</span>
+        <span class="c1"># such as [28, 29, 10, 4], 4</span>
+        <span class="c1"># It means the sentance is  28, 29, 4, 10, 4.</span>
+        <span class="k">yield</span> <span class="n">read_next_from_file</span><span class="p">(</span><span class="n">f</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>这个任务的配置为:</p>
+<div class="highlight-default"><div class="highlight"><pre><span></span><span class="o">...</span>  <span class="c1"># the settings and define data provider is omitted.</span>
+<span class="n">DICT_DIM</span> <span class="o">=</span> <span class="mi">3000</span>  <span class="c1"># dictionary dimension.</span>
+<span class="n">word_ids</span> <span class="o">=</span> <span class="n">data_layer</span><span class="p">(</span><span class="s1">&#39;word_ids&#39;</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">DICT_DIM</span><span class="p">)</span>
+
+<span class="n">emb</span> <span class="o">=</span> <span class="n">embedding_layer</span><span class="p">(</span>
+    <span class="nb">input</span><span class="o">=</span><span class="n">word_ids</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span> <span class="n">param_attr</span><span class="o">=</span><span class="n">ParamAttr</span><span class="p">(</span><span class="n">sparse_update</span><span class="o">=</span><span class="kc">True</span><span class="p">))</span>
+<span class="n">emb_sum</span> <span class="o">=</span> <span class="n">pooling_layer</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">emb</span><span class="p">,</span> <span class="n">pooling_type</span><span class="o">=</span><span class="n">SumPooling</span><span class="p">())</span>
+<span class="n">predict</span> <span class="o">=</span> <span class="n">fc_layer</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">emb_sum</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">DICT_DIM</span><span class="p">,</span> <span class="n">act</span><span class="o">=</span><span class="n">Softmax</span><span class="p">())</span>
+<span class="n">outputs</span><span class="p">(</span>
+    <span class="n">classification_cost</span><span class="p">(</span>
+        <span class="nb">input</span><span class="o">=</span><span class="n">predict</span><span class="p">,</span> <span class="n">label</span><span class="o">=</span><span class="n">data_layer</span><span class="p">(</span>
+            <span class="s1">&#39;label&#39;</span><span class="p">,</span> <span class="n">size</span><span class="o">=</span><span class="n">DICT_DIM</span><span class="p">)))</span>
+</pre></div>
+</div>
+</div>
+<div class="section" id="id8">
+<h3><a class="toc-backref" href="#id17">利用更多的计算资源</a><a class="headerlink" href="#id8" title="永久链接至标题">¶</a></h3>
+<p>利用更多的计算资源可以分为一下几个方式来进行:</p>
+<ul class="simple">
+<li>单机CPU训练<ul>
+<li>使用多线程训练。设置命令行参数 <code class="code docutils literal"><span class="pre">trainer_count</span></code>。</li>
+</ul>
+</li>
+<li>单机GPU训练<ul>
+<li>使用显卡训练。设置命令行参数 <code class="code docutils literal"><span class="pre">use_gpu</span></code>。</li>
+<li>使用多块显卡训练。设置命令行参数 <code class="code docutils literal"><span class="pre">use_gpu</span></code> 和 <code class="code docutils literal"><span class="pre">trainer_count</span></code> 。</li>
+</ul>
+</li>
+<li>多机训练<ul>
+<li>请参考 <a class="reference internal" href="../../howto/usage/cluster/cluster_train_cn.html#cluster-train"><span class="std std-ref">运行分布式训练</span></a> 。</li>
+</ul>
+</li>
+</ul>
+</div>
+</div>
+<div class="section" id="gpu">
+<h2><a class="toc-backref" href="#id18">3. 如何指定GPU设备</a><a class="headerlink" href="#gpu" title="永久链接至标题">¶</a></h2>
+<p>例如机器上有4块GPU，编号从0开始，指定使用2、3号GPU：</p>
+<ul class="simple">
+<li>方式1：通过 <a class="reference external" href="http://www.acceleware.com/blog/cudavisibledevices-masking-gpus">CUDA_VISIBLE_DEVICES</a> 环境变量来指定特定的GPU。</li>
+</ul>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>env <span class="nv">CUDA_VISIBLE_DEVICES</span><span class="o">=</span><span class="m">2</span>,3 paddle train --use_gpu<span class="o">=</span><span class="nb">true</span> --trainer_count<span class="o">=</span><span class="m">2</span>
+</pre></div>
+</div>
+<ul class="simple">
+<li>方式2：通过命令行参数 <code class="docutils literal"><span class="pre">--gpu_id</span></code> 指定。</li>
+</ul>
+<div class="highlight-bash"><div class="highlight"><pre><span></span>paddle train --use_gpu<span class="o">=</span><span class="nb">true</span> --trainer_count<span class="o">=</span><span class="m">2</span> --gpu_id<span class="o">=</span><span class="m">2</span>
+</pre></div>
+</div>
+</div>
+<div class="section" id="floating-point-exception">
+<h2><a class="toc-backref" href="#id19">4. 训练过程中出现 <code class="code docutils literal"><span class="pre">Floating</span> <span class="pre">point</span> <span class="pre">exception</span></code>, 训练因此退出怎么办?</a><a class="headerlink" href="#floating-point-exception" title="永久链接至标题">¶</a></h2>
+<p>Paddle二进制在运行时捕获了浮点数异常，只要出现浮点数异常(即训练过程中出现NaN或者Inf)，立刻退出。浮点异常通常的原因是浮点数溢出、除零等问题。
+主要原因包括两个方面:</p>
+<ul class="simple">
+<li>训练过程中参数或者训练过程中的梯度尺度过大，导致参数累加，乘除等时候，导致了浮点数溢出。</li>
+<li>模型一直不收敛，发散到了一个数值特别大的地方。</li>
+<li>训练数据有问题，导致参数收敛到了一些奇异的情况。或者输入数据尺度过大，有些特征的取值达到数百万，这时进行矩阵乘法运算就可能导致浮点数溢出。</li>
+</ul>
+<p>这里有两种有效的解决方法：</p>
+<ol class="arabic simple">
+<li>设置 <code class="code docutils literal"><span class="pre">gradient_clipping_threshold</span></code> 参数，示例代码如下：</li>
+</ol>
+<div class="highlight-python"><div class="highlight"><pre><span></span>
+</pre></div>
+</div>
+<dl class="docutils">
+<dt>optimizer = paddle.optimizer.RMSProp(</dt>
+<dd>learning_rate=1e-3,
+gradient_clipping_threshold=10.0,
+regularization=paddle.optimizer.L2Regularization(rate=8e-4))</dd>
+</dl>
+<p>具体可以参考  <a class="reference external" href="https://github.com/PaddlePaddle/models/blob/develop/nmt_without_attention/train.py#L35">nmt_without_attention</a> 示例。</p>
+<ol class="arabic simple" start="2">
+<li>设置 <code class="code docutils literal"><span class="pre">error_clipping_threshold</span></code> 参数，示例代码如下：</li>
+</ol>
+<div class="highlight-python"><div class="highlight"><pre><span></span>
+</pre></div>
+</div>
+<dl class="docutils">
+<dt>decoder_inputs = paddle.layer.fc(</dt>
+<dd><p class="first">act=paddle.activation.Linear(),
+size=decoder_size * 3,
+bias_attr=False,
+input=[context, current_word],
+layer_attr=paddle.attr.ExtraLayerAttribute(</p>
+<blockquote class="last">
+<div>error_clipping_threshold=100.0))</div></blockquote>
+</dd>
+</dl>
+<p>完整代码可以参考示例 <a class="reference external" href="https://github.com/PaddlePaddle/book/blob/develop/08.machine_translation/train.py#L66">machine translation</a> 。</p>
+<p>两种方法的区别：</p>
+<ol class="arabic simple">
+<li>两者都是对梯度的截断，但截断时机不同，前者在 <code class="code docutils literal"><span class="pre">optimzier</span></code> 更新网络参数时应用；后者在激活函数反向计算时被调用；</li>
+<li>截断对象不同：前者截断可学习参数的梯度，后者截断回传给前层的梯度;</li>
+</ol>
+<p>除此之外，还可以通过减小学习律或者对数据进行归一化处理来解决这类问题。</p>
+</div>
+<div class="section" id="infer-layer">
+<h2><a class="toc-backref" href="#id20">5.  如何调用 infer 接口输出多个layer的预测结果</a><a class="headerlink" href="#infer-layer" title="永久链接至标题">¶</a></h2>
+<ul class="simple">
+<li>将需要输出的层作为 <code class="code docutils literal"><span class="pre">paddle.inference.Inference()</span></code> 接口的 <code class="code docutils literal"><span class="pre">output_layer</span></code> 参数输入，代码如下：</li>
+</ul>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">inferer</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">inference</span><span class="o">.</span><span class="n">Inference</span><span class="p">(</span><span class="n">output_layer</span><span class="o">=</span><span class="p">[</span><span class="n">layer1</span><span class="p">,</span> <span class="n">layer2</span><span class="p">],</span> <span class="n">parameters</span><span class="o">=</span><span class="n">parameters</span><span class="p">)</span>
+</pre></div>
+</div>
+<ul class="simple">
+<li>指定要输出的字段进行输出。以输出 <code class="code docutils literal"><span class="pre">value</span></code> 字段为例，代码如下：</li>
+</ul>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">out</span> <span class="o">=</span> <span class="n">inferer</span><span class="o">.</span><span class="n">infer</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">data_batch</span><span class="p">,</span> <span class="n">field</span><span class="o">=</span><span class="p">[</span><span class="s2">&quot;value&quot;</span><span class="p">])</span>
+</pre></div>
+</div>
+<p>需要注意的是：</p>
+<ul class="simple">
+<li>如果指定了2个layer作为输出层，实际上需要的输出结果是两个矩阵；</li>
+<li>假设第一个layer的输出A是一个 N1 * M1 的矩阵，第二个 Layer 的输出B是一个 N2 * M2 的矩阵；</li>
+<li>paddle.v2 默认会将A和B 横向拼接，当N1 和 N2 大小不一样时，会报如下的错误：</li>
+</ul>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="ne">ValueError</span><span class="p">:</span> <span class="nb">all</span> <span class="n">the</span> <span class="nb">input</span> <span class="n">array</span> <span class="n">dimensions</span> <span class="k">except</span> <span class="k">for</span> <span class="n">the</span> <span class="n">concatenation</span> <span class="n">axis</span> <span class="n">must</span> <span class="n">match</span> <span class="n">exactly</span>
+</pre></div>
+</div>
+<p>多个层的输出矩阵的高度不一致导致拼接失败，这种情况常常发生在：</p>
+<ul class="simple">
+<li>同时输出序列层和非序列层；</li>
+<li>多个输出层处理多个不同长度的序列;</li>
+</ul>
+<p>此时可以在调用infer接口时通过设置 <code class="code docutils literal"><span class="pre">flatten_result=False</span></code> , 跳过“拼接”步骤，来解决上面的问题。这时，infer接口的返回值是一个python list:</p>
+<ul class="simple">
+<li>list 中元素的个数等于网络中输出层的个数；</li>
+<li>list 中每个元素是一个layer的输出结果矩阵，类型是numpy的ndarray；</li>
+<li>每一个layer输出矩阵的高度，在非序列输入时：等于样本数；序列输入时等于：输入序列中元素的总数；宽度等于配置中layer的size；</li>
+</ul>
+</div>
+</div>
+
+
+           </div>
+          </div>
+          <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../cluster/index_cn.html" class="btn btn-neutral float-right" title="集群训练与预测" accesskey="n">Next <span class="fa fa-arrow-circle-right"></span></a>
+      
+      
+        <a href="../parameter/index_cn.html" class="btn btn-neutral" title="参数设置" accesskey="p"><span class="fa fa-arrow-circle-left"></span> Previous</a>
+      
+    </div>
+  
+
+  <hr/>
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2016, PaddlePaddle developers.
+
+    </p>
+  </div>
+  Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>. 
+
+</footer>
+
+        </div>
+      </div>
+
+    </section>
+
+  </div>
+  
+
+
+  
+
+    <script type="text/javascript">
+        var DOCUMENTATION_OPTIONS = {
+            URL_ROOT:'../../',
+            VERSION:'',
+            COLLAPSE_INDEX:false,
+            FILE_SUFFIX:'.html',
+            HAS_SOURCE:  true,
+            SOURCELINK_SUFFIX: ".txt",
+        };
+    </script>
+      <script type="text/javascript" src="../../_static/jquery.js"></script>
+      <script type="text/javascript" src="../../_static/underscore.js"></script>
+      <script type="text/javascript" src="../../_static/doctools.js"></script>
+      <script type="text/javascript" src="../../_static/translations.js"></script>
+      <script type="text/javascript" src="https://cdn.bootcss.com/mathjax/2.7.0/MathJax.js"></script>
+       
+  
+
+  
+  
+    <script type="text/javascript" src="../../_static/js/theme.js"></script>
+  
+  
+  <script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/js/bootstrap.min.js" integrity="sha384-Tc5IQib027qvyjSMfHjOMaLkfuWVxZxUPnCJA7l2mCWNIpG9mGCD8wGNIcPD7Txa" crossorigin="anonymous"></script>
+  <script src="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/js/perfect-scrollbar.jquery.min.js"></script>
+  <script src="../../_static/js/paddle_doc_init.js"></script> 
+
+</body>
+</html>
\ No newline at end of file
diff --git a/develop/doc_cn/faq/model/index_cn.html b/develop/doc_cn/faq/model/index_cn.html
new file mode 100644
index 0000000000000000000000000000000000000000..d8b0fda461bd2f8e55e454ee399fd772fc0bbabc
--- /dev/null
+++ b/develop/doc_cn/faq/model/index_cn.html
@@ -0,0 +1,355 @@
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>模型配置 &mdash; PaddlePaddle  文档</title>
+  
+
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  
+  
+    <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  
+
+  
+  
+        <link rel="index" title="索引"
+              href="../../genindex.html"/>
+        <link rel="search" title="搜索" href="../../search.html"/>
+    <link rel="top" title="PaddlePaddle  文档" href="../../index.html"/>
+        <link rel="up" title="FAQ" href="../index_cn.html"/>
+        <link rel="next" title="参数设置" href="../parameter/index_cn.html"/>
+        <link rel="prev" title="编译安装与单元测试" href="../build_and_install/index_cn.html"/> 
+
+  <link rel="stylesheet" href="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/css/perfect-scrollbar.min.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/override.css" type="text/css" />
+  <script>
+  var _hmt = _hmt || [];
+  (function() {
+    var hm = document.createElement("script");
+    hm.src = "//hm.baidu.com/hm.js?b9a314ab40d04d805655aab1deee08ba";
+    var s = document.getElementsByTagName("script")[0]; 
+    s.parentNode.insertBefore(hm, s);
+  })();
+  </script>
+
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+</head>
+
+<body class="wy-body-for-nav" role="document">
+
+  
+  <header class="site-header">
+    <div class="site-logo">
+      <a href="/"><img src="../../_static/images/PP_w.png"></a>
+    </div>
+    <div class="site-nav-links">
+      <div class="site-menu">
+        <a class="fork-on-github" href="https://github.com/PaddlePaddle/Paddle" target="_blank"><i class="fa fa-github"></i>Fork me on Github</a>
+        <div class="language-switcher dropdown">
+          <a type="button" data-toggle="dropdown">
+            <span>English</span>
+            <i class="fa fa-angle-up"></i>
+            <i class="fa fa-angle-down"></i>
+          </a>
+          <ul class="dropdown-menu">
+            <li><a href="/doc_cn">中文</a></li>
+            <li><a href="/doc">English</a></li>
+          </ul>
+        </div>
+        <ul class="site-page-links">
+          <li><a href="/">Home</a></li>
+        </ul>
+      </div>
+      <div class="doc-module">
+        
+        <ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">FAQ</a></li>
+</ul>
+
+        
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>        
+      </div>
+    </div>
+  </header>
+  
+  <div class="main-content-wrap">
+
+    
+    <nav class="doc-menu-vertical" role="navigation">
+        
+          
+          <ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../getstarted/build_and_install/index_cn.html">安装与编译</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/docker_install_cn.html">PaddlePaddle的Docker容器使用方式</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/cmake/build_from_source_cn.html">PaddlePaddle的编译选项</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../getstarted/concepts/use_concepts_cn.html">基本使用概念</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cmd_parameter/index_cn.html">设置命令行参数</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/use_case_cn.html">使用案例</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/arguments_cn.html">参数概述</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/detail_introduction_cn.html">细节描述</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cluster/cluster_train_cn.html">运行分布式训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_basis_cn.html">Kubernetes 简介</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_cn.html">Kubernetes单机训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_distributed_cn.html">Kubernetes分布式训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/build_cn.html">编译PaddlePaddle和运行单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/write_docs_cn.html">如何贡献/修改文档</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/contribute_to_paddle_cn.html">如何贡献代码</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/deep_model/rnn/index_cn.html">RNN相关模型</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/rnn_config_cn.html">RNN配置</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/recurrent_group_cn.html">Recurrent Group教程</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hierarchical_layer_cn.html">支持双层序列作为输入的Layer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hrnn_rnn_api_compare_cn.html">单双层RNN API对比介绍</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/optimization/gpu_profiling_cn.html">GPU性能分析与调优</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/model_configs.html">模型配置</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/activation.html">Activation</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/layer.html">Layers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/evaluators.html">Evaluators</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/optimizer.html">Optimizer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/pooling.html">Pooling</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/networks.html">Networks</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/attr.html">Parameter Attribute</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/data.html">数据访问</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
+</ul>
+</li>
+<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">FAQ</a><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2 current"><a class="current reference internal" href="#">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
+</ul>
+
+        
+    </nav>
+    
+    <section class="doc-content-wrap">
+
+      
+
+ 
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+  <ul class="wy-breadcrumbs">
+      
+        <li><a href="../index_cn.html">FAQ</a> > </li>
+      
+    <li>模型配置</li>
+  </ul>
+</div>
+      
+      <div class="wy-nav-content" id="doc-content">
+        <div class="rst-content">
+          <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
+           <div itemprop="articleBody">
+            
+  <div class="section" id="id1">
+<h1><a class="toc-backref" href="#id2">模型配置</a><a class="headerlink" href="#id1" title="永久链接至标题">¶</a></h1>
+<div class="contents topic" id="contents">
+<p class="topic-title first">Contents</p>
+<ul class="simple">
+<li><a class="reference internal" href="#id1" id="id2">模型配置</a><ul>
+<li><a class="reference internal" href="#duplicated-layer-name" id="id3">1. 出现 <code class="code docutils literal"><span class="pre">Duplicated</span> <span class="pre">layer</span> <span class="pre">name</span></code> 错误怎么办</a></li>
+<li><a class="reference internal" href="#paddle-layer-memory-name" id="id4">2. <code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 的参数 <code class="code docutils literal"><span class="pre">name</span></code> 如何使用</a></li>
+<li><a class="reference internal" href="#drop-out" id="id5">3. 两种使用 drop_out 的方法有何区别</a></li>
+<li><a class="reference internal" href="#recurrent-layer" id="id6">4. 不同的 recurrent layer 的区别</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<div class="section" id="duplicated-layer-name">
+<h2><a class="toc-backref" href="#id3">1. 出现 <code class="code docutils literal"><span class="pre">Duplicated</span> <span class="pre">layer</span> <span class="pre">name</span></code> 错误怎么办</a><a class="headerlink" href="#duplicated-layer-name" title="永久链接至标题">¶</a></h2>
+<p>出现该错误的原因一般是用户对不同layer的参数 <code class="code docutils literal"><span class="pre">name</span></code> 设置了相同的取值。遇到该错误时，先找出参数 <code class="code docutils literal"><span class="pre">name</span></code> 取值相同的layer，然后将这些layer的参数 <code class="code docutils literal"><span class="pre">name</span></code> 设置为不同的值。</p>
+</div>
+<div class="section" id="paddle-layer-memory-name">
+<h2><a class="toc-backref" href="#id4">2. <code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 的参数 <code class="code docutils literal"><span class="pre">name</span></code> 如何使用</a><a class="headerlink" href="#paddle-layer-memory-name" title="永久链接至标题">¶</a></h2>
+<ul class="simple">
+<li><code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 用于获取特定layer上一时间步的输出，该layer是通过参数 <code class="code docutils literal"><span class="pre">name</span></code> 指定，即，<code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 会关联参数 <code class="code docutils literal"><span class="pre">name</span></code> 取值相同的layer，并将该layer上一时间步的输出作为自身当前时间步的输出。</li>
+<li>PaddlePaddle的所有layer都有唯一的name，用户通过参数 <code class="code docutils literal"><span class="pre">name</span></code> 设定，当用户没有显式设定时，PaddlePaddle会自动设定。而 <code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 不是真正的layer，其name由参数 <code class="code docutils literal"><span class="pre">memory_name</span></code> 设定，当用户没有显式设定时，PaddlePaddle会自动设定。<code class="code docutils literal"><span class="pre">paddle.layer.memory</span></code> 的参数 <code class="code docutils literal"><span class="pre">name</span></code> 用于指定其要关联的layer，需要用户显式设定。</li>
+</ul>
+</div>
+<div class="section" id="drop-out">
+<h2><a class="toc-backref" href="#id5">3. 两种使用 drop_out 的方法有何区别</a><a class="headerlink" href="#drop-out" title="永久链接至标题">¶</a></h2>
+<ul>
+<li><p class="first">在PaddlePaddle中使用dropout有两种方式</p>
+<ul class="simple">
+<li>在相应layer的 <code class="code docutils literal"><span class="pre">layer_atter</span></code> 设置 <code class="code docutils literal"><span class="pre">drop_rate</span></code>，以 <code class="code docutils literal"><span class="pre">paddle.layer.fc</span></code> 为例，代码如下：</li>
+</ul>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">fc</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">layer</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">,</span> <span class="n">layer_attr</span><span class="o">=</span><span class="n">paddle</span><span class="o">.</span><span class="n">attr</span><span class="o">.</span><span class="n">ExtraLayerAttribute</span><span class="p">(</span><span class="n">drop_rate</span><span class="o">=</span><span class="mf">0.5</span><span class="p">))</span>
+</pre></div>
+</div>
+<ul class="simple">
+<li>使用 <code class="code docutils literal"><span class="pre">paddle.layer.dropout</span></code>，以 <code class="code docutils literal"><span class="pre">paddle.layer.fc</span></code> 为例，代码如下：</li>
+</ul>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">fc</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">layer</span><span class="o">.</span><span class="n">fc</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="nb">input</span><span class="p">)</span>
+<span class="n">drop_fc</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">layer</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">fc</span><span class="p">,</span> <span class="n">dropout_rate</span><span class="o">=</span><span class="mf">0.5</span><span class="p">)</span>
+</pre></div>
+</div>
+</li>
+<li><p class="first"><code class="code docutils literal"><span class="pre">paddle.layer.dropout</span></code> 实际上使用了 <code class="code docutils literal"><span class="pre">paddle.layer.add_to</span></code>，并在该layer里采用第一种方式设置 <code class="code docutils literal"><span class="pre">drop_rate</span></code> 来使用dropout的。这种方式对内存消耗较大。</p>
+</li>
+<li><p class="first">PaddlePaddle在激活函数里实现dropout，而不是在layer里实现。</p>
+</li>
+<li><p class="first"><code class="code docutils literal"><span class="pre">paddle.layer.lstmemory</span></code>、<code class="code docutils literal"><span class="pre">paddle.layer.grumemory</span></code>、<code class="code docutils literal"><span class="pre">paddle.layer.recurrent</span></code> 不是通过一般的方式来实现对输出的激活，所以不能采用第一种方式在这几个layer里设置 <code class="code docutils literal"><span class="pre">drop_rate</span></code> 来使用dropout。若要对这几个layer使用dropout，可采用第二种方式，即使用 <code class="code docutils literal"><span class="pre">paddle.layer.dropout</span></code>。</p>
+</li>
+</ul>
+</div>
+<div class="section" id="recurrent-layer">
+<h2><a class="toc-backref" href="#id6">4. 不同的 recurrent layer 的区别</a><a class="headerlink" href="#recurrent-layer" title="永久链接至标题">¶</a></h2>
+<p>以LSTM为例，在PaddlePaddle中包含以下 recurrent layer：</p>
+<ul class="simple">
+<li><code class="code docutils literal"><span class="pre">paddle.layer.lstmemory</span></code></li>
+<li><code class="code docutils literal"><span class="pre">paddle.networks.simple_lstm</span></code></li>
+<li><code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_group</span></code></li>
+<li><code class="code docutils literal"><span class="pre">paddle.networks.bidirectional_lstm</span></code></li>
+</ul>
+<p>按照具体实现方式可以归纳为2类：</p>
+<ol class="arabic simple">
+<li>由 recurrent_group 实现的 recurrent layer：</li>
+</ol>
+<blockquote>
+<div><ul class="simple">
+<li>用户在使用这一类recurrent layer时，可以访问由recurrent unit在一个时间步内计算得到的中间值（例如：hidden states, memory cells等）；</li>
+<li>上述的 <code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_group</span></code> 是这一类的 recurrent layer ；</li>
+</ul>
+</div></blockquote>
+<ol class="arabic simple" start="2">
+<li>将recurrent layer作为一个整体来实现：</li>
+</ol>
+<blockquote>
+<div><ul class="simple">
+<li>用户在使用这一类recurrent layer，只能访问它们的输出值；</li>
+<li>上述的 <code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_group</span></code> 、 <code class="code docutils literal"><span class="pre">paddle.networks.simple_lstm</span></code> 和 <code class="code docutils literal"><span class="pre">paddle.networks.bidirectional_lstm</span></code> 属于这一类的实现；</li>
+</ul>
+</div></blockquote>
+<p>将recurrent layer作为一个整体来实现， 能够针对CPU和GPU的计算做更多优化， 所以相比于recurrent group的实现方式， 第二类 recurrent layer 计算效率更高。 在实际应用中，如果用户不需要访问LSTM的中间变量，而只需要获得recurrent layer计算的输出，我们建议使用第二类实现。</p>
+<p>此外，关于LSTM, PaddlePaddle中还包含 <code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_unit</span></code> 这一计算单元：</p>
+<blockquote>
+<div><ul class="simple">
+<li>不同于上述介绍的recurrent layer , <code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_unit</span></code> 定义了LSTM单元在一个时间步内的计算过程，它并不是一个完整的recurrent layer，也不能接收序列数据作为输入；</li>
+<li><code class="code docutils literal"><span class="pre">paddle.networks.lstmemory_unit</span></code> 只能在recurrent_group中作为step function使用；</li>
+</ul>
+</div></blockquote>
+</div>
+</div>
+
+
+           </div>
+          </div>
+          <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../parameter/index_cn.html" class="btn btn-neutral float-right" title="参数设置" accesskey="n">Next <span class="fa fa-arrow-circle-right"></span></a>
+      
+      
+        <a href="../build_and_install/index_cn.html" class="btn btn-neutral" title="编译安装与单元测试" accesskey="p"><span class="fa fa-arrow-circle-left"></span> Previous</a>
+      
+    </div>
+  
+
+  <hr/>
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2016, PaddlePaddle developers.
+
+    </p>
+  </div>
+  Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>. 
+
+</footer>
+
+        </div>
+      </div>
+
+    </section>
+
+  </div>
+  
+
+
+  
+
+    <script type="text/javascript">
+        var DOCUMENTATION_OPTIONS = {
+            URL_ROOT:'../../',
+            VERSION:'',
+            COLLAPSE_INDEX:false,
+            FILE_SUFFIX:'.html',
+            HAS_SOURCE:  true,
+            SOURCELINK_SUFFIX: ".txt",
+        };
+    </script>
+      <script type="text/javascript" src="../../_static/jquery.js"></script>
+      <script type="text/javascript" src="../../_static/underscore.js"></script>
+      <script type="text/javascript" src="../../_static/doctools.js"></script>
+      <script type="text/javascript" src="../../_static/translations.js"></script>
+      <script type="text/javascript" src="https://cdn.bootcss.com/mathjax/2.7.0/MathJax.js"></script>
+       
+  
+
+  
+  
+    <script type="text/javascript" src="../../_static/js/theme.js"></script>
+  
+  
+  <script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/js/bootstrap.min.js" integrity="sha384-Tc5IQib027qvyjSMfHjOMaLkfuWVxZxUPnCJA7l2mCWNIpG9mGCD8wGNIcPD7Txa" crossorigin="anonymous"></script>
+  <script src="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/js/perfect-scrollbar.jquery.min.js"></script>
+  <script src="../../_static/js/paddle_doc_init.js"></script> 
+
+</body>
+</html>
\ No newline at end of file
diff --git a/develop/doc_cn/faq/parameter/index_cn.html b/develop/doc_cn/faq/parameter/index_cn.html
new file mode 100644
index 0000000000000000000000000000000000000000..c259aa4d4eae8ab8527c395d4b0a19ae95e38361
--- /dev/null
+++ b/develop/doc_cn/faq/parameter/index_cn.html
@@ -0,0 +1,439 @@
+
+
+<!DOCTYPE html>
+<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
+<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
+<head>
+  <meta charset="utf-8">
+  
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  
+  <title>参数设置 &mdash; PaddlePaddle  文档</title>
+  
+
+  
+  
+
+  
+
+  
+  
+    
+
+  
+
+  
+  
+    <link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
+  
+
+  
+  
+        <link rel="index" title="索引"
+              href="../../genindex.html"/>
+        <link rel="search" title="搜索" href="../../search.html"/>
+    <link rel="top" title="PaddlePaddle  文档" href="../../index.html"/>
+        <link rel="up" title="FAQ" href="../index_cn.html"/>
+        <link rel="next" title="本地训练与预测" href="../local/index_cn.html"/>
+        <link rel="prev" title="模型配置" href="../model/index_cn.html"/> 
+
+  <link rel="stylesheet" href="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/css/perfect-scrollbar.min.css" type="text/css" />
+  <link rel="stylesheet" href="../../_static/css/override.css" type="text/css" />
+  <script>
+  var _hmt = _hmt || [];
+  (function() {
+    var hm = document.createElement("script");
+    hm.src = "//hm.baidu.com/hm.js?b9a314ab40d04d805655aab1deee08ba";
+    var s = document.getElementsByTagName("script")[0]; 
+    s.parentNode.insertBefore(hm, s);
+  })();
+  </script>
+
+  
+
+  
+  <script src="../../_static/js/modernizr.min.js"></script>
+
+</head>
+
+<body class="wy-body-for-nav" role="document">
+
+  
+  <header class="site-header">
+    <div class="site-logo">
+      <a href="/"><img src="../../_static/images/PP_w.png"></a>
+    </div>
+    <div class="site-nav-links">
+      <div class="site-menu">
+        <a class="fork-on-github" href="https://github.com/PaddlePaddle/Paddle" target="_blank"><i class="fa fa-github"></i>Fork me on Github</a>
+        <div class="language-switcher dropdown">
+          <a type="button" data-toggle="dropdown">
+            <span>English</span>
+            <i class="fa fa-angle-up"></i>
+            <i class="fa fa-angle-down"></i>
+          </a>
+          <ul class="dropdown-menu">
+            <li><a href="/doc_cn">中文</a></li>
+            <li><a href="/doc">English</a></li>
+          </ul>
+        </div>
+        <ul class="site-page-links">
+          <li><a href="/">Home</a></li>
+        </ul>
+      </div>
+      <div class="doc-module">
+        
+        <ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a></li>
+<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">FAQ</a></li>
+</ul>
+
+        
+<div role="search">
+  <form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
+    <input type="text" name="q" placeholder="Search docs" />
+    <input type="hidden" name="check_keywords" value="yes" />
+    <input type="hidden" name="area" value="default" />
+  </form>
+</div>        
+      </div>
+    </div>
+  </header>
+  
+  <div class="main-content-wrap">
+
+    
+    <nav class="doc-menu-vertical" role="navigation">
+        
+          
+          <ul class="current">
+<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../getstarted/build_and_install/index_cn.html">安装与编译</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/docker_install_cn.html">PaddlePaddle的Docker容器使用方式</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/cmake/build_from_source_cn.html">PaddlePaddle的编译选项</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../getstarted/concepts/use_concepts_cn.html">基本使用概念</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="../../howto/index_cn.html">进阶指南</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cmd_parameter/index_cn.html">设置命令行参数</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/use_case_cn.html">使用案例</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/arguments_cn.html">参数概述</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/usage/cmd_parameter/detail_introduction_cn.html">细节描述</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/cluster/cluster_train_cn.html">运行分布式训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_basis_cn.html">Kubernetes 简介</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_cn.html">Kubernetes单机训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/usage/k8s/k8s_distributed_cn.html">Kubernetes分布式训练</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/build_cn.html">编译PaddlePaddle和运行单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/write_docs_cn.html">如何贡献/修改文档</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/dev/contribute_to_paddle_cn.html">如何贡献代码</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/deep_model/rnn/index_cn.html">RNN相关模型</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/rnn_config_cn.html">RNN配置</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/recurrent_group_cn.html">Recurrent Group教程</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hierarchical_layer_cn.html">支持双层序列作为输入的Layer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../howto/deep_model/rnn/hrnn_rnn_api_compare_cn.html">单双层RNN API对比介绍</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../howto/optimization/gpu_profiling_cn.html">GPU性能分析与调优</a></li>
+</ul>
+</li>
+<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/model_configs.html">模型配置</a><ul>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/activation.html">Activation</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/layer.html">Layers</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/evaluators.html">Evaluators</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/optimizer.html">Optimizer</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/pooling.html">Pooling</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/networks.html">Networks</a></li>
+<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/attr.html">Parameter Attribute</a></li>
+</ul>
+</li>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/data.html">数据访问</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
+</ul>
+</li>
+<li class="toctree-l1 current"><a class="reference internal" href="../index_cn.html">FAQ</a><ul class="current">
+<li class="toctree-l2"><a class="reference internal" href="../build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2 current"><a class="current reference internal" href="#">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
+</ul>
+
+        
+    </nav>
+    
+    <section class="doc-content-wrap">
+
+      
+
+ 
+
+
+
+
+
+
+
+<div role="navigation" aria-label="breadcrumbs navigation">
+  <ul class="wy-breadcrumbs">
+      
+        <li><a href="../index_cn.html">FAQ</a> > </li>
+      
+    <li>参数设置</li>
+  </ul>
+</div>
+      
+      <div class="wy-nav-content" id="doc-content">
+        <div class="rst-content">
+          <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
+           <div itemprop="articleBody">
+            
+  <div class="section" id="id1">
+<h1><a class="toc-backref" href="#id6">参数设置</a><a class="headerlink" href="#id1" title="永久链接至标题">¶</a></h1>
+<div class="contents topic" id="contents">
+<p class="topic-title first">Contents</p>
+<ul class="simple">
+<li><a class="reference internal" href="#id1" id="id6">参数设置</a><ul>
+<li><a class="reference internal" href="#sgd" id="id7">1. 如何选择SGD算法的学习率</a></li>
+<li><a class="reference internal" href="#learning-rate-annealing" id="id8">2. 如何设置学习率退火（learning rate annealing）</a></li>
+<li><a class="reference internal" href="#id2" id="id9">3. 如何初始化参数</a></li>
+<li><a class="reference internal" href="#id3" id="id10">4. 如何共享参数</a></li>
+<li><a class="reference internal" href="#id4" id="id11">5. 如何加载预训练参数</a></li>
+<li><a class="reference internal" href="#id5" id="id12">6. 存储的参数格式是什么，如何和明文进行相互转化</a></li>
+<li><a class="reference internal" href="#a-protocol-message-was-rejected-because-it-was-too-big" id="id13">7. A protocol message was rejected because it was too big</a></li>
+</ul>
+</li>
+</ul>
+</div>
+<div class="section" id="sgd">
+<h2><a class="toc-backref" href="#id7">1. 如何选择SGD算法的学习率</a><a class="headerlink" href="#sgd" title="永久链接至标题">¶</a></h2>
+<p>在采用sgd/async_sgd进行训练时，一个重要的问题是选择正确的learning_rate。如果learning_rate太大，那么训练有可能不收敛，如果learning_rate太小，那么收敛可能很慢，导致训练时间过长。</p>
+<p>通常做法是从一个比较大的learning_rate开始试，如果不收敛，那减少学习率10倍继续试验，直到训练收敛为止。那么如何判断训练不收敛呢？可以估计出如果模型采用不变的输出最小的cost0是多少。</p>
+<p>如果训练过程的的cost明显高于这个常数输出的cost，那么我们可以判断为训练不收敛。举一个例子，假如我们是三分类问题，采用multi-class-cross-entropy作为cost，数据中0,1,2三类的比例为 <code class="code docutils literal"><span class="pre">0.2,</span> <span class="pre">0.5,</span> <span class="pre">0.3</span></code> , 那么常数输出所能达到的最小cost是 <code class="code docutils literal"><span class="pre">-(0.2*log(0.2)+0.5*log(0.5)+0.3*log(0.3))=1.03</span></code> 。如果训练一个pass（或者更早）后，cost还大于这个数，那么可以认为训练不收敛，应该降低学习率。</p>
+</div>
+<div class="section" id="learning-rate-annealing">
+<h2><a class="toc-backref" href="#id8">2. 如何设置学习率退火（learning rate annealing）</a><a class="headerlink" href="#learning-rate-annealing" title="永久链接至标题">¶</a></h2>
+<p>在相应的优化算法里设置learning_rate_schedule及相关参数，以使用Adam算法为例，代码如下：</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span>
+    <span class="n">learning_rate</span><span class="o">=</span><span class="mf">1e-3</span><span class="p">,</span>
+    <span class="n">learning_rate_decay_a</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span>
+    <span class="n">learning_rate_decay_b</span><span class="o">=</span><span class="mf">0.75</span><span class="p">,</span>
+    <span class="n">learning_rate_schedule</span><span class="o">=</span><span class="s2">&quot;poly&quot;</span><span class="p">,)</span>
+</pre></div>
+</div>
+<p>PaddlePaddle目前支持8种learning_rate_schedule，这8种learning_rate_schedule及其对应学习率计算方式如下：</p>
+<ul>
+<li><p class="first">&#8220;constant&#8221;</p>
+<p>lr = learning_rate</p>
+</li>
+<li><p class="first">&#8220;poly&#8221;</p>
+<p>lr = learning_rate * pow(1 + learning_rate_decay_a * num_samples_processed, -learning_rate_decay_b)</p>
+<p>其中，num_samples_processed为已训练样本数，下同。</p>
+</li>
+<li><p class="first">&#8220;caffe_poly&#8221;</p>
+<p>lr = learning_rate * pow(1.0 - num_samples_processed / learning_rate_decay_a, learning_rate_decay_b)</p>
+</li>
+<li><p class="first">&#8220;exp&#8221;</p>
+<p>lr = learning_rate * pow(learning_rate_decay_a, num_samples_processed / learning_rate_decay_b)</p>
+</li>
+<li><p class="first">&#8220;discexp&#8221;</p>
+<p>lr = learning_rate * pow(learning_rate_decay_a, floor(num_samples_processed / learning_rate_decay_b))</p>
+</li>
+<li><p class="first">&#8220;linear&#8221;</p>
+<p>lr = max(learning_rate - learning_rate_decay_a * num_samples_processed, learning_rate_decay_b)</p>
+</li>
+<li><p class="first">&#8220;manual&#8221;</p>
+<p>这是一种按已训练样本数分段取值的学习率退火方法。使用该learning_rate_schedule时，用户通过参数 <code class="code docutils literal"><span class="pre">learning_rate_args</span></code> 设置学习率衰减因子分段函数，当前的学习率为所设置 <code class="code docutils literal"><span class="pre">learning_rate</span></code> 与当前的衰减因子的乘积。以使用Adam算法为例，代码如下：</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span>
+    <span class="n">learning_rate</span><span class="o">=</span><span class="mf">1e-3</span><span class="p">,</span>
+    <span class="n">learning_rate_schedule</span><span class="o">=</span><span class="s2">&quot;manual&quot;</span><span class="p">,</span>
+    <span class="n">learning_rate_args</span><span class="o">=</span><span class="s2">&quot;1000:1.0,2000:0.9,3000:0.8&quot;</span><span class="p">,)</span>
+</pre></div>
+</div>
+<p>在该示例中，当已训练样本数小于等于1000时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">1.0</span></code>；当已训练样本数大于1000小于等于2000时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">0.9</span></code>；当已训练样本数大于2000时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">0.8</span></code>。</p>
+</li>
+<li><p class="first">&#8220;pass_manual&#8221;</p>
+<p>这是一种按已训练pass数分段取值的学习率退火方法。使用该learning_rate_schedule时，用户通过参数 <code class="code docutils literal"><span class="pre">learning_rate_args</span></code> 设置学习率衰减因子分段函数，当前的学习率为所设置 <code class="code docutils literal"><span class="pre">learning_rate</span></code> 与当前的衰减因子的乘积。以使用Adam算法为例，代码如下：</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">optimizer</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">optimizer</span><span class="o">.</span><span class="n">Adam</span><span class="p">(</span>
+    <span class="n">learning_rate</span><span class="o">=</span><span class="mf">1e-3</span><span class="p">,</span>
+    <span class="n">learning_rate_schedule</span><span class="o">=</span><span class="s2">&quot;manual&quot;</span><span class="p">,</span>
+    <span class="n">learning_rate_args</span><span class="o">=</span><span class="s2">&quot;1:1.0,2:0.9,3:0.8&quot;</span><span class="p">,)</span>
+</pre></div>
+</div>
+<p>在该示例中，当已训练pass数小于等于1时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">1.0</span></code>；当已训练pass数大于1小于等于2时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">0.9</span></code>；当已训练pass数大于2时，学习率为 <code class="code docutils literal"><span class="pre">1e-3</span> <span class="pre">*</span> <span class="pre">0.8</span></code>。</p>
+</li>
+</ul>
+</div>
+<div class="section" id="id2">
+<h2><a class="toc-backref" href="#id9">3. 如何初始化参数</a><a class="headerlink" href="#id2" title="永久链接至标题">¶</a></h2>
+<p>默认情况下，PaddlePaddle使用均值0，标准差为 <span class="math">\(\frac{1}{\sqrt{d}}\)</span> 来初始化参数。其中 <span class="math">\(d\)</span> 为参数矩阵的宽度。这种初始化方式在一般情况下不会产生很差的结果。如果用户想要自定义初始化方式，PaddlePaddle目前提供两种参数初始化的方式:</p>
+<ul class="simple">
+<li>高斯分布。将 <code class="code docutils literal"><span class="pre">param_attr</span></code> 设置成 <code class="code docutils literal"><span class="pre">param_attr=ParamAttr(initial_mean=0.0,</span> <span class="pre">initial_std=1.0)</span></code></li>
+<li>均匀分布。将 <code class="code docutils literal"><span class="pre">param_attr</span></code> 设置成 <code class="code docutils literal"><span class="pre">param_attr=ParamAttr(initial_max=1.0,</span> <span class="pre">initial_min=-1.0)</span></code></li>
+</ul>
+<p>比如设置一个全连接层的参数初始化方式和bias初始化方式，可以使用如下代码。</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">hidden</span> <span class="o">=</span> <span class="n">fc_layer</span><span class="p">(</span><span class="nb">input</span><span class="o">=</span><span class="n">ipt</span><span class="p">,</span> <span class="n">param_attr</span><span class="o">=</span><span class="n">ParamAttr</span><span class="p">(</span><span class="n">initial_max</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">initial_min</span><span class="o">=-</span><span class="mf">1.0</span><span class="p">),</span>
+                  <span class="n">bias_attr</span><span class="o">=</span><span class="n">ParamAttr</span><span class="p">(</span><span class="n">initial_mean</span><span class="o">=</span><span class="mf">1.0</span><span class="p">,</span> <span class="n">initial_std</span><span class="o">=</span><span class="mf">0.0</span><span class="p">))</span>
+</pre></div>
+</div>
+<p>上述代码将bias全部初始化为1.0, 同时将参数初始化为 <code class="code docutils literal"><span class="pre">[1.0,</span> <span class="pre">-1.0]</span></code> 的均匀分布。</p>
+</div>
+<div class="section" id="id3">
+<h2><a class="toc-backref" href="#id10">4. 如何共享参数</a><a class="headerlink" href="#id3" title="永久链接至标题">¶</a></h2>
+<p>PaddlePaddle的参数使用名字 <code class="code docutils literal"><span class="pre">name</span></code> 作为参数的ID，相同名字的参数，会共享参数。设置参数的名字，可以使用 <code class="code docutils literal"><span class="pre">ParamAttr(name=&quot;YOUR_PARAM_NAME&quot;)</span></code> 来设置。更方便的设置方式，是使得要共享的参数使用同样的 <code class="code docutils literal"><span class="pre">ParamAttr</span></code> 对象。</p>
+<p>简单的全连接网络，参数共享的配置示例为:</p>
+<p>这里 <code class="code docutils literal"><span class="pre">hidden_a</span></code> 和 <code class="code docutils literal"><span class="pre">hidden_b</span></code> 使用了同样的parameter和bias。并且softmax层的两个输入也使用了同样的参数 <code class="code docutils literal"><span class="pre">softmax_param</span></code>。</p>
+</div>
+<div class="section" id="id4">
+<h2><a class="toc-backref" href="#id11">5. 如何加载预训练参数</a><a class="headerlink" href="#id4" title="永久链接至标题">¶</a></h2>
+<ul class="simple">
+<li>对加载预训练参数的层，设置其参数属性 <code class="code docutils literal"><span class="pre">is_static=True</span></code>，使该层的参数在训练过程中保持不变。以embedding层为例，代码如下：</li>
+</ul>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">emb_para</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">attr</span><span class="o">.</span><span class="n">Param</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="n">is_static</span><span class="o">=</span><span class="bp">True</span><span class="p">)</span>
+<span class="n">paddle</span><span class="o">.</span><span class="n">layer</span><span class="o">.</span><span class="n">embedding</span><span class="p">(</span><span class="n">size</span><span class="o">=</span><span class="n">word_dim</span><span class="p">,</span> <span class="nb">input</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">param_attr</span><span class="o">=</span><span class="n">emb_para</span><span class="p">)</span>
+</pre></div>
+</div>
+<ul class="simple">
+<li>从模型文件将预训练参数载入 <code class="code docutils literal"><span class="pre">numpy.array</span></code>，在创建parameters后，使用 <code class="code docutils literal"><span class="pre">parameters.set()</span></code> 加载预训练参数。PaddlePaddle保存的模型参数文件前16字节为头信息，用户将参数载入 <code class="code docutils literal"><span class="pre">numpy.array</span></code> 时须从第17字节开始。以embedding层为例，代码如下：</li>
+</ul>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">load_parameter</span><span class="p">(</span><span class="n">file_name</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">):</span>
+    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">file_name</span><span class="p">,</span> <span class="s1">&#39;rb&#39;</span><span class="p">)</span> <span class="k">as</span> <span class="n">f</span><span class="p">:</span>
+        <span class="n">f</span><span class="o">.</span><span class="n">read</span><span class="p">(</span><span class="mi">16</span><span class="p">)</span>  <span class="c1"># skip header.</span>
+        <span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">fromfile</span><span class="p">(</span><span class="n">f</span><span class="p">,</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">)</span>
+
+<span class="n">parameters</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">.</span><span class="n">parameters</span><span class="o">.</span><span class="n">create</span><span class="p">(</span><span class="n">my_cost</span><span class="p">)</span>
+<span class="n">parameters</span><span class="o">.</span><span class="n">set</span><span class="p">(</span><span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="n">load_parameter</span><span class="p">(</span><span class="n">emb_param_file</span><span class="p">,</span> <span class="mi">30000</span><span class="p">,</span> <span class="mi">256</span><span class="p">))</span>
+</pre></div>
+</div>
+</div>
+<div class="section" id="id5">
+<h2><a class="toc-backref" href="#id12">6. 存储的参数格式是什么，如何和明文进行相互转化</a><a class="headerlink" href="#id5" title="永久链接至标题">¶</a></h2>
+<p>PaddlePaddle保存的模型参数文件内容由16字节头信息和网络参数两部分组成。头信息中，1~4字节表示PaddlePaddle版本信息，请直接填充0；5~8字节表示每个参数占用的字节数，当保存的网络参数为float类型时为4，double类型时为8；9~16字节表示保存的参数总个数。</p>
+<p>将PaddlePaddle保存的模型参数还原回明文时，可以使用相应数据类型的 <code class="code docutils literal"><span class="pre">numpy.array</span></code> 加载具体网络参数，此时可以跳过PaddlePaddle模型参数文件的头信息。若在PaddlePaddle编译时，未指定按照double精度编译，默认情况下按照float精度计算，保存的参数也是float类型。这时在使用 <code class="code docutils literal"><span class="pre">numpy.array</span></code> 时，一般设置 <code class="code docutils literal"><span class="pre">dtype=float32</span></code> 。示例如下：</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">read_parameter</span><span class="p">(</span><span class="n">fname</span><span class="p">,</span> <span class="n">width</span><span class="p">):</span>
+    <span class="n">s</span> <span class="o">=</span> <span class="nb">open</span><span class="p">(</span><span class="n">fname</span><span class="p">)</span><span class="o">.</span><span class="n">read</span><span class="p">()</span>
+    <span class="c1"># skip header</span>
+    <span class="n">vec</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">fromstring</span><span class="p">(</span><span class="n">s</span><span class="p">[</span><span class="mi">16</span><span class="p">:],</span> <span class="n">dtype</span><span class="o">=</span><span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">)</span>
+    <span class="c1"># width is the size of the corresponding layer</span>
+    <span class="n">np</span><span class="o">.</span><span class="n">savetxt</span><span class="p">(</span><span class="n">fname</span> <span class="o">+</span> <span class="s2">&quot;.csv&quot;</span><span class="p">,</span> <span class="n">vec</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">width</span><span class="p">,</span> <span class="o">-</span><span class="mi">1</span><span class="p">),</span>
+            <span class="n">fmt</span><span class="o">=</span><span class="s2">&quot;</span><span class="si">%.6f</span><span class="s2">&quot;</span><span class="p">,</span> <span class="n">delimiter</span><span class="o">=</span><span class="s2">&quot;,&quot;</span><span class="p">)</span>
+</pre></div>
+</div>
+<p>将明文参数转化为PaddlePaddle可加载的模型参数时，首先构造头信息，再写入网络参数。下面的代码将随机生成的矩阵转化为可以被PaddlePaddle加载的模型参数。</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="k">def</span> <span class="nf">gen_rand_param</span><span class="p">(</span><span class="n">param_file</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">height</span><span class="p">,</span> <span class="n">need_trans</span><span class="p">):</span>
+    <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">seed</span><span class="p">()</span>
+    <span class="n">header</span> <span class="o">=</span> <span class="n">struct</span><span class="o">.</span><span class="n">pack</span><span class="p">(</span><span class="s2">&quot;iil&quot;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">4</span><span class="p">,</span> <span class="n">height</span> <span class="o">*</span> <span class="n">width</span><span class="p">)</span>
+    <span class="n">param</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">float32</span><span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">rand</span><span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">))</span>
+    <span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="n">param_file</span><span class="p">,</span> <span class="s2">&quot;w&quot;</span><span class="p">)</span> <span class="k">as</span> <span class="n">fparam</span><span class="p">:</span>
+        <span class="n">fparam</span><span class="o">.</span><span class="n">write</span><span class="p">(</span><span class="n">header</span> <span class="o">+</span> <span class="n">param</span><span class="o">.</span><span class="n">tostring</span><span class="p">())</span>
+</pre></div>
+</div>
+</div>
+<div class="section" id="a-protocol-message-was-rejected-because-it-was-too-big">
+<h2><a class="toc-backref" href="#id13">7. A protocol message was rejected because it was too big</a><a class="headerlink" href="#a-protocol-message-was-rejected-because-it-was-too-big" title="永久链接至标题">¶</a></h2>
+<p>如果在训练NLP相关模型时，出现以下错误：</p>
+<div class="highlight-bash"><div class="highlight"><pre><span></span><span class="o">[</span>libprotobuf ERROR google/protobuf/io/coded_stream.cc:171<span class="o">]</span> A protocol message was rejected because it was too big <span class="o">(</span>more than <span class="m">67108864</span> bytes<span class="o">)</span>.  To increase the limit <span class="o">(</span>or to disable these warnings<span class="o">)</span>, see CodedInputStream::SetTotalBytesLimit<span class="o">()</span> in google/protobuf/io/coded_stream.h.
+F1205 <span class="m">14</span>:59:50.295174 <span class="m">14703</span> TrainerConfigHelper.cpp:59<span class="o">]</span> Check failed: m-&gt;conf.ParseFromString<span class="o">(</span>configProtoStr<span class="o">)</span>
+</pre></div>
+</div>
+<p>可能的原因是：传给dataprovider的某一个args过大，一般是由于直接传递大字典导致的。错误的define_py_data_sources2类似：</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">src_dict</span> <span class="o">=</span> <span class="nb">dict</span><span class="p">()</span>
+<span class="k">for</span> <span class="n">line_count</span><span class="p">,</span> <span class="n">line</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="nb">open</span><span class="p">(</span><span class="n">src_dict_path</span><span class="p">,</span> <span class="s2">&quot;r&quot;</span><span class="p">)):</span>
+   <span class="n">src_dict</span><span class="p">[</span><span class="n">line</span><span class="o">.</span><span class="n">strip</span><span class="p">()]</span> <span class="o">=</span> <span class="n">line_count</span>
+
+<span class="n">define_py_data_sources2</span><span class="p">(</span>
+   <span class="n">train_list</span><span class="p">,</span>
+   <span class="n">test_list</span><span class="p">,</span>
+   <span class="n">module</span><span class="o">=</span><span class="s2">&quot;dataprovider&quot;</span><span class="p">,</span>
+   <span class="n">obj</span><span class="o">=</span><span class="s2">&quot;process&quot;</span><span class="p">,</span>
+   <span class="n">args</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;src_dict&quot;</span><span class="p">:</span> <span class="n">src_dict</span><span class="p">})</span>
+</pre></div>
+</div>
+<p>解决方案是：将字典的地址作为args传给dataprovider，然后在dataprovider里面根据该地址加载字典。即define_py_data_sources2应改为：</p>
+<div class="highlight-python"><div class="highlight"><pre><span></span><span class="n">define_py_data_sources2</span><span class="p">(</span>
+   <span class="n">train_list</span><span class="p">,</span>
+   <span class="n">test_list</span><span class="p">,</span>
+   <span class="n">module</span><span class="o">=</span><span class="s2">&quot;dataprovider&quot;</span><span class="p">,</span>
+   <span class="n">obj</span><span class="o">=</span><span class="s2">&quot;process&quot;</span><span class="p">,</span>
+   <span class="n">args</span><span class="o">=</span><span class="p">{</span><span class="s2">&quot;src_dict_path&quot;</span><span class="p">:</span> <span class="n">src_dict_path</span><span class="p">})</span>
+</pre></div>
+</div>
+<p>完整源码可参考 <a class="reference external" href="https://github.com/PaddlePaddle/Paddle/tree/develop/demo/seqToseq">seqToseq</a> 示例。</p>
+</div>
+</div>
+
+
+           </div>
+          </div>
+          <footer>
+  
+    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
+      
+        <a href="../local/index_cn.html" class="btn btn-neutral float-right" title="本地训练与预测" accesskey="n">Next <span class="fa fa-arrow-circle-right"></span></a>
+      
+      
+        <a href="../model/index_cn.html" class="btn btn-neutral" title="模型配置" accesskey="p"><span class="fa fa-arrow-circle-left"></span> Previous</a>
+      
+    </div>
+  
+
+  <hr/>
+
+  <div role="contentinfo">
+    <p>
+        &copy; Copyright 2016, PaddlePaddle developers.
+
+    </p>
+  </div>
+  Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>. 
+
+</footer>
+
+        </div>
+      </div>
+
+    </section>
+
+  </div>
+  
+
+
+  
+
+    <script type="text/javascript">
+        var DOCUMENTATION_OPTIONS = {
+            URL_ROOT:'../../',
+            VERSION:'',
+            COLLAPSE_INDEX:false,
+            FILE_SUFFIX:'.html',
+            HAS_SOURCE:  true,
+            SOURCELINK_SUFFIX: ".txt",
+        };
+    </script>
+      <script type="text/javascript" src="../../_static/jquery.js"></script>
+      <script type="text/javascript" src="../../_static/underscore.js"></script>
+      <script type="text/javascript" src="../../_static/doctools.js"></script>
+      <script type="text/javascript" src="../../_static/translations.js"></script>
+      <script type="text/javascript" src="https://cdn.bootcss.com/mathjax/2.7.0/MathJax.js"></script>
+       
+  
+
+  
+  
+    <script type="text/javascript" src="../../_static/js/theme.js"></script>
+  
+  
+  <script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/js/bootstrap.min.js" integrity="sha384-Tc5IQib027qvyjSMfHjOMaLkfuWVxZxUPnCJA7l2mCWNIpG9mGCD8wGNIcPD7Txa" crossorigin="anonymous"></script>
+  <script src="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/js/perfect-scrollbar.jquery.min.js"></script>
+  <script src="../../_static/js/paddle_doc_init.js"></script> 
+
+</body>
+</html>
\ No newline at end of file
diff --git a/develop/doc_cn/genindex.html b/develop/doc_cn/genindex.html
index d377c904a82aaa37520ee2925dd128f53e775042..7b34d895019c4c7c5ba58691389d47d3414b6885 100644
--- a/develop/doc_cn/genindex.html
+++ b/develop/doc_cn/genindex.html
@@ -155,7 +155,14 @@
 <li class="toctree-l2"><a class="reference internal" href="api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/getstarted/basic_usage/index_cn.html b/develop/doc_cn/getstarted/basic_usage/index_cn.html
index a3b02cb6f96e5bf3803f60a54253a212cbf33605..3f071dac6cded42fe197d183fe8f7015b00f9f8c 100644
--- a/develop/doc_cn/getstarted/basic_usage/index_cn.html
+++ b/develop/doc_cn/getstarted/basic_usage/index_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/getstarted/build_and_install/cmake/build_from_source_cn.html b/develop/doc_cn/getstarted/build_and_install/cmake/build_from_source_cn.html
index 2597d99992c9a5bd710adaca03129bab6a6cf522..09ea067b9e672563c5ab712d630be0edcdf0283c 100644
--- a/develop/doc_cn/getstarted/build_and_install/cmake/build_from_source_cn.html
+++ b/develop/doc_cn/getstarted/build_and_install/cmake/build_from_source_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/getstarted/build_and_install/docker_install_cn.html b/develop/doc_cn/getstarted/build_and_install/docker_install_cn.html
index 2bf4d885e6541f64797efe501ea630e5d79afad2..7f696afecae3aa21ebb5389236b850faf6533f9a 100644
--- a/develop/doc_cn/getstarted/build_and_install/docker_install_cn.html
+++ b/develop/doc_cn/getstarted/build_and_install/docker_install_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/getstarted/build_and_install/index_cn.html b/develop/doc_cn/getstarted/build_and_install/index_cn.html
index 379504277dbe6c8c818cec88c1f23258a1819780..85ac6cab9ae79cce000b2929a09bbae3ca711b0f 100644
--- a/develop/doc_cn/getstarted/build_and_install/index_cn.html
+++ b/develop/doc_cn/getstarted/build_and_install/index_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/getstarted/concepts/use_concepts_cn.html b/develop/doc_cn/getstarted/concepts/use_concepts_cn.html
index f19cfc8459c0370c8d0068eb6555b9c12f1aec99..e009ad33d9ee16720bbc63828f35c04f9e401859 100644
--- a/develop/doc_cn/getstarted/concepts/use_concepts_cn.html
+++ b/develop/doc_cn/getstarted/concepts/use_concepts_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/getstarted/index_cn.html b/develop/doc_cn/getstarted/index_cn.html
index 466829a7ba50a249224cf666ba2dbfe153f2737d..753c0a2dda1a6846ca5fc40db62b9176005fac7a 100644
--- a/develop/doc_cn/getstarted/index_cn.html
+++ b/develop/doc_cn/getstarted/index_cn.html
@@ -156,7 +156,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/cross_compiling/cross_compiling_for_android_cn.html b/develop/doc_cn/howto/cross_compiling/cross_compiling_for_android_cn.html
index 950dd1ea48e61f12d6b4781f6288eedbf3f42731..2a1706facb19410a64757332dc3d5d8e736d0bc9 100644
--- a/develop/doc_cn/howto/cross_compiling/cross_compiling_for_android_cn.html
+++ b/develop/doc_cn/howto/cross_compiling/cross_compiling_for_android_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/cross_compiling/cross_compiling_for_raspberry_cn.html b/develop/doc_cn/howto/cross_compiling/cross_compiling_for_raspberry_cn.html
index 4babf29652c7cadab41a1caf0c774f1efd84a48d..46cc230e95b3d79603d5d850f33b7fac32589a17 100644
--- a/develop/doc_cn/howto/cross_compiling/cross_compiling_for_raspberry_cn.html
+++ b/develop/doc_cn/howto/cross_compiling/cross_compiling_for_raspberry_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/deep_model/rnn/hierarchical_layer_cn.html b/develop/doc_cn/howto/deep_model/rnn/hierarchical_layer_cn.html
index 50eed7403ac82163add15187e498ce27c280fea9..c9a332767ffd477146ce386a3f1db475d4620850 100644
--- a/develop/doc_cn/howto/deep_model/rnn/hierarchical_layer_cn.html
+++ b/develop/doc_cn/howto/deep_model/rnn/hierarchical_layer_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/deep_model/rnn/hrnn_rnn_api_compare_cn.html b/develop/doc_cn/howto/deep_model/rnn/hrnn_rnn_api_compare_cn.html
index eafb68323d791d2631ae4f56e577833542a294f7..d5e4129a36de062e4c3772be1ba4af1b3ac75548 100644
--- a/develop/doc_cn/howto/deep_model/rnn/hrnn_rnn_api_compare_cn.html
+++ b/develop/doc_cn/howto/deep_model/rnn/hrnn_rnn_api_compare_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/deep_model/rnn/index_cn.html b/develop/doc_cn/howto/deep_model/rnn/index_cn.html
index 3f1fb4df57e8798213fc755b78429c57c78179ee..e49714797e37445e60ed104a3580264bb8e05e50 100644
--- a/develop/doc_cn/howto/deep_model/rnn/index_cn.html
+++ b/develop/doc_cn/howto/deep_model/rnn/index_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/deep_model/rnn/recurrent_group_cn.html b/develop/doc_cn/howto/deep_model/rnn/recurrent_group_cn.html
index 99409297903fc02b3c0fc21b7aa3d21a9e6a27d6..9638428c38643979f653fa845e624a211015848d 100644
--- a/develop/doc_cn/howto/deep_model/rnn/recurrent_group_cn.html
+++ b/develop/doc_cn/howto/deep_model/rnn/recurrent_group_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/deep_model/rnn/rnn_config_cn.html b/develop/doc_cn/howto/deep_model/rnn/rnn_config_cn.html
index b88519373b0c0fe7931351baf1edb7af116feae0..e8bf609799c4b8308895195ab53fa4566bf675f8 100644
--- a/develop/doc_cn/howto/deep_model/rnn/rnn_config_cn.html
+++ b/develop/doc_cn/howto/deep_model/rnn/rnn_config_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/dev/build_cn.html b/develop/doc_cn/howto/dev/build_cn.html
index f41a28b75a72d2345def63ead60924d4ca886950..e79abbd8e26aefcf726dca5926ee56f7da3bdbc3 100644
--- a/develop/doc_cn/howto/dev/build_cn.html
+++ b/develop/doc_cn/howto/dev/build_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/dev/contribute_to_paddle_cn.html b/develop/doc_cn/howto/dev/contribute_to_paddle_cn.html
index 4febadcc5e4b0ff932e2847b0b3b354d229530b2..0f8254bdac279acbcf4665fb5cfdcce8c4b6ca36 100644
--- a/develop/doc_cn/howto/dev/contribute_to_paddle_cn.html
+++ b/develop/doc_cn/howto/dev/contribute_to_paddle_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/dev/new_layer_cn.html b/develop/doc_cn/howto/dev/new_layer_cn.html
index c247a8afe0a12ce4407ad1e3f5c6f5352310992c..0362f350cc15d0c918398dde383fa591a1a903bd 100644
--- a/develop/doc_cn/howto/dev/new_layer_cn.html
+++ b/develop/doc_cn/howto/dev/new_layer_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/dev/new_op_cn.html b/develop/doc_cn/howto/dev/new_op_cn.html
index b72308cde152e619c9dbf083b5447a9f009f254e..01e260ebe57a1860c819bdf9cc3b82a6c49d58d1 100644
--- a/develop/doc_cn/howto/dev/new_op_cn.html
+++ b/develop/doc_cn/howto/dev/new_op_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/dev/use_eigen_cn.html b/develop/doc_cn/howto/dev/use_eigen_cn.html
index 4a75e1677f3475f530585ccbf94e178bce6c0200..7579f7af1d9f9270e17e2803075e905b6b92c549 100644
--- a/develop/doc_cn/howto/dev/use_eigen_cn.html
+++ b/develop/doc_cn/howto/dev/use_eigen_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/dev/write_docs_cn.html b/develop/doc_cn/howto/dev/write_docs_cn.html
index 5e657d01f1e0886b971702a6c039028b6aa7f0c5..703736902c14e60a2702060fb5fe1bf49db07634 100644
--- a/develop/doc_cn/howto/dev/write_docs_cn.html
+++ b/develop/doc_cn/howto/dev/write_docs_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/index_cn.html b/develop/doc_cn/howto/index_cn.html
index d0e529b39524833d73665ffcbdd0ac9c16a45182..9b5fcf15e5baed1f13f21f5af5547b8bae116cad 100644
--- a/develop/doc_cn/howto/index_cn.html
+++ b/develop/doc_cn/howto/index_cn.html
@@ -156,7 +156,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/optimization/gpu_profiling_cn.html b/develop/doc_cn/howto/optimization/gpu_profiling_cn.html
index 8e38bc8b90c3397b79be3e3fcfce992e7650883d..3e7149714576e8e607947277ae7c871a2a33a24a 100644
--- a/develop/doc_cn/howto/optimization/gpu_profiling_cn.html
+++ b/develop/doc_cn/howto/optimization/gpu_profiling_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/usage/cluster/cluster_train_cn.html b/develop/doc_cn/howto/usage/cluster/cluster_train_cn.html
index ead9b44494827aab8f6c13679a0834d342eca1be..8a0ed421530afc14b3a0843ded66837ded04a01c 100644
--- a/develop/doc_cn/howto/usage/cluster/cluster_train_cn.html
+++ b/develop/doc_cn/howto/usage/cluster/cluster_train_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/usage/cmd_parameter/arguments_cn.html b/develop/doc_cn/howto/usage/cmd_parameter/arguments_cn.html
index 0add275ddb8b712428a5a8d8bc19b97a19f1d389..663b9df23cd3eb36946049b2873a9bc7ae24982d 100644
--- a/develop/doc_cn/howto/usage/cmd_parameter/arguments_cn.html
+++ b/develop/doc_cn/howto/usage/cmd_parameter/arguments_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/usage/cmd_parameter/detail_introduction_cn.html b/develop/doc_cn/howto/usage/cmd_parameter/detail_introduction_cn.html
index 53b8e419b5973450a44648294f77693b6da36b33..4959e893d222368f24ca1a6ed76a98b3008f8273 100644
--- a/develop/doc_cn/howto/usage/cmd_parameter/detail_introduction_cn.html
+++ b/develop/doc_cn/howto/usage/cmd_parameter/detail_introduction_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/usage/cmd_parameter/index_cn.html b/develop/doc_cn/howto/usage/cmd_parameter/index_cn.html
index 252bfeb8d34a5d908f003ff1d317a1352937d816..71ffb2d10542f5a902dca4e4729a583b0e5911ca 100644
--- a/develop/doc_cn/howto/usage/cmd_parameter/index_cn.html
+++ b/develop/doc_cn/howto/usage/cmd_parameter/index_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/usage/cmd_parameter/use_case_cn.html b/develop/doc_cn/howto/usage/cmd_parameter/use_case_cn.html
index 19e503afb0ddeac73b1134a36a2a0000ad582ae3..e6c47db2eab7c1e5736478a3178d4e044db2c0d6 100644
--- a/develop/doc_cn/howto/usage/cmd_parameter/use_case_cn.html
+++ b/develop/doc_cn/howto/usage/cmd_parameter/use_case_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/usage/k8s/k8s_basis_cn.html b/develop/doc_cn/howto/usage/k8s/k8s_basis_cn.html
index bf947981c5987a51a981bdcb081192788a2d51bf..bc493a9f84b6e45dbe84ea2a9f28900bf1f2b99e 100644
--- a/develop/doc_cn/howto/usage/k8s/k8s_basis_cn.html
+++ b/develop/doc_cn/howto/usage/k8s/k8s_basis_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/usage/k8s/k8s_cn.html b/develop/doc_cn/howto/usage/k8s/k8s_cn.html
index 7d475cb15db8415e0bc8d16743ebecaa693e3c0e..1a33b76430ed6c9a4e320316b36579cbc60a1c25 100644
--- a/develop/doc_cn/howto/usage/k8s/k8s_cn.html
+++ b/develop/doc_cn/howto/usage/k8s/k8s_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/usage/k8s/k8s_distributed_cn.html b/develop/doc_cn/howto/usage/k8s/k8s_distributed_cn.html
index 61e58e9f7b18cee95015e9d9bd10a08aca8d6646..5591f9a3bd07f5e2f848e601cb7ad53c18933fcc 100644
--- a/develop/doc_cn/howto/usage/k8s/k8s_distributed_cn.html
+++ b/develop/doc_cn/howto/usage/k8s/k8s_distributed_cn.html
@@ -157,7 +157,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/usage/k8s/src/k8s_data/README.html b/develop/doc_cn/howto/usage/k8s/src/k8s_data/README.html
index a076601ad436b2495d122205d8f0fadfdecc3b7b..bfdba92bf51809ee2bbfc9054134468ae4736427 100644
--- a/develop/doc_cn/howto/usage/k8s/src/k8s_data/README.html
+++ b/develop/doc_cn/howto/usage/k8s/src/k8s_data/README.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/howto/usage/k8s/src/k8s_train/README.html b/develop/doc_cn/howto/usage/k8s/src/k8s_train/README.html
index 87f8fb6746e3207124e7b2647c236ea05d73d731..3eadecc0d8194148e56cc12f2d373bfd9988d740 100644
--- a/develop/doc_cn/howto/usage/k8s/src/k8s_train/README.html
+++ b/develop/doc_cn/howto/usage/k8s/src/k8s_train/README.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../../../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../../../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../../../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../../../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../../../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/index_cn.html b/develop/doc_cn/index_cn.html
index 52ed7b511f626360606736540ffff05ba88c2b5e..00be7085025471dc0e1ac5d8e144fc3ceaafc5ca 100644
--- a/develop/doc_cn/index_cn.html
+++ b/develop/doc_cn/index_cn.html
@@ -155,7 +155,14 @@
 <li class="toctree-l2"><a class="reference internal" href="api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/objects.inv b/develop/doc_cn/objects.inv
index 6380de30b8dfae966c8b48c83a7d9168c223dad7..1d637ae0b21c3ce2881e32bf0e76d9d1c9420975 100644
Binary files a/develop/doc_cn/objects.inv and b/develop/doc_cn/objects.inv differ
diff --git a/develop/doc_cn/search.html b/develop/doc_cn/search.html
index 2a13d4963c5b7737b332c2c69337c2e4e96581b1..8fae4fc8d6e322b093b8a4b01cfd581d08d34c38 100644
--- a/develop/doc_cn/search.html
+++ b/develop/doc_cn/search.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/searchindex.js b/develop/doc_cn/searchindex.js
index 705f8594f0b138e4cb09522e7314fa278701e6a5..fc22346f32889ce2f718a7816d277b1da47ce8a7 100644
--- a/develop/doc_cn/searchindex.js
+++ b/develop/doc_cn/searchindex.js
@@ -1 +1 @@
-Search.setIndex({docnames:["api/index_cn","api/v1/data_provider/dataprovider_cn","api/v1/data_provider/pydataprovider2_cn","api/v1/index_cn","api/v1/predict/swig_py_paddle_cn","api/v2/config/activation","api/v2/config/attr","api/v2/config/evaluators","api/v2/config/layer","api/v2/config/networks","api/v2/config/optimizer","api/v2/config/pooling","api/v2/data","api/v2/model_configs","api/v2/run_logic","design/api","design/auto_gradient_check","design/block","design/build_system/README","design/cluster_train/README","design/cluster_train/checkpointing","design/cluster_train/data_dispatch","design/cluster_train/large_model_dist_train","design/cluster_train/master_server","design/cluster_train/pserver_client","design/cluster_train/remote_parameter_updater","design/cluster_train/save_model","design/cluster_train/submit-job","design/file_manager/README","design/file_manager/pfs/pfsclient","design/functions_operators_layers","design/graph","design/if_else_op","design/multi_language_interface/00.why_plain_c","design/multi_language_interface/01.inference_implementation","design/ops/rnn","design/parameters_in_cpp","design/program","design/reader/README","design/refactor/distributed_architecture","design/refactor/parameter_server","design/refactorization","design/releasing_process","design/scope","design/simple_op_design","design/var_desc","faq/index_cn","getstarted/basic_usage/index_cn","getstarted/build_and_install/cmake/build_from_source_cn","getstarted/build_and_install/docker_install_cn","getstarted/build_and_install/index_cn","getstarted/concepts/use_concepts_cn","getstarted/index_cn","howto/cross_compiling/cross_compiling_for_android_cn","howto/cross_compiling/cross_compiling_for_raspberry_cn","howto/deep_model/rnn/hierarchical_layer_cn","howto/deep_model/rnn/hrnn_rnn_api_compare_cn","howto/deep_model/rnn/index_cn","howto/deep_model/rnn/recurrent_group_cn","howto/deep_model/rnn/rnn_config_cn","howto/dev/build_cn","howto/dev/contribute_to_paddle_cn","howto/dev/new_layer_cn","howto/dev/new_op_cn","howto/dev/use_eigen_cn","howto/dev/write_docs_cn","howto/index_cn","howto/optimization/gpu_profiling_cn","howto/usage/cluster/cluster_train_cn","howto/usage/cmd_parameter/arguments_cn","howto/usage/cmd_parameter/detail_introduction_cn","howto/usage/cmd_parameter/index_cn","howto/usage/cmd_parameter/use_case_cn","howto/usage/k8s/k8s_basis_cn","howto/usage/k8s/k8s_cn","howto/usage/k8s/k8s_distributed_cn","howto/usage/k8s/src/k8s_data/README","howto/usage/k8s/src/k8s_train/README","index_cn","survey/cluster_bootstrapping_tools","tutorials/embedding_model/index_cn","tutorials/image_classification/index_cn","tutorials/imagenet_model/resnet_model_cn","tutorials/index_cn","tutorials/quick_start/index_cn","tutorials/rec/ml_dataset_cn","tutorials/rec/ml_regression_cn","tutorials/semantic_role_labeling/index_cn","tutorials/sentiment_analysis/index_cn","tutorials/text_generation/index_cn"],envversion:50,filenames:["api/index_cn.rst","api/v1/data_provider/dataprovider_cn.rst","api/v1/data_provider/pydataprovider2_cn.rst","api/v1/index_cn.rst","api/v1/predict/swig_py_paddle_cn.rst","api/v2/config/activation.rst","api/v2/config/attr.rst","api/v2/config/evaluators.rst","api/v2/config/layer.rst","api/v2/config/networks.rst","api/v2/config/optimizer.rst","api/v2/config/pooling.rst","api/v2/data.rst","api/v2/model_configs.rst","api/v2/run_logic.rst","design/api.md","design/auto_gradient_check.md","design/block.md","design/build_system/README.md","design/cluster_train/README.md","design/cluster_train/checkpointing.md","design/cluster_train/data_dispatch.md","design/cluster_train/large_model_dist_train.md","design/cluster_train/master_server.md","design/cluster_train/pserver_client.md","design/cluster_train/remote_parameter_updater.md","design/cluster_train/save_model.md","design/cluster_train/submit-job.md","design/file_manager/README.md","design/file_manager/pfs/pfsclient.md","design/functions_operators_layers.md","design/graph.md","design/if_else_op.md","design/multi_language_interface/00.why_plain_c.md","design/multi_language_interface/01.inference_implementation.md","design/ops/rnn.md","design/parameters_in_cpp.md","design/program.md","design/reader/README.md","design/refactor/distributed_architecture.md","design/refactor/parameter_server.md","design/refactorization.md","design/releasing_process.md","design/scope.md","design/simple_op_design.md","design/var_desc.md","faq/index_cn.rst","getstarted/basic_usage/index_cn.rst","getstarted/build_and_install/cmake/build_from_source_cn.rst","getstarted/build_and_install/docker_install_cn.rst","getstarted/build_and_install/index_cn.rst","getstarted/concepts/use_concepts_cn.rst","getstarted/index_cn.rst","howto/cross_compiling/cross_compiling_for_android_cn.md","howto/cross_compiling/cross_compiling_for_raspberry_cn.md","howto/deep_model/rnn/hierarchical_layer_cn.rst","howto/deep_model/rnn/hrnn_rnn_api_compare_cn.rst","howto/deep_model/rnn/index_cn.rst","howto/deep_model/rnn/recurrent_group_cn.md","howto/deep_model/rnn/rnn_config_cn.rst","howto/dev/build_cn.md","howto/dev/contribute_to_paddle_cn.md","howto/dev/new_layer_cn.rst","howto/dev/new_op_cn.md","howto/dev/use_eigen_cn.md","howto/dev/write_docs_cn.rst","howto/index_cn.rst","howto/optimization/gpu_profiling_cn.rst","howto/usage/cluster/cluster_train_cn.md","howto/usage/cmd_parameter/arguments_cn.md","howto/usage/cmd_parameter/detail_introduction_cn.md","howto/usage/cmd_parameter/index_cn.rst","howto/usage/cmd_parameter/use_case_cn.md","howto/usage/k8s/k8s_basis_cn.md","howto/usage/k8s/k8s_cn.md","howto/usage/k8s/k8s_distributed_cn.md","howto/usage/k8s/src/k8s_data/README.md","howto/usage/k8s/src/k8s_train/README.md","index_cn.rst","survey/cluster_bootstrapping_tools.md","tutorials/embedding_model/index_cn.md","tutorials/image_classification/index_cn.md","tutorials/imagenet_model/resnet_model_cn.md","tutorials/index_cn.md","tutorials/quick_start/index_cn.rst","tutorials/rec/ml_dataset_cn.md","tutorials/rec/ml_regression_cn.rst","tutorials/semantic_role_labeling/index_cn.md","tutorials/sentiment_analysis/index_cn.md","tutorials/text_generation/index_cn.md"],objects:{},objnames:{},objtypes:{},terms:{"00012\u7684\u6a21\u578b\u6709\u7740\u6700\u9ad8\u7684bleu\u503c27":89,"0005\u4e58\u4ee5batch":81,"000\u4e2a\u5df2\u6807\u6ce8\u8fc7\u7684\u9ad8\u6781\u6027\u7535\u5f71\u8bc4\u8bba\u7528\u4e8e\u8bad\u7ec3":88,"000\u4e2a\u7528\u4e8e\u6d4b\u8bd5":88,"000\u4e2atxt\u6587\u4ef6":88,"000\u4f4d\u7528\u6237\u5bf94":85,"000\u5e45\u56fe\u50cf\u4e0a\u6d4b\u8bd5\u4e86\u6a21\u578b\u7684\u5206\u7c7b\u9519\u8bef\u7387":82,"000\u5f20\u7070\u5ea6\u56fe\u7247\u7684\u6570\u5b57\u5206\u7c7b\u6570\u636e\u96c6":2,"000\u6761\u8bc4\u4ef7":85,"000\u90e8\u7535\u5f71\u76841":85,"00186201e":4,"00m":67,"01852v1":8,"03m":67,"0424m":67,"0473v3":9,"05d":81,"0630u":67,"06u":67,"0810u":67,"08823112e":4,"0957m":67,"0\u4e4b\u540e\u7684\u7248\u672c":49,"0\u53f7\u8bad\u7ec3\u8282\u70b9\u662f\u4e3b\u8bad\u7ec3\u8282\u70b9":70,"0\u5c42\u5e8f\u5217":55,"0\u8868\u793a\u7b2c\u4e00\u6b21\u7ecf\u8fc7\u8bad\u7ec3\u96c6":88,"0ab":8,"0rc1":42,"0rc2":42,"10000\u5f20\u4f5c\u4e3a\u6d4b\u8bd5\u96c6":81,"10007_10":88,"10014_7":88,"100m":46,"101\u5c42\u548c152\u5c42\u7684\u7f51\u7edc\u7ed3\u6784\u4e2d":82,"101\u5c42\u548c152\u5c42\u7684\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6\u53ef\u53c2\u7167":82,"101\u5c42\u7f51\u7edc\u6a21\u578b":82,"10\u4e2d\u7684\u8bad\u7ec3\u96c6\u548c\u6d4b\u8bd5\u96c6":81,"10\u6570\u636e\u96c6":81,"10\u6570\u636e\u96c6\u5305\u542b60000\u5f2032x32\u7684\u5f69\u8272\u56fe\u7247":81,"10\u6570\u636e\u96c6\u7684\u5b98\u65b9\u7f51\u5740":81,"10\u6570\u636e\u96c6\u8bad\u7ec3\u4e00\u4e2a\u5377\u79ef\u795e\u7ecf\u7f51\u7edc":81,"10g":27,"1150u":67,"11\u5b9e\u73b0\u4e86c":34,"11e6":74,"12194102e":4,"124n":67,"128\u7ef4\u548c256\u7ef4":80,"12\u64cd\u4f5c\u7cfb\u7edf":46,"13m":74,"1490u":67,"14\u4ee5\u4e0a":54,"14\u6570\u636e\u96c6":89,"14\u6570\u636e\u96c6\u4e0a\u5f97\u5230\u826f\u597d\u8868\u73b0\u7684\u8bad\u7ec3\u8fc7\u7a0b":89,"14\u8fd9\u79cd\u5199\u6cd5\u5c06\u4f1a\u6d4b\u8bd5\u6a21\u578b":72,"152\u5c42\u7f51\u7edc\u6a21\u578b":82,"15501715e":4,"1550u":67,"15\u884c":56,"1636k":89,"16\u5b57\u8282\u8868\u793a\u4fdd\u5b58\u7684\u53c2\u6570\u603b\u4e2a\u6570":46,"16u":67,"173m":82,"173n":67,"1770u":67,"18\u5c81\u4ee5\u4e0b":85,"18e457ce3d362ff5f3febf8e7f85ffec852f70f3b629add10aed84f930a68750":74,"197u":67,"1\u7684\u5c42\u4e4b\u5916":72,"1\u7a00\u758f\u6570\u636e":62,"1\u8f6e\u5b58\u50a8\u7684\u6240\u6709\u6a21\u578b":72,"1\u9664\u4ee5batch":81,"1m\u6570\u636e\u96c6\u4e2d":86,"1m\u7684\u5b57\u6bb5\u914d\u7f6e\u6587\u4ef6\u5728\u76ee\u5f55":86,"200\u6570\u636e\u96c6\u4e0a\u4f7f\u7528vgg\u6a21\u578b\u8bad\u7ec3\u4e00\u4e2a\u9e1f\u7c7b\u56fe\u7247\u5206\u7c7b\u6a21\u578b":81,"210u":67,"211839e770f7b538e2d8":9,"215n":67,"228u":67,"234m":82,"24\u5c81":85,"2520u":67,"25639710e":4,"25k":84,"2680u":67,"26\u884c":56,"27787406e":4,"279n":67,"27m":67,"285m":67,"2863m":67,"28\u7684\u56fe\u7247\u50cf\u7d20\u7070\u5ea6\u503c":2,"28\u7ef4\u7684\u7a20\u5bc6\u6d6e\u70b9\u6570\u5411\u91cf\u548c\u4e00\u4e2a":2,"28m":67,"2977m":67,"29997\u4e2a\u6700\u9ad8\u9891\u5355\u8bcd\u548c3\u4e2a\u7279\u6b8a\u7b26\u53f7":89,"2\u4e09\u7c7b\u7684\u6bd4\u4f8b\u4e3a":46,"2\u4e2a\u6d6e\u70b9\u6570":47,"2\u5206\u522b\u4ee3\u88683\u4e2a\u8282\u70b9\u7684trainer":75,"2\u610f\u5473\u77400\u53f7\u548c1\u53f7gpu\u5c06\u4f1a\u4f7f\u7528\u6570\u636e\u5e76\u884c\u6765\u8ba1\u7b97fc1\u548cfc2\u5c42":72,"2\u8fd9\u51e0\u4e2a\u76ee\u5f55\u8868\u793apaddlepaddle\u8282\u70b9\u4e0etrain":75,"2nd":8,"302n":67,"30u":67,"3206326\u4e2a\u8bcd\u548c4\u4e2a\u7279\u6b8a\u6807\u8bb0":80,"32777140e":4,"328n":67,"32\u7ef4":80,"32u":67,"32x32":12,"331n":67,"3320u":67,"34\u5c81":85,"35\u65f6":89,"36540484e":4,"36u":67,"3710m":67,"3768m":67,"387u":67,"38u":67,"3920u":67,"39u":67,"3\u4e2a\u7279\u6b8a\u7b26\u53f7":89,"3\u53f7gpu":46,"4035m":67,"4090u":67,"4096mb":70,"4279m":67,"43630644e":4,"43u":67,"448a5b355b84":74,"44\u5c81":85,"4560u":67,"4563m":67,"45u":67,"4650u":67,"4726m":67,"473m":74,"48565123e":4,"48684503e":4,"49316648e":4,"49\u5c81":85,"4\u5b57\u8282\u8868\u793apaddlepaddle\u7248\u672c\u4fe1\u606f":46,"4gb":70,"4yf":8,"500\u4e2atxt\u6587\u4ef6":88,"500m":46,"50\u5c42":82,"50\u5c42\u7f51\u7edc\u6a21\u578b":82,"51111044e":4,"514u":67,"525n":67,"526u":67,"53018653e":4,"536u":67,"5460u":67,"5470u":67,"54u":67,"55\u5c81":85,"55g":89,"5690m":67,"573u":67,"578n":67,"5798m":67,"586u":67,"58s":74,"5969m":67,"5\u4e2a\u6d4b\u8bd5\u6837\u4f8b\u548c2\u4e2a\u751f\u6210\u5f0f\u6837\u4f8b":80,"5delta":8,"6080u":67,"6082v4":8,"6140u":67,"6305m":67,"639u":67,"64\u5e73\u53f0\u4e3a\u4f8b":[53,54],"64\u73af\u5883":54,"64\u7ef4":80,"655u":67,"6780u":67,"6810u":67,"682u":67,"6970u":67,"6\u4e07\u4ebf\u6b21\u6d6e\u70b9\u8fd0\u7b97\u6bcf\u79d2":67,"6\u4e2a\u8282\u70b9":68,"6\u5143\u4e0a\u4e0b\u6587\u4f5c\u4e3a\u8f93\u5165\u5c42":80,"704u":67,"70634608e":4,"7090u":67,"72296313e":4,"72u":67,"73u":67,"75u":67,"760u":67,"767u":67,"783n":67,"784u":67,"78m":67,"7\u548cpip":46,"7\u7248\u672c\u5f00\u59cb":53,"7eamaa":12,"7kb":74,"8250u":67,"8300u":67,"830n":67,"849m":67,"85625684e":4,"861u":67,"864k":89,"8661m":67,"877\u4e2a\u88ab\u5411\u91cf\u5316\u7684\u8bcd":80,"877\u884c":80,"892m":67,"8\u5b57\u8282\u8868\u793a\u6bcf\u4e2a\u53c2\u6570\u5360\u7528\u7684\u5b57\u8282\u6570":46,"901n":67,"90u":67,"918u":67,"9247m":67,"924n":67,"9261m":67,"93137714e":4,"9330m":67,"94u":67,"9530m":67,"96644767e":4,"983m":67,"988u":67,"997u":67,"99982715e":4,"99m":82,"99u":67,"9\u4e2d\u7684\u4e00\u4e2a\u6570\u5b57":2,"9f18":74,"\u0233":47,"\u03b5":47,"\u4e00":56,"\u4e00\u4e2a0\u5c42\u5e8f\u5217":55,"\u4e00\u4e2a0\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u6269\u5c55\u6210\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":55,"\u4e00\u4e2a\u4e0d\u5171\u4eab\u7684\u4f8b\u5b50\u662f":63,"\u4e00\u4e2a\u4f8b\u5b50\u662f\u623f\u4ea7\u4f30\u503c":47,"\u4e00\u4e2a\u5178\u578b\u7684\u795e\u7ecf\u7f51\u7edc\u5982\u4e0b\u56fe\u6240\u793a":81,"\u4e00\u4e2a\u5178\u578b\u7684chunk\u5982\u4e0b\u6240\u793a":28,"\u4e00\u4e2a\u5206\u5e03\u5f0f\u7684\u5b58\u50a8\u7cfb\u7edf":73,"\u4e00\u4e2a\u5206\u5e03\u5f0fpaddle\u8bad\u7ec3\u4efb\u52a1\u4e2d\u7684\u6bcf\u4e2a\u8fdb\u7a0b\u90fd\u53ef\u4ee5\u4ececeph\u8bfb\u53d6\u6570\u636e":74,"\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":55,"\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217\u6216\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":55,"\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u6269\u5c55\u6210\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":55,"\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217\u8fdb\u5165":58,"\u4e00\u4e2a\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u5305\u542b\u5982\u4e0b\u5c42":81,"\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":55,"\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217\u6216\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":55,"\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u53d8\u6210\u4e00\u4e2a0\u5c42\u5e8f\u5217":55,"\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u53d8\u6210\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":55,"\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217\u8fdb\u5165":58,"\u4e00\u4e2a\u53cc\u5c42rnn\u7531\u591a\u4e2a\u5355\u5c42rnn\u7ec4\u6210":58,"\u4e00\u4e2a\u53ef\u8c03\u7528\u7684\u51fd\u6570":58,"\u4e00\u4e2a\u57fa\u672c\u7684\u5e94\u7528\u573a\u666f\u662f\u533a\u5206\u7ed9\u5b9a\u6587\u672c\u7684\u8912\u8d2c\u4e24\u6781\u6027":88,"\u4e00\u4e2a\u5bb9\u5668\u5c31\u662f\u4e00\u4e2a\u64cd\u4f5c\u7cfb\u7edf\u7684\u8fdb\u7a0b":49,"\u4e00\u4e2a\u6216\u591a\u4e2a":73,"\u4e00\u4e2a\u6570\u636e\u96c6\u5927\u90e8\u5206\u5e8f\u5217\u957f\u5ea6\u662f100":46,"\u4e00\u4e2a\u6587\u4ef6":2,"\u4e00\u4e2a\u662f\u6d6e\u70b9\u8ba1\u7b97\u91cf":67,"\u4e00\u4e2a\u662f\u76f4\u63a5\u628a\u4e00\u4e2a\u5bb9\u5668\u8f6c\u6362\u6210\u955c\u50cf":49,"\u4e00\u4e2a\u72ec\u7acb\u7684\u5143\u7d20":55,"\u4e00\u4e2a\u72ec\u7acb\u7684\u8bcd\u8bed":55,"\u4e00\u4e2a\u795e\u7ecf\u7f51\u7edc\u7684\u6a21\u578b\u7531\u5927\u91cf\u7684\u53c2\u6570\u7ec4\u6210":20,"\u4e00\u4e2a\u7b80\u5355\u7684\u4f8b\u5b50\u5982":88,"\u4e00\u4e2a\u7ec8\u7aef\u8fd0\u884cvi":49,"\u4e00\u4e2a\u7f51\u7edc\u5c42\u7684\u524d\u5411\u4f20\u64ad\u90e8\u5206\u628a\u8f93\u5165\u8f6c\u5316\u4e3a\u76f8\u5e94\u7684\u8f93\u51fa":62,"\u4e00\u4e2a\u7f51\u7edc\u5c42\u7684\u53c2\u6570\u662f\u5728":62,"\u4e00\u4e2a\u7f51\u7edc\u5c42\u7684c":62,"\u4e00\u4e2a\u8f93\u51fa\u7ec4\u6210":63,"\u4e00\u4e2a\u91cd\u8981\u7684\u95ee\u9898\u662f\u9009\u62e9\u6b63\u786e\u7684learning_r":46,"\u4e00\u4e2achunk\u7531\u6240\u5728\u7684\u6587\u4ef6\u504f\u79fb":28,"\u4e00\u4e2adocker\u955c\u50cf\u662f\u4e00\u4e2a\u6253\u5305\u597d\u7684\u8f6f\u4ef6":49,"\u4e00\u4e2agpu\u8bbe\u5907\u4e0a\u4e0d\u5141\u8bb8\u914d\u7f6e\u591a\u4e2a\u6a21\u578b":70,"\u4e00\u4e2alabel":56,"\u4e00\u4e2alogging\u5bf9\u8c61":2,"\u4e00\u4e2amemory\u5305\u542b":59,"\u4e00\u4e2apass\u610f\u5473\u7740paddlepaddle\u8bad\u7ec3\u6570\u636e\u96c6\u4e2d\u7684\u6240\u6709\u6837\u672c\u88ab\u904d\u5386\u4e00\u6b21":87,"\u4e00\u4e2apass\u8868\u793a\u8fc7\u4e00\u904d\u6240\u6709\u8bad\u7ec3\u6837\u672c":84,"\u4e00\u4e2apod\u4e2d\u7684\u6240\u6709\u5bb9\u5668\u4f1a\u88ab\u8c03\u5ea6\u5230\u540c\u4e00\u4e2anode\u4e0a":73,"\u4e00\u4e2aposix\u517c\u5bb9\u7684\u6587\u4ef6\u7cfb\u7edf":28,"\u4e00\u4eba":56,"\u4e00\u53e5\u8bdd\u662f\u7531\u8bcd\u8bed\u6784\u6210\u7684\u5e8f\u5217":58,"\u4e00\u53f0\u673a\u5668\u4e0a\u9762\u7684\u7ebf\u7a0b\u6570\u91cf":86,"\u4e00\u53f0\u7535\u8111":60,"\u4e00\u65e9":56,"\u4e00\u662fbatch":46,"\u4e00\u6761\u6837\u672c":2,"\u4e00\u6837\u7684\u65b9\u5f0f":60,"\u4e00\u6837\u8bbe\u7f6e":68,"\u4e00\u6b21\u4f5c\u4e1a\u79f0\u4e3a\u4e00\u4e2ajob":73,"\u4e00\u6b21\u6027\u676f\u5b50":56,"\u4e00\u6b21yield\u8c03\u7528":2,"\u4e00\u79cd\u5e38\u7528\u7684\u505a\u6cd5\u662f\u7528\u5b66\u4e60\u7684\u6a21\u578b\u5bf9\u53e6\u5916\u4e00\u7ec4\u6d4b\u8bd5\u6570\u636e\u8fdb\u884c\u9884\u6d4b":47,"\u4e00\u79cd\u5e38\u7528\u7684cmake\u914d\u7f6e\u5982\u4e0b":53,"\u4e00\u7bc7\u8bba\u6587":89,"\u4e00\u7ea7\u76ee\u5f55":[88,89],"\u4e00\u81f4":[55,56],"\u4e00\u822c\u4e0d\u5141\u8bb8\u518d\u4ece":42,"\u4e00\u822c\u4ece":61,"\u4e00\u822c\u5728paddlepaddle\u4e2d":56,"\u4e00\u822c\u60c5\u51b5\u4e0b":[1,47],"\u4e00\u822c\u63a8\u8350\u8bbe\u7f6e\u6210true":2,"\u4e00\u822c\u662f\u7531\u4e8e\u76f4\u63a5\u4f20\u9012\u5927\u5b57\u5178\u5bfc\u81f4\u7684":46,"\u4e00\u822c\u6765\u8bf4":59,"\u4e00\u822c\u800c\u8a00":89,"\u4e00\u822c\u8868\u793a":56,"\u4e00\u822c\u8bbe\u7f6e":46,"\u4e00\u884c\u4e3a\u4e00\u4e2a\u6837\u672c":84,"\u4e09\u79cd\u5e8f\u5217\u6a21\u5f0f":[2,51],"\u4e09\u7ea7\u76ee\u5f55":[88,89],"\u4e0a":61,"\u4e0a\u4e0b\u6587\u5927\u5c0f\u8bbe\u7f6e\u4e3a1\u7684\u4e00\u4e2a\u6837\u672c\u7684\u7279\u5f81\u5982\u4e0b":87,"\u4e0a\u4f20\u5230cloud\u6216\u8005\u4e0b\u8f7d\u5230\u672c\u5730\u7684\u65f6\u95f4\u53ef\u80fd\u6bd4\u8f83\u957f":28,"\u4e0a\u4f20\u65b9\u6cd5":42,"\u4e0a\u53d1\u8868\u7684\u8bc4\u8bba\u5206\u6210\u6b63\u9762\u8bc4\u8bba\u548c\u8d1f\u9762\u8bc4\u8bba\u4e24\u7c7b":88,"\u4e0a\u56fe\u4e2d\u865a\u7ebf\u7684\u8fde\u63a5":56,"\u4e0a\u56fe\u63cf\u8ff0\u4e86\u4e00\u4e2a3\u8282\u70b9\u7684\u5206\u5e03\u5f0f\u8bad\u7ec3\u573a\u666f":75,"\u4e0a\u6ce8\u518c\u4e00\u4e0b":28,"\u4e0a\u7f16\u8bd1\u5f88\u6162":60,"\u4e0a\u7f16\u8bd1android\u5e73\u53f0\u4e0a\u9002\u7528\u7684paddlepaddle\u5e93":53,"\u4e0a\u7f51":56,"\u4e0a\u8fd0\u884c":49,"\u4e0a\u8ff0\u4ee3\u7801\u5c06bias\u5168\u90e8\u521d\u59cb\u5316\u4e3a1":46,"\u4e0a\u8ff0\u547d\u4ee4\u7f16\u8bd1\u51fa\u4e00\u4e2a":60,"\u4e0a\u8ff0\u7684":46,"\u4e0a\u8ff0\u7684\u4ee3\u7801\u7247\u6bb5\u5305\u542b\u4e86\u4e24\u79cd\u65b9\u6cd5":67,"\u4e0a\u8ff0\u811a\u672c\u4f7f\u7528":68,"\u4e0a\u9762\u7684\u4ee3\u7801\u5728":63,"\u4e0a\u9762\u7684\u4ee3\u7801\u9996\u5148\u5bfc\u5165\u4f9d\u8d56\u7684\u5305":63,"\u4e0b":[63,81],"\u4e0b\u540c":46,"\u4e0b\u56fe\u4e2d\u5c31\u5c55\u793a\u4e86\u4e00\u4e9b\u5173\u4e8e\u5185\u5b58\u6570\u636e\u8fc1\u5f99\u548c\u8ba1\u7b97\u8d44\u6e90\u5229\u7528\u7387\u7684\u5efa\u8bae":67,"\u4e0b\u56fe\u5c55\u793a\u4e86\u6240\u6709\u7684\u56fe\u7247\u7c7b\u522b":81,"\u4e0b\u56fe\u5c55\u793a\u4e86\u65f6\u95f4\u6269\u5c55\u76842\u5c42":87,"\u4e0b\u56fe\u5c55\u793a\u7684\u662f\u57fa\u4e8e\u6b8b\u5dee\u7684\u8fde\u63a5\u65b9\u5f0f":82,"\u4e0b\u56fe\u662f\u4e00\u4e2a\u5168\u8fde\u63a5\u5c42\u7684\u793a\u610f\u56fe":62,"\u4e0b\u5b58\u653e\u516c\u5171\u6570\u636e\u96c6\u5408":21,"\u4e0b\u6587\u4ee5nlp\u4efb\u52a1\u4e3a\u4f8b":58,"\u4e0b\u6587\u4f7f\u7528":75,"\u4e0b\u6587\u5c31\u662f\u7528job\u7c7b\u578b\u7684\u8d44\u6e90\u6765\u8fdb\u884c\u8bad\u7ec3":74,"\u4e0b\u6b21":56,"\u4e0b\u7684":75,"\u4e0b\u7684\u4f5c\u4e3a\u7f16\u8bd1\u5de5\u5177":54,"\u4e0b\u8868\u5c55\u793a\u4e86batch":82,"\u4e0b\u8f7d":28,"\u4e0b\u8f7d\u5230\u672c\u5730":28,"\u4e0b\u8f7d\u5b8c\u6570\u636e\u540e":74,"\u4e0b\u8f7d\u6570\u636e\u96c6":81,"\u4e0b\u8f7draspberri":54,"\u4e0b\u8f7dwmt":89,"\u4e0b\u9762\u4e3e\u4e2a\u7b80\u5355\u7684\u4f8b\u5b50":67,"\u4e0b\u9762\u4ecb\u7ecd\u4ecb\u7ecd":63,"\u4e0b\u9762\u4ecb\u7ecd\u9884\u5904\u7406\u8fc7\u7a0b\u5177\u4f53\u7684\u6b65\u9aa4":86,"\u4e0b\u9762\u4ee5\u77e9\u9635\u4e58\u64cd\u4f5c":63,"\u4e0b\u9762\u4ee5addop\u4e3a\u4f8b\u8bf4\u660etensor\u7684\u4f7f\u7528\u8fc7\u7a0b":64,"\u4e0b\u9762\u5148\u7b80\u8981\u4ecb\u7ecd\u4e00\u4e0b\u672c\u6587\u7528\u5230\u7684\u51e0\u4e2akubernetes\u6982\u5ff5":73,"\u4e0b\u9762\u5206\u522b\u4ecb\u7ecd\u67d0\u4e00\u7c7b\u6587\u4ef6\u7684\u5b9e\u73b0\u65b9\u5f0f":34,"\u4e0b\u9762\u5217\u51fa\u4e86":59,"\u4e0b\u9762\u5217\u51fa\u4e86\u5168\u8fde\u63a5\u5c42\u7684\u68af\u5ea6\u68c0\u67e5\u5355\u5143\u6d4b\u8bd5":62,"\u4e0b\u9762\u5c31\u6839\u636e\u8fd9\u51e0\u4e2a\u6b65\u9aa4\u5206\u522b\u4ecb\u7ecd":75,"\u4e0b\u9762\u6211\u4eec\u7ed9\u51fa\u4e86\u4e00\u4e2a\u914d\u7f6e\u793a\u4f8b":81,"\u4e0b\u9762\u662f":63,"\u4e0b\u9762\u662f\u4e00\u4e2a\u8bef\u5dee\u66f2\u7ebf\u56fe\u7684\u793a\u4f8b":81,"\u4e0b\u9762\u662f\u5bf9":63,"\u4e0b\u9762\u662fcifar":81,"\u4e0b\u9762\u7684\u4ee3\u7801\u5c06\u968f\u673a\u751f\u6210\u7684\u77e9\u9635\u8f6c\u5316\u4e3a\u53ef\u4ee5\u88abpaddlepaddle\u52a0\u8f7d\u7684\u6a21\u578b\u53c2\u6570":46,"\u4e0b\u9762\u7684\u4ee3\u7801\u7247\u6bb5\u5b9e\u73b0\u4e86":62,"\u4e0b\u9762\u7684\u4f8b\u5b50\u4f7f\u7528\u4e86":82,"\u4e0b\u9762\u7684\u4f8b\u5b50\u540c\u6837\u4f7f\u7528\u4e86":82,"\u4e0b\u9762\u7684\u547d\u4ee4\u628a\u5f53\u524d\u76ee\u5f55\u6302\u8f7d\u5230\u4e86\u5bb9\u5668\u4e2d\u7684":49,"\u4e0b\u9762\u7684\u70b9\u5b9e\u73b0\u4e86mulop\u7684\u5b9a\u4e49":63,"\u4e0b\u9762\u7ed9\u51fa\u4e86\u4e00\u4e2a\u4f8b\u5b50":62,"\u4e0b\u9762\u7ed9\u51fa\u5728\u4e09\u7ef4\u7a7a\u95f4\u4e2d\u4f7f\u7528\u7ebf\u6027\u56de\u5f52\u62df\u5408\u4e00\u6761\u76f4\u7ebf\u7684\u4f8b\u5b50":51,"\u4e0b\u9762\u811a\u672c\u7b26\u5408paddlepaddle\u671f\u5f85\u7684\u8bfb\u53d6\u6570\u636e\u7684python\u7a0b\u5e8f\u7684\u6a21\u5f0f":47,"\u4e0b\u9762\u89e3\u91ca\u4ee3\u7801\u4e2d\u4e00\u4e9b\u5173\u952e\u7684\u5730\u65b9":63,"\u4e0b\u9762\u8fd9\u4e9blayer\u80fd\u591f\u63a5\u53d7\u53cc\u5c42\u5e8f\u5217\u4f5c\u4e3a\u8f93\u5165":55,"\u4e0d":56,"\u4e0d\u4e00\u5b9a\u548c\u65f6\u95f4\u6709\u5173\u7cfb":2,"\u4e0d\u4e00\u81f4\u7684\u7531pfsclient\u4e0b\u8f7d\u6216\u8005\u4f20\u8f93chunk\u5b8c\u6210":28,"\u4e0d\u4f1a\u4fdd\u7559\u5728\u78c1\u76d8\u4e0a":60,"\u4e0d\u4f1a\u518d\u4ece":46,"\u4e0d\u4f1a\u865a\u62df\u4efb\u4f55\u786c\u4ef6":60,"\u4e0d\u4f7f\u7528\u9759\u6001\u5e93":33,"\u4e0d\u4f7f\u7528\u989d\u5916\u7a7a\u95f4":62,"\u4e0d\u4f7f\u7528c":33,"\u4e0d\u4f7f\u7528swig":33,"\u4e0d\u5141\u8bb8\u4e00\u4e2a\u6587\u4ef6\u4e2d\u5305\u542b\u591a\u4e2aop":63,"\u4e0d\u5171\u4eab\u5219\u4e0d\u52a0":63,"\u4e0d\u5171\u4eab\u7684\u4f8b\u5b50\u53ef\u4ee5\u53c2\u8003":63,"\u4e0d\u5305\u542b\u5728\u5b57\u5178\u4e2d\u7684\u5355\u8bcd":89,"\u4e0d\u540c":87,"\u4e0d\u540c\u4e3b\u673a":73,"\u4e0d\u540c\u4e8e\u4e0a\u8ff0\u4ecb\u7ecd\u7684recurr":46,"\u4e0d\u540c\u4e8eop\u7684\u7f16\u8bd1\u6d4b\u8bd5":63,"\u4e0d\u540c\u4ea7\u54c1":88,"\u4e0d\u540c\u53c2\u6570\u670d\u52a1\u5668\u4e0a\u6570\u636e\u5927\u5c0f\u7684\u6700\u5927\u503c\u4e0e\u6700\u5c0f\u503c\u7684\u6bd4\u7387":70,"\u4e0d\u540c\u5c42\u7684\u7279\u5f81\u7531\u5206\u53f7":82,"\u4e0d\u540c\u65f6\u95f4\u6b65\u7684\u8f93\u5165\u662f\u4e0d\u540c\u7684":59,"\u4e0d\u540c\u7248\u672c\u7684\u7f16\u8bd1\u5668\u4e4b\u95f4":33,"\u4e0d\u540c\u7684\u4f18\u5316\u7b97\u6cd5\u9700\u8981\u4f7f\u7528\u4e0d\u540c\u5927\u5c0f\u7684\u5185\u5b58":46,"\u4e0d\u540c\u7684\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf":75,"\u4e0d\u540c\u7684\u6570\u636e\u7c7b\u578b\u548c\u5e8f\u5217\u6a21\u5f0f\u8fd4\u56de\u7684\u683c\u5f0f\u4e0d\u540c":[2,51],"\u4e0d\u540c\u7a7a\u95f4\u7684\u8d44\u6e90\u540d\u53ef\u4ee5\u91cd\u590d":73,"\u4e0d\u540c\u8bbe\u5907":63,"\u4e0d\u540c\u8bed\u8a00\u7684\u63a5\u53e3\u9002\u5e94\u4e0d\u540c\u8bed\u8a00\u7684\u7279\u6027":33,"\u4e0d\u540c\u8f93\u5165\u542b\u6709\u7684\u5b50\u53e5":58,"\u4e0d\u540c\u8f93\u5165\u5e8f\u5217\u542b\u6709\u7684\u8bcd\u8bed\u6570\u5fc5\u987b\u4e25\u683c\u76f8\u7b49":58,"\u4e0d\u540cdataprovider\u5bf9\u6bd4\u5982\u4e0b":56,"\u4e0d\u540cpod\u4e4b\u95f4\u53ef\u4ee5\u901a\u8fc7ip\u5730\u5740\u8bbf\u95ee":73,"\u4e0d\u540crank\u7684tensor\u662f\u4e0d\u540c\u7c7b\u578b":64,"\u4e0d\u5728":34,"\u4e0d\u5bb9\u6613\u51fa\u9519":28,"\u4e0d\u5c11":56,"\u4e0d\u5d4c\u5165\u5176\u4ed6\u8bed\u8a00\u89e3\u91ca\u5668":33,"\u4e0d\u5d4c\u5165python\u89e3\u91ca\u5668":33,"\u4e0d\u5e94\u8be5\u88ab\u62c6\u89e3":58,"\u4e0d\u5fc5\u518d\u5c06\u4efb\u610f\u957f\u5ea6\u6e90\u8bed\u53e5\u4e2d\u7684\u6240\u6709\u4fe1\u606f\u538b\u7f29\u81f3\u4e00\u4e2a\u5b9a\u957f\u7684\u5411\u91cf\u4e2d":89,"\u4e0d\u6307\u5b9a\u65f6":58,"\u4e0d\u63d0\u4f9b\u5206\u5e03\u5f0f\u5b58\u50a8":73,"\u4e0d\u652f\u6301":87,"\u4e0d\u662f\u4e00\u6761\u5e8f\u5217":[2,51],"\u4e0d\u662f\u771f\u6b63\u7684layer":46,"\u4e0d\u662f\u901a\u8fc7\u4e00\u822c\u7684\u65b9\u5f0f\u6765\u5b9e\u73b0\u5bf9\u8f93\u51fa\u7684\u6fc0\u6d3b":46,"\u4e0d\u663e\u793a\u7684\u5199\u6bcf\u4e2a\u7c7b\u5177\u4f53\u5305\u542b\u4ec0\u4e48":33,"\u4e0d\u6ee1\u8db3\u94a9\u5b50\u7684":61,"\u4e0d\u7528mount\u7684\u65b9\u5f0f\u6765\u8bbf\u95ee\u6570\u636e":21,"\u4e0d\u7f13\u5b58\u4efb\u4f55\u6570\u636e":2,"\u4e0d\u80fd\u4fee\u6539op\u7684\u6210\u5458\u53d8\u91cf":63,"\u4e0d\u80fd\u592a\u968f\u610f":61,"\u4e0d\u80fd\u88ab\u63d0\u4ea4\u5230":61,"\u4e0d\u8fc7":56,"\u4e0d\u8fc7\u5b9e\u9645\u4e0a\u662f\u8fd0\u884c\u5728\u4e00\u4e2a":60,"\u4e0d\u8fdc":56,"\u4e0d\u9519":56,"\u4e0d\u9700\u8981\u4f9d\u8d56\u5176\u4ed6\u4efb\u4f55\u8f6f\u4ef6\u4e86":60,"\u4e0e":[63,75,80,89],"\u4e0e\u4e4b\u76f8\u5bf9\u7684\u662flocal":28,"\u4e0e\u529f\u80fd\u5206\u652f\u4e0d\u540c\u7684\u662f":42,"\u4e0e\u5355\u5c42rnn\u7684\u914d\u7f6e\u7c7b\u4f3c":56,"\u4e0e\u53ef\u80fd\u6709\u7684":42,"\u4e0e\u5728":87,"\u4e0e\u5f53\u524d\u7684\u8870\u51cf\u56e0\u5b50\u7684\u4e58\u79ef":46,"\u4e0e\u672c\u5730\u8bad\u7ec3\u76f8\u540c":68,"\u4e0e\u6b64\u4e0d\u540c\u7684\u662f":75,"\u4e0e\u7ffb\u8bd1":89,"\u4e0e\u8bad\u7ec3\u4e0d\u540c":88,"\u4e0e\u8bad\u7ec3\u6a21\u578b\u4e0d\u540c\u7684\u662f":89,"\u4e0e\u8fd9\u4e2a\u8bad\u7ec3\u6570\u636e\u4ea4\u4e92\u7684layer":46,"\u4e0eimdb\u7f51\u7ad9\u63d0\u4f9b\u7684\u4e00\u81f4":85,"\u4e0ejob":75,"\u4e0eoperator\u524d\u5411\u8ba1\u7b97\u7684\u8f93\u51fa\u8fdb\u884c\u5bf9\u6bd4":63,"\u4e0eoperator\u6ce8\u518c\u65f6\u6ce8\u518c\u7684\u7c7b\u578b\u4e00\u81f4":63,"\u4e0etime":85,"\u4e14":56,"\u4e14\u4e0d\u6392\u9664commit\u4e4b\u95f4\u7684\u4fee\u6539\u5b58\u5728\u76f8\u4e92\u8986\u76d6\u7684\u60c5\u51b5":61,"\u4e14\u589e\u52a0\u4e00\u4e2a\u7b2c\u4e09\u65b9\u8bed\u8a00":33,"\u4e14\u5e8f\u5217\u7684\u6bcf\u4e00\u4e2a\u5143\u7d20\u8fd8\u662f\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217":[2,51],"\u4e14\u652f\u6301\u90e8\u7f72\u5230":73,"\u4e14\u6bcf\u4e2a\u53e5\u5b50\u8868\u793a\u4e3a\u5bf9\u5e94\u7684\u8bcd\u8868\u7d22\u5f15\u6570\u7ec4":56,"\u4e14\u8c03\u7528\u65f6\u4e0d\u80fd\u629b\u51fa\u5f02\u5e38\u6216\u51fa\u73b0\u8fd0\u884c\u65f6\u9519\u8bef":34,"\u4e14\u9ed8\u8ba4\u5728\u8bad\u7ec3\u96c6\u4e0a\u6784\u5efa\u5b57\u5178":88,"\u4e14c99\u652f\u6301bool\u7c7b\u578b\u548c\u5b9a\u957f\u6574\u6570":33,"\u4e14c99\u76f8\u5bf9\u4e8ec11\u4f7f\u7528\u66f4\u52a0\u5e7f\u6cdb":33,"\u4e24":56,"\u4e24\u4e2a\u5217\u8868\u6587\u4ef6":68,"\u4e24\u4e2a\u5b50\u76ee\u5f55\u4e0b":65,"\u4e24\u4e2a\u5d4c\u5957\u7684":58,"\u4e24\u4e2a\u64cd\u4f5c":67,"\u4e24\u4e2a\u6587\u4ef6\u5939":81,"\u4e24\u4e2a\u8f93\u5165\u7279\u5f81\u5728\u8fd9\u4e2a\u6d41\u7a0b\u4e2d\u8d77\u7740\u81f3\u5173\u91cd\u8981\u7684\u4f5c\u7528":87,"\u4e24\u4e2a\u8f93\u5165\u7684\u5b50\u5e8f\u5217\u957f\u5ea6\u4e5f\u5e76\u4e0d\u76f8\u540c":56,"\u4e24\u4e2a\u90e8\u5206":65,"\u4e24\u79cd\u65b9\u6cd5\u7684\u533a\u522b":46,"\u4e24\u79cd\u7c7b\u522b":84,"\u4e24\u8005\u5747\u4e3a\u7eaf\u6587\u672c\u6587\u4ef6":1,"\u4e24\u8005\u90fd\u662f\u5bf9\u68af\u5ea6\u7684\u622a\u65ad":46,"\u4e25\u683c\u7684\u547d\u540d\u89c4\u8303pep":42,"\u4e2a":84,"\u4e2a\u5185\u5b58\u6c60\u5b9e\u9645\u4e0a\u51b3\u5b9a\u4e86shuffle\u7684\u7c92\u5ea6":46,"\u4e2a\u5355\u8bcd":89,"\u4e2a\u6027\u5316\u63a8\u8350":[42,83],"\u4e2a\u6279\u6b21\u540e\u6253\u5370\u4e00\u4e2a":86,"\u4e2a\u6279\u6b21\u7684\u53c2\u6570\u5e73\u5747\u503c\u8fdb\u884c\u6d4b\u8bd5":70,"\u4e2a\u6a21\u578b\u6d4b\u8bd5\u6570\u636e":70,"\u4e2d":[33,34,46,62,63,64,75,81,84,86,87,88],"\u4e2d\u4e0d\u8981\u6dfb\u52a0\u5927\u6587\u4ef6\u7b49":61,"\u4e2d\u4ecb\u7ecd\u7684\u65b9\u6cd5":80,"\u4e2d\u4efb\u610f\u7b2ci\u884c\u7684\u53e5\u5b50\u4e4b\u95f4\u90fd\u5fc5\u987b\u6709\u7740\u4e00\u4e00\u5bf9\u5e94\u7684\u5173\u7cfb":89,"\u4e2d\u4efb\u610f\u7b2ci\u884c\u7684\u53e5\u5b50\u4e4b\u95f4\u90fd\u6709\u7740\u4e00\u4e00\u5bf9\u5e94\u7684\u5173\u7cfb":89,"\u4e2d\u5143\u7d20\u7684\u4e2a\u6570\u7b49\u4e8e\u7f51\u7edc\u4e2d\u8f93\u51fa\u5c42\u7684\u4e2a\u6570":46,"\u4e2d\u5173\u4e8e\u65f6\u95f4\u9012\u5f52\u795e\u7ecf\u7f51\u7edc\u7684\u4ecb\u7ecd":56,"\u4e2d\u5199\u5165json\u5185\u5bb9":20,"\u4e2d\u5305\u542b\u4e00\u4e2aandroid\u7248\u672c\u7684\u5e93":53,"\u4e2d\u5305\u542b\u4e00\u4e2araspberri":54,"\u4e2d\u5305\u542b\u4e86\u8bad\u7ec3\u6a21\u578b\u7684\u57fa\u672c\u547d\u4ee4":84,"\u4e2d\u5305\u542b\u5982\u4e0b\u8868\u6240\u793a\u76843\u4e2a\u6587\u4ef6\u5939":89,"\u4e2d\u5305\u542bc":[53,54],"\u4e2d\u5355\u5143\u6d4b\u8bd5\u7684\u4e00\u90e8\u5206":61,"\u4e2d\u5355\u5143\u6d4b\u8bd5\u80fd\u987a\u5229\u901a\u8fc7":61,"\u4e2d\u5b89\u88c5":68,"\u4e2d\u5b8c\u5168\u4e00\u81f4":33,"\u4e2d\u5b8c\u6210":88,"\u4e2d\u5b9a\u4e49":59,"\u4e2d\u5b9a\u4e49\u4f7f\u7528\u54ea\u79cddataprovid":1,"\u4e2d\u5b9a\u4e49\u548c\u4f7f\u7528":58,"\u4e2d\u5b9e\u73b0\u7684\u7ed3\u6784\u4f53":34,"\u4e2d\u5bfc\u51fa\u9884\u5b9a\u4e49\u7684\u7f51\u7edc":88,"\u4e2d\u5c55\u793a\u4e86\u5982\u4f55\u4f7f\u7528python\u6765\u63d0\u53d6\u7279\u5f81":82,"\u4e2d\u6307\u5b9a":70,"\u4e2d\u6307\u5b9a\u7684\u540d\u5b57":72,"\u4e2d\u6307\u5b9a\u7684\u5c42\u987a\u5e8f\u4e00\u81f4":82,"\u4e2d\u63d0\u4f9b\u4e86\u4e00\u4e9b\u5168\u5c40\u51fd\u6570\u7528\u6765\u5b9e\u73b0paddl":64,"\u4e2d\u63d0\u51fa\u7684resnet\u7f51\u7edc\u7ed3\u6784\u57282015\u5e74imagenet\u5927\u89c4\u6a21\u89c6\u89c9\u8bc6\u522b\u7ade\u8d5b":82,"\u4e2d\u641c\u7d22\u8fd9\u51e0\u4e2a\u5e93":48,"\u4e2d\u6587\u6587\u6863":65,"\u4e2d\u6587\u6587\u6863\u76ee\u5f55":65,"\u4e2d\u6587\u7ef4\u57fa\u767e\u79d1\u9875\u9762":56,"\u4e2d\u6709\u8bb8\u591a\u7684\u7279\u5f81":85,"\u4e2d\u6bcf\u4e2a\u5143\u7d20\u662f\u4e00\u4e2alayer\u7684\u8f93\u51fa\u7ed3\u679c\u77e9\u9635":46,"\u4e2d\u6bcf\u4e2apod\u7684ip\u5730\u5740":75,"\u4e2d\u6bcf\u5c42\u7684\u6570\u503c\u7edf\u8ba1":70,"\u4e2d\u6dfb\u52a0\u4e24\u4e2a\u8f93\u5165":63,"\u4e2d\u7684":[64,82],"\u4e2d\u7684\u4e00\u884c":[2,61],"\u4e2d\u7684\u5185\u5bb9":87,"\u4e2d\u7684\u5bf9\u5e94\u5206\u652f\u5373\u53ef":61,"\u4e2d\u7684\u63a5\u53e3":86,"\u4e2d\u7684\u6570\u636e":82,"\u4e2d\u7684\u6570\u636e\u662f\u5426\u4e3a\u5e8f\u5217\u6a21\u5f0f":86,"\u4e2d\u7684\u6570\u636e\u8fdb\u884c\u9884\u6d4b":82,"\u4e2d\u7684\u6570\u6910\u96c6\u7684\u7ed3\u6784\u5982\u4e0b":88,"\u4e2d\u7684\u6bcf\u4e00\u884c\u547d\u4ee4":86,"\u4e2d\u7684\u751f\u6210\u7ed3\u679c\u5982\u4e0b\u6240\u793a":89,"\u4e2d\u7684\u7528\u6237\u8bc1\u4e66":73,"\u4e2d\u7684\u7b2ci\u884c":89,"\u4e2d\u7684\u8bf4\u660e":2,"\u4e2d\u7684\u8fd9\u4e9b\u6570\u636e\u6587\u4ef6":85,"\u4e2d\u770b\u5230\u4e0b\u9762\u7684\u6587\u4ef6":88,"\u4e2d\u83b7\u53d6":75,"\u4e2d\u8ba4\u771f\u8bbe\u7f6e":68,"\u4e2d\u8bbe\u7f6e":68,"\u4e2d\u8bbe\u7f6e\u7684\u6240\u6709\u8282\u70b9":68,"\u4e2d\u8be6\u7ec6\u4ecb\u7ecd":62,"\u4e2d\u8bfb\u53d6":2,"\u4e2d\u8c03\u7528":63,"\u4e2d\u8fd0\u884c\u4efb\u52a1\u7684\u89d2\u5ea6":21,"\u4e2d\u914d\u7f6e\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":59,"\u4e2d\u914d\u7f6e\u7684\u6548\u679c\u4e00\u81f4":2,"\u4e34\u65f6\u53d8\u91cf\u7b49\u7b49":46,"\u4e3a":[2,53,54,59,63],"\u4e3a0":2,"\u4e3a\u4e86\u4f7f":63,"\u4e3a\u4e86\u4f7f\u7528\u63d0\u524d\u7f16\u5199\u7684\u811a\u672c":88,"\u4e3a\u4e86\u4f7f\u8bc4\u5ba1\u4eba\u5728\u8bc4\u5ba1\u4ee3\u7801\u65f6\u66f4\u597d\u5730\u4e13\u6ce8\u4e8e\u4ee3\u7801\u672c\u8eab":61,"\u4e3a\u4e86\u4fdd\u8bc1\u6548\u7387":62,"\u4e3a\u4e86\u4fdd\u8bc1gpu\u9a71\u52a8\u80fd\u591f\u5728\u955c\u50cf\u91cc\u9762\u6b63\u5e38\u8fd0\u884c":49,"\u4e3a\u4e86\u5145\u5206\u7684\u968f\u673a\u6253\u4e71\u8bad\u7ec3\u96c6":88,"\u4e3a\u4e86\u5b8c\u6210\u5206\u5e03\u5f0f\u673a\u5668\u5b66\u4e60\u8bad\u7ec3\u4efb\u52a1":73,"\u4e3a\u4e86\u5c01\u88c5\u80fd\u591f\u6b63\u786e\u5de5\u4f5c":62,"\u4e3a\u4e86\u5e94\u5bf9\u4ee5\u4e0a\u7684\u95ee\u9898":28,"\u4e3a\u4e86\u5f00\u53d1paddlepaddl":60,"\u4e3a\u4e86\u63cf\u8ff0\u65b9\u4fbf":58,"\u4e3a\u4e86\u65b9\u4fbf\u5728\u56fd\u5185\u7684\u5f00\u53d1\u8005\u4e0b\u8f7ddocker\u955c\u50cf":49,"\u4e3a\u4e86\u65b9\u4fbf\u5927\u5bb6":61,"\u4e3a\u4e86\u65b9\u4fbf\u8d77\u89c1":68,"\u4e3a\u4e86\u66b4\u9732\u7684\u63a5\u53e3\u5c3d\u91cf\u7b80\u5355":34,"\u4e3a\u4e86\u6ee1\u8db3\u8bad\u7ec3":68,"\u4e3a\u4e86\u7b80\u5316cmake\u914d\u7f6e":[53,54],"\u4e3a\u4e86\u8fbe\u5230\u6027\u80fd\u6700\u4f18":67,"\u4e3a\u4e86\u8fd8\u539f":47,"\u4e3a\u4ec0\u4e48\u7528":60,"\u4e3a\u4f8b":[46,63,84],"\u4e3a\u4f8b\u521b\u5efa\u5206\u5e03\u5f0f\u7684\u5355\u8fdb\u7a0b\u8bad\u7ec3":68,"\u4e3a\u4f8b\u6765\u4ecb\u7ecd\u5982\u4f55\u5199\u5e26kernel\u7684oper":63,"\u4e3a\u4f8b\u8fdb\u884c\u9884\u6d4b":84,"\u4e3a\u53c2\u6570\u77e9\u9635\u7684\u5bbd\u5ea6":46,"\u4e3a\u60a8\u505a\u6027\u80fd\u8c03\u4f18\u63d0\u4f9b\u4e86\u65b9\u5411":67,"\u4e3a\u60f3\u4fee\u6b63\u8bcd\u5411\u91cf\u6a21\u578b\u7684\u7528\u6237\u63d0\u4f9b\u4e86\u5c06\u6587\u672c\u8bcd\u5411\u91cf\u6a21\u578b\u8f6c\u6362\u4e3a\u4e8c\u8fdb\u5236\u6a21\u578b\u7684\u547d\u4ee4":80,"\u4e3a\u65b9\u4fbf\u4f5c\u4e1a\u542f\u52a8\u63d0\u4f9b\u4e86\u4e24\u4e2a\u72ec\u7279\u7684\u547d\u4ee4\u9009\u9879":68,"\u4e3a\u6b64":74,"\u4e3a\u6bcf\u4e2aop\u521b\u5efa\u5355\u72ec\u7684":63,"\u4e3a\u8f93\u51fa\u5206\u914d\u5185\u5b58":62,"\u4e3a\u96c6\u7fa4\u4f5c\u4e1a\u8bbe\u7f6e\u989d\u5916\u7684":68,"\u4e3aconst\u51fd\u6570":63,"\u4e3ajson\u6216yaml\u683c\u5f0f":86,"\u4e3aoutput_\u7533\u8bf7\u5185\u5b58":62,"\u4e3b\u8981\u4e3a\u5f00\u53d1\u8005\u4f7f\u7528":70,"\u4e3b\u8981\u529f\u80fd\u5305\u62ec":28,"\u4e3b\u8981\u5305\u62ec\u4ee5\u4e0b\u4e94\u4e2a\u6b65\u9aa4":4,"\u4e3b\u8981\u5305\u62ec\u56db\u79cd\u7c7b\u578b":51,"\u4e3b\u8981\u539f\u56e0":56,"\u4e3b\u8981\u539f\u56e0\u5305\u62ec\u4e24\u4e2a\u65b9\u9762":46,"\u4e3b\u8981\u539f\u56e0\u662f\u589e\u52a0\u4e86\u521d\u59cb\u5316\u673a\u5236":2,"\u4e3b\u8981\u6765\u81ea\u5317\u7f8e\u6d32":81,"\u4e3b\u8981\u7528\u4e8epython":63,"\u4e3b\u8981\u804c\u8d23\u5728\u4e8e\u5c06\u8bad\u7ec3\u6570\u636e\u4f20\u5165\u5185\u5b58\u6216\u8005\u663e\u5b58":84,"\u4e3e\u4e00\u4e2a\u4f8b\u5b50":46,"\u4e3e\u4f8b":46,"\u4e3e\u4f8b\u8bf4\u660e":56,"\u4e4b\u524d":61,"\u4e4b\u524d\u914d\u7f6e\u6587\u4ef6\u4e2d":84,"\u4e4b\u540e":[47,51,62],"\u4e4b\u540e\u4f60\u4f1a\u5f97\u5230\u8bad\u7ec3":68,"\u4e4b\u540e\u4f7f\u7528":62,"\u4e4b\u540e\u4f7f\u7528\u77e9\u9635\u8fd0\u7b97\u51fd\u6570\u6765\u8ba1\u7b97":62,"\u4e4b\u540e\u521d\u59cb\u5316\u6240\u6709\u7684\u6743\u91cd\u77e9\u9635":62,"\u4e4b\u540e\u5b9a\u4e49\u7684":81,"\u4e4b\u5916\u7684\u6240\u6709\u5934\u6587\u4ef6":34,"\u4e4b\u7c7b\u7684\u7a0b\u5e8f\u6765\u7f16\u8bd1\u6e90\u7801":60,"\u4e4b\u95f4\u7684\u8ddd\u79bb":47,"\u4e4b\u95f4\u7684\u8fd0\u7b97\u662f\u72ec\u7acb\u7684":58,"\u4e58\u4e0a\u8f93\u51fa\u7684\u68af\u5ea6":62,"\u4e58\u9664\u7b49\u65f6\u5019":46,"\u4e5d\u4e2a":87,"\u4e5f":56,"\u4e5f\u4e0d\u4f7f\u7528\u5176\u4ed6\u52a8\u6001\u5e93":33,"\u4e5f\u4e0d\u5b58\u5728\u4e00\u4e2asubseq\u76f4\u63a5\u751f\u6210\u4e0b\u4e00\u4e2asubseq\u7684\u60c5\u51b5":58,"\u4e5f\u4e0d\u5e94\u8be5\u62a5\u9519":34,"\u4e5f\u4e0d\u751f\u6210":34,"\u4e5f\u4e0d\u80fd\u63a5\u6536\u5e8f\u5217\u6570\u636e\u4f5c\u4e3a\u8f93\u5165":46,"\u4e5f\u4f1a\u5360\u7528\u78c1\u76d8":60,"\u4e5f\u53ef\u4ee5\u4f7f\u7528":61,"\u4e5f\u53ef\u4ee5\u53bb\u6389\u8fd9\u4e9b\u8bc1\u4e66\u7684\u914d\u7f6e":73,"\u4e5f\u53ef\u4ee5\u5728\u5f00\u53d1\u955c\u50cf\u4e2d\u542f\u52a8\u4e00\u4e2asshd\u670d\u52a1":49,"\u4e5f\u53ef\u4ee5\u662f\u4e00\u4e2a\u8bcd\u8bed":58,"\u4e5f\u53ef\u4ee5\u8bf4\u662f\u67d0\u4e9b\u7279\u5b9a\u6307\u4ee4\u7684\u4f7f\u7528\u60c5\u51b5":67,"\u4e5f\u53ef\u4ee5\u901a\u8fc7\u4fee\u6539":75,"\u4e5f\u53ef\u4ee5\u901a\u8fc7saving_period_by_batches\u8bbe\u7f6e\u6bcf\u9694\u591a\u5c11batch\u4fdd\u5b58\u4e00\u6b21\u6a21\u578b":84,"\u4e5f\u53ef\u4ee5\u914d\u7f6e\u4e0d\u540c\u7684\u91cd\u8bd5\u673a\u5236":73,"\u4e5f\u53ef\u5199\u6210":63,"\u4e5f\u53ef\u901a\u8fc7\u4ee5\u4e0b\u547d\u4ee4\u83b7\u53d6":[53,54],"\u4e5f\u5c31\u662f":61,"\u4e5f\u5c31\u662f\u5c06\u8bcd\u5411\u91cf\u6a21\u578b\u8fdb\u4e00\u6b65\u6f14\u5316\u4e3a\u4e09\u4e2a\u65b0\u6b65\u9aa4":84,"\u4e5f\u5c31\u662f\u672c\u5730\u7684\u6e90\u7801\u6811\u6839\u76ee\u5f55\u91cc\u7684":60,"\u4e5f\u5c31\u662f\u81ea\u5df1\u7528\u6237\u540d\u4e0b\u7684":61,"\u4e5f\u5c31\u662f\u8bf4":[70,72,80],"\u4e5f\u5f97\u5230\u4e00\u4e2a\u7528\u6237\u7279\u5f81":86,"\u4e5f\u63cf\u8ff0\u4e86\u5bb9\u5668\u9700\u8981\u4f7f\u7528\u7684\u5b58\u50a8\u5377\u6302\u8f7d\u7684\u60c5\u51b5":75,"\u4e5f\u652f\u6301cpu\u7684\u6027\u80fd\u5206\u6790":67,"\u4e5f\u662f\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217":56,"\u4e5f\u662f\u5bb9\u5668\u4e0enode\u4e4b\u95f4\u5171\u4eab\u6587\u4ef6\u7684\u65b9\u5f0f":73,"\u4e5f\u662fdecoder\u5faa\u73af\u5c55\u5f00\u7684\u4f9d\u636e":58,"\u4e5f\u662fpaddlepaddle\u6240\u80fd\u591f\u4fdd\u8bc1\u7684shuffle\u7c92\u5ea6":2,"\u4e5f\u6ca1\u7528":46,"\u4e5f\u79f0\u4e3arnn\u6a21\u578b":84,"\u4e5f\u8bb8\u662f\u56e0\u4e3a\u9700\u8981\u5b89\u88c5":81,"\u4e5f\u9700\u8981\u4e24\u6b21\u968f\u673a\u9009\u62e9\u5230\u76f8\u540cgenerator\u7684\u65f6\u5019":2,"\u4e66\u5199":33,"\u4e7e":56,"\u4e86":[56,60],"\u4e86\u89e3\u60a8\u7684\u786c\u4ef6":67,"\u4e86\u89e3\u66f4\u591a\u7ec6\u8282":59,"\u4e86\u89e3\u66f4\u591a\u8be6\u7ec6\u4fe1\u606f":59,"\u4e8c\u7ea7\u76ee\u5f55":[88,89],"\u4e8c\u7ef4\u77e9\u9635":82,"\u4e8c\u8005\u8bed\u610f\u4e0a\u5b8c\u5168\u4e00\u81f4":56,"\u4e8c\u8fdb\u5236":80,"\u4e92\u76f8\u901a\u4fe1":73,"\u4e92\u8054\u7f51\u7535\u5f71\u6570\u636e\u5e93":88,"\u4e94\u661f\u7ea7":56,"\u4e9a\u9a6c\u900a":88,"\u4ea4\u53c9\u7f16\u8bd1\u5de5\u5177\u94fe":54,"\u4ea4\u53c9\u7f16\u8bd1\u5de5\u5177\u94fe\u4e3a":53,"\u4ea4\u53c9\u7f16\u8bd1android\u7248\u672c\u7684paddlepaddle\u5e93\u65f6":53,"\u4ea4\u53c9\u7f16\u8bd1raspberri":54,"\u4ea4\u7531cmake\u7cfb\u7edf\u672c\u8eab\u6765\u5904\u7406":53,"\u4ea4\u901a":56,"\u4ea4\u901a\u4fbf\u5229":56,"\u4ea6\u53ef\u4ee5\u901a\u8fc7\u624b\u52a8\u8bbe\u7f6e":[53,54],"\u4eab\u53d7\u60a8\u7684\u65c5\u7a0b":49,"\u4eba\u8138\u8bc6\u522b":21,"\u4ec0\u4e48":86,"\u4ec0\u4e48\u662f":60,"\u4ec5\u4ec5\u4f7f\u7528":33,"\u4ec5\u4ec5\u662f\u4e00\u4e9b\u5173\u952e\u8bcd":88,"\u4ec5\u4ec5\u662f\u4e24\u4e2a\u5168\u8fde\u63a5\u5c42":86,"\u4ec5\u4ec5\u662f\u7b80\u5355\u7684\u5d4c\u5165":86,"\u4ec5\u5305\u542b\u8bad\u7ec3\u548c\u6d4b\u8bd5\u6570\u6910\u96c6":88,"\u4ec5\u5728\u8fdc\u7a0b\u7a00\u758f\u8bad\u7ec3\u65f6\u6709\u6548":62,"\u4ec5\u5bf9\u7a00\u758f\u6570\u636e\u6709\u6548":62,"\u4ec5\u9700\u8981\u77e5\u9053\u5982\u4f55\u4ece":2,"\u4ecb\u7ecd\u4e86\u4e00\u79cd\u901a\u8fc7ssh\u8fdc\u7a0b\u5206\u53d1\u4efb\u52a1":75,"\u4ecb\u7ecd\u4ea4\u53c9\u7f16\u8bd1android\u5e73\u53f0\u4e0a\u9002\u7528\u7684paddlepaddle\u5e93\u7684\u65b9\u6cd5\u548c\u6b65\u9aa4":53,"\u4ecb\u7ecd\u4ea4\u53c9\u7f16\u8bd1raspberri":54,"\u4ecb\u7ecd\u5206\u5e03\u5f0f\u8bad\u7ec3\u4e4b\u524d":73,"\u4ecb\u7ecdpaddlepaddle\u7684\u57fa\u672c\u4f7f\u7528\u65b9\u6cd5":84,"\u4ece":[42,46,67,87],"\u4ece0\u5230num":70,"\u4ece\u4e00\u4e2aword\u751f\u6210\u4e0b\u4e00\u4e2aword":58,"\u4ece\u5185\u6838\u51fd\u6570\u7684\u89d2\u5ea6":67,"\u4ece\u56fe\u4e2d\u53ef\u4ee5\u770b\u5230":47,"\u4ece\u5916\u90e8\u7f51\u7ad9\u4e0a\u4e0b\u8f7d\u7684\u539f\u59cb\u6570\u6910\u96c6":88,"\u4ece\u5927\u5230\u5c0f":89,"\u4ece\u6570\u636e\u63d0\u4f9b\u7a0b\u5e8f\u52a0\u8f7d\u5b9e\u4f8b":87,"\u4ece\u6a21\u578b\u6587\u4ef6\u5c06\u9884\u8bad\u7ec3\u53c2\u6570\u8f7d\u5165":46,"\u4ece\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u6765\u770b":56,"\u4ece\u6bcf\u4e2a\u5355\u8bcd\u5de6\u53f3\u4e24\u7aef\u5206\u522b\u83b7\u53d6k\u4e2a\u76f8\u90bb\u7684\u5355\u8bcd":84,"\u4ece\u6e90\u7801\u4ea4\u53c9\u7f16\u8bd1paddlepaddl":[53,54],"\u4ece\u78c1\u76d8\u6587\u4ef6\u4e2d\u52a0\u8f7duuid\u6587\u4ef6\u540d\u7684\u68c0\u67e5\u70b9\u5feb\u7167\u6587\u4ef6":20,"\u4ece\u7b2c0\u4e2a\u8bc4\u4f30\u5230\u5f53\u524d\u8bc4\u4f30\u4e2d":89,"\u4ece\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u7684\u5e73\u5747\u635f\u5931":88,"\u4ece\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u7684\u5e73\u5747cost":89,"\u4ece\u800c\u53ef\u4ee5\u505a\u4e00\u4e9b\u4e0e\u8ba1\u7b97\u91cd\u53e0\u7684\u5de5\u4f5c":62,"\u4ece\u800c\u53ef\u4ee5\u62df\u5408\u4efb\u610f\u7684\u51fd\u6570\u6765\u5b66\u4e60\u590d\u6742\u7684\u6570\u636e\u5173\u7cfb":47,"\u4ece\u800c\u5f15\u53d1\u5176\u4ed6\u8282\u70b9\u65e0\u6cd5\u8fde\u63a5\u5bfc\u81f4":46,"\u4ece\u800c\u751f\u6210\u591a\u4e2agener":2,"\u4ece\u800c\u80fd\u591f\u88abpaddlepaddl":84,"\u4ece\u800c\u9632\u6b62\u8fc7\u62df\u5408":1,"\u4ece\u8be5\u94fe\u63a5":89,"\u4ece\u8bed\u4e49\u4e0a\u770b":58,"\u4ece\u8f93\u5165\u6570\u636e\u4e0a\u770b":56,"\u4ece\u8f93\u51fa\u65e5\u5fd7\u53ef\u4ee5\u770b\u5230":47,"\u4ece\u9884\u8bad\u7ec3\u6a21\u578b\u4e2d":80,"\u4ececmake":53,"\u4eceetcd\u4e2d\u8bfb\u53d6\u8282\u70b9":20,"\u4ecestart":70,"\u4ecetest":89,"\u4ed3\u5e93\u7684\u8fdc\u7a0b\u4e3b\u673a":61,"\u4ed4\u7ec6\u89c2\u5bdf":82,"\u4ed6\u4e3b\u8981\u5305\u542b\u4e86\u5b9e\u9645\u66b4\u9732\u7684\u7c7b\u578b\u7ed3\u6784":34,"\u4ed6\u4eec\u5206\u522b\u662f":56,"\u4ed6\u4eec\u5728\u81ea\u5df1\u7684":60,"\u4ed6\u4eec\u5728paddle\u7684\u6587\u6863\u548capi\u4e2d\u662f\u4e00\u4e2a\u6982\u5ff5":56,"\u4ed6\u4eec\u63d0\u51fa\u6b8b\u5dee\u5b66\u4e60\u7684\u6846\u67b6\u6765\u7b80\u5316\u7f51\u7edc\u7684\u8bad\u7ec3":82,"\u4ed6\u662f\u5c06":34,"\u4ed6\u7684\u76ee\u6807\u662f\u4f7f\u7528c":33,"\u4ee3\u66ff":75,"\u4ee3\u7801":86,"\u4ee3\u7801\u4e2d9":56,"\u4ee3\u7801\u5982\u4e0b":[46,59],"\u4ee3\u7801\u6ce8\u91ca\u8bf7\u9075\u5b88":61,"\u4ee3\u7801\u751f\u6210\u7684\u7b26\u53f7\u53ef\u80fd\u4e0d\u4e00\u81f4":33,"\u4ee3\u7801\u793a\u4f8b\u5982\u4e0b":63,"\u4ee3\u8868\u5bbf\u4e3b\u673a\u76ee\u5f55":75,"\u4ee3\u8868\u7f16\u53f7":86,"\u4ee3\u8868\u8fd9\u4e2ashard\u7684\u6700\u5927index":21,"\u4ee3\u8868shard\u7684index":21,"\u4ee5":46,"\u4ee5\u4e0a":61,"\u4ee5\u4e0a\u4ee3\u7801\u7684reader\u8f93\u51fa\u7684data":21,"\u4ee5\u4e0a\u547d\u4ee4\u4f1a\u5728\u5f53\u524d\u76ee\u5f55\u4e0b\u751f\u6210100\u4e2a\u6587\u4ef6":21,"\u4ee5\u4e0a\u547d\u4ee4\u4f1a\u6309\u7167\u751f\u6210\u7684dockerfile\u628a\u751f\u6210\u7684\u7a0b\u5e8f\u62f7\u8d1d\u5230\u751f\u4ea7\u955c\u50cf\u4e2d\u5e76\u505a\u76f8\u5e94\u7684\u914d\u7f6e":49,"\u4ee5\u4e0a\u547d\u4ee4\u4f1a\u7f16\u8bd1paddlepaddl":49,"\u4ee5\u4e0a\u65b9\u6cd5\u5728gpu\u955c\u50cf\u91cc\u4e5f\u80fd\u7528":49,"\u4ee5\u4e0b":[21,86],"\u4ee5\u4e0b\u4ee3\u7801\u7247\u6bb5\u5b9a\u4e49":59,"\u4ee5\u4e0b\u547d\u4ee4\u542f\u52a8\u4e00\u4e2a":60,"\u4ee5\u4e0b\u6211\u4eec\u7ffb\u8bd1\u6570\u636e\u96c6\u7f51\u7ad9\u4e2dreadme\u6587\u4ef6\u7684\u63cf\u8ff0":85,"\u4ee5\u4e0b\u6307\u4ee4\u80fd\u68c0\u67e5linux\u7535\u8111\u662f\u5426\u652f\u6301avx":49,"\u4ee5\u4e0b\u6559\u7a0b\u5c06\u6307\u5bfc\u60a8\u63d0\u4ea4\u4ee3\u7801":61,"\u4ee5\u4e0b\u662f\u5bf9\u4e0a\u8ff0\u6570\u636e\u52a0\u8f7d\u7684\u89e3\u91ca":84,"\u4ee5\u4e0b\u6b65\u9aa4\u57fa\u4e8e":68,"\u4ee5\u4e0b\u793a\u8303\u5982\u4f55\u4f7f\u7528\u9884\u8bad\u7ec3\u7684\u4e2d\u6587\u5b57\u5178\u548c\u8bcd\u5411\u91cf\u8fdb\u884c\u77ed\u8bed\u6539\u5199":80,"\u4ee5\u4e0b\u9009\u9879\u5fc5\u987b\u5728":68,"\u4ee5\u4ea4\u4e92\u5bb9\u5668\u65b9\u5f0f\u8fd0\u884c\u5f00\u53d1\u955c\u50cf":49,"\u4ee5\u4f7f\u7528adam\u7b97\u6cd5\u4e3a\u4f8b":46,"\u4ee5\u4fbf\u6211\u4eec\u53ef\u4ee5\u628a\u66f4\u591a\u7684\u7cbe\u529b\u653e\u5230\u903b\u8f91\u672c\u8eab\u4e0a":28,"\u4ee5\u4fbf\u7528\u6237":68,"\u4ee5\u4fdd\u8bc1\u68af\u5ea6\u7684\u6b63\u786e\u8ba1\u7b97":62,"\u4ee5\u4fdd\u8bc1\u68af\u5ea6\u8ba1\u7b97\u7684\u6b63\u786e\u6027":62,"\u4ee5\u4fdd\u8bc1\u7f16\u8bd1\u9ad8\u6548":60,"\u4ee5\u5206\u7c7b\u6765\u81ea":88,"\u4ee5\u53ca":[60,62],"\u4ee5\u53ca\u4f7f\u7528\u5b50\u5e8f\u5217\u6765\u5b9a\u4e49\u5206\u7ea7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u67b6\u6784":59,"\u4ee5\u53ca\u5207\u6362\u673a\u5668\u65f6\u9700\u8981\u65b0\u5b89\u88c5\u7684\u8f9b\u82e6":60,"\u4ee5\u53ca\u53cc\u5c42\u5e8f\u5217":55,"\u4ee5\u53ca\u5728wmt":89,"\u4ee5\u53ca\u5982\u4f55\u5728\u5c42\u4e4b\u95f4\u8fdb\u884c\u8fde\u63a5":81,"\u4ee5\u53ca\u751f\u6210\u521b\u5efa\u751f\u4ea7\u955c\u50cf\u7684dockerfil":49,"\u4ee5\u53ca\u76ee\u6807\u673a\u7248openblas\u5e93":54,"\u4ee5\u53ca\u76f8\u5173\u7684\u5c5e\u6027\u53c2\u6570":63,"\u4ee5\u53ca\u8ba1\u7b97\u903b\u8f91\u5728\u5e8f\u5217\u4e0a\u7684\u5faa\u73af\u5c55\u5f00":58,"\u4ee5\u53ca\u8f93\u5165\u7684\u68af\u5ea6":62,"\u4ee5\u53caandroid":53,"\u4ee5\u53caandroid\u6240\u9700":53,"\u4ee5\u53canumpi":21,"\u4ee5\u53capaddle\u5982\u4f55\u5904\u7406\u591a\u79cd\u7c7b\u578b\u7684\u8f93\u5165":86,"\u4ee5\u53carelu":62,"\u4ee5\u63d0\u4f9b\u4e00\u4e9b\u9ed8\u8ba4\u7684\u7f16\u8bd1\u5668\u548c\u7f16\u8bd1\u53c2\u6570\u76f8\u5173\u914d\u7f6e":[53,54],"\u4ee5\u76f8\u5bf9\u8def\u5f84\u5f15\u7528":1,"\u4ee5\u786e\u4fdd\u6240\u6709\u7684\u7b2c\u4e09\u65b9\u4f9d\u8d56\u5e93\u548cpaddlepaddle\u4ee3\u7801\u90fd\u662f\u9488\u5bf9\u65b0\u7684cmake\u914d\u7f6e\u91cd\u65b0\u7f16\u8bd1\u7684":[53,54],"\u4ee5\u83b7\u5f97\u66f4\u597d\u7684\u7f51\u7edc\u6027\u80fd":68,"\u4ee5\u8f93\u51fa":46,"\u4ee5\u9017\u53f7":80,"\u4ee5\u9017\u53f7\u95f4\u9694":70,"\u4ee5eigentensor\u4e3a\u4f8b":64,"\u4ee5embedding\u5c42\u4e3a\u4f8b":46,"\u4ee5lstm\u4e3a\u4f8b":46,"\u4ef7\u683c":56,"\u4efb\u52a1":86,"\u4efb\u52a1\u6765\u7ec8\u6b62\u96c6\u7fa4\u4f5c\u4e1a":68,"\u4efb\u610f\u5c06\u4e00\u4e9b\u6570\u636e\u7ec4\u5408\u6210\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217":56,"\u4efb\u610f\u65f6\u523b\u53ea\u53ef\u80fd\u540c\u65f6\u6709\u4e00\u53f0\u670d\u52a1\u5668\u6545\u969c":20,"\u4f18\u5316":88,"\u4f18\u5316\u5668\u5219\u7528\u94fe\u5f0f\u6cd5\u5219\u6765\u5bf9\u6bcf\u4e2a\u53c2\u6570\u8ba1\u7b97\u635f\u5931\u51fd\u6570\u7684\u68af\u5ea6":62,"\u4f1a\u5148\u8fdb\u884c\u53c2\u6570\u7684\u521d\u59cb\u5316\u4e0e\u89e3\u6790":75,"\u4f1a\u5171\u4eab\u53c2\u6570":46,"\u4f1a\u5173\u8054\u53c2\u6570":46,"\u4f1a\u52a0\u8f7d\u4e0a\u4e00\u8f6e\u7684\u53c2\u6570":70,"\u4f1a\u53d8\u6210\u8bcd\u8868\u4e2d\u7684\u4f4d\u7f6e":56,"\u4f1a\u542f\u52a8pserver\u4e0etrainer\u8fdb\u7a0b":75,"\u4f1a\u5728\u5f53\u524d\u76ee\u5f55\u751f\u6210\u4e24\u4e2a\u5b50\u76ee\u5f55":65,"\u4f1a\u5bf9\u6bcf\u4e00\u4e2a\u6fc0\u6d3b\u6682\u5b58\u4e00\u4e9b\u6570\u636e":46,"\u4f1a\u5bf9\u8fd9\u7c7b\u8f93\u5165\u8fdb\u884c\u62c6\u89e3":58,"\u4f1a\u5bfc\u81f4\u4e0d\u540c\u7248\u672cpython\u5728\u4e00\u4e2a\u8fdb\u7a0b\u91cc\u7684bug":33,"\u4f1a\u5c06\u6bcf\u4e2a\u65f6\u95f4\u6b65\u7684\u8f93\u51fa\u62fc\u63a5":58,"\u4f1a\u5c06\u7b2c\u4e00\u4e2a":46,"\u4f1a\u6210\u4e3astep\u51fd\u6570\u7684\u8f93\u5165":58,"\u4f1a\u6254\u5230\u8fd9\u6761\u6570\u636e":2,"\u4f1a\u62a5\u5982\u4e0b\u7684\u9519\u8bef":46,"\u4f1a\u62a5\u9519":58,"\u4f1a\u6839\u636e\u547d\u4ee4\u884c\u53c2\u6570\u6307\u5b9a\u7684\u6d4b\u8bd5\u65b9\u5f0f":1,"\u4f1a\u6839\u636einput_types\u68c0\u67e5\u6570\u636e\u7684\u5408\u6cd5\u6027":2,"\u4f1a\u76f4\u63a5\u62a5\u9519\u9000\u51fa":33,"\u4f1a\u76f8\u5e94\u5730\u6539\u53d8\u8f93\u51fa\u7684\u5c3a\u5bf8":62,"\u4f1a\u81ea\u52a8\u5173\u95ed\u5bf9\u5e94\u7684issu":61,"\u4f1a\u81ea\u9002\u5e94\u5730\u4ece\u8fd9\u4e9b\u5411\u91cf\u4e2d\u9009\u62e9\u4e00\u4e2a\u5b50\u96c6\u51fa\u6765":89,"\u4f1a\u83b7\u53d6\u5f53\u524dnamespace\u4e0b\u7684\u6240\u6709pod":75,"\u4f1a\u88ab\u62c6\u89e3\u4e3a\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":58,"\u4f1a\u88ab\u62c6\u89e3\u4e3a\u975e\u5e8f\u5217":58,"\u4f1a\u88abpickle\u5e8f\u5217\u5316\u6210\u5b57\u7b26\u4e32":21,"\u4f1a\u9020\u6210\u90ae\u4ef6\u707e\u96be":61,"\u4f20\u5165":[2,21],"\u4f20\u5165\u4e0a\u4e00\u6b65\u89e3\u6790\u51fa\u6765\u7684\u6a21\u578b\u914d\u7f6e\u5c31\u53ef\u4ee5\u521b\u5efa\u4e00\u4e2a":4,"\u4f20\u5165\u9884\u6d4b\u6570\u636e":4,"\u4f20\u7ed9dataprovider\u7684\u67d0\u4e00\u4e2aargs\u8fc7\u5927":46,"\u4f20\u9012\u7ed9\u914d\u7f6e\u6587\u4ef6\u7684\u53c2\u6570":70,"\u4f46":34,"\u4f46\u4e0d\u66b4\u9732":34,"\u4f46\u4e0d\u7528\u4e8e\u8ba1\u7b97\u68af\u5ea6":62,"\u4f46\u4e0d\u9700\u8981\u63d0\u524d\u521b\u5efa":70,"\u4f46\u4e8e\u53cc\u5c42\u5e8f\u5217\u7684lstm\u6765\u8bf4":56,"\u4f46\u548c\u5355\u5c42rnn\u4e0d\u540c":56,"\u4f46\u5728\u8d77\u521d\u7684\u51e0\u8f6e\u8bad\u7ec3\u4e2d\u5b83\u4eec\u90fd\u5728\u5feb\u901f\u903c\u8fd1\u771f\u5b9e\u503c":47,"\u4f46\u5b50\u53e5\u542b\u6709\u7684\u8bcd\u8bed\u6570\u53ef\u4ee5\u4e0d\u76f8\u7b49":58,"\u4f46\u5c3d\u91cf\u8bf7\u4fdd\u6301\u7f16\u8bd1\u548c\u8fd0\u884c\u4f7f\u7528\u7684cudnn\u662f\u540c\u4e00\u4e2a\u7248\u672c":48,"\u4f46\u5e76\u6ca1\u6709\u7ecf\u8fc7\u56de\u5f52\u6d4b\u8bd5":42,"\u4f46\u5e8f\u5217\u8f93\u51fa\u65f6":56,"\u4f46\u5f53\u8c03\u7528\u8fc7\u4e00\u6b21\u540e":2,"\u4f46\u622a\u65ad\u65f6\u673a\u4e0d\u540c":46,"\u4f46\u6240\u6709fork\u7684\u7248\u672c\u5e93\u7684\u6240\u6709\u5206\u652f\u90fd\u76f8\u5f53\u4e8e\u7279\u6027\u5206\u652f":42,"\u4f46\u662f":[46,56],"\u4f46\u662f2008\u5e74\u4e4b\u524d\u751f\u4ea7\u7684\u65e7\u7535\u8111\u4e0d\u652f\u6301avx":49,"\u4f46\u662f\u4e5f\u6ca1\u6709\u5fc5\u8981\u5220\u9664\u65e0\u7528\u7684\u6587\u4ef6":68,"\u4f46\u662f\u53c8\u8fc7\u4e8e\u7410\u788e":34,"\u4f46\u662f\u5927\u90e8\u5206\u53c2\u6570\u662f\u4e3a\u5f00\u53d1\u8005\u63d0\u4f9b\u7684":69,"\u4f46\u662f\u5b50\u5e8f\u5217\u7684\u6570\u76ee\u5fc5\u987b\u4e00\u6837":56,"\u4f46\u662f\u6211\u4eec\u5e76\u4e0d\u63a8\u8350\u8fd9\u79cd\u65b9\u6cd5":49,"\u4f46\u662f\u652f\u6301avx\u6307\u4ee4\u96c6":61,"\u4f46\u662f\u65b9\u4fbf\u8c03\u8bd5\u548c\u6d4bbenchmark":48,"\u4f46\u662f\u662f\u8fd0\u884c\u5728\u72ec\u7acb\u7684\u8fdb\u7a0b\u7a7a\u95f4":49,"\u4f46\u662f\u6bcf\u4e2a\u6837\u672c\u4ec5\u5305\u542b\u51e0\u4e2a\u8bcd":72,"\u4f46\u662f\u7a81\u7136\u6709\u4e00\u4e2a10000\u957f\u7684\u5e8f\u5217":46,"\u4f46\u662f\u865a\u62df\u7684\u4e0d\u4ec5\u4ec5\u662f":60,"\u4f46\u662f\u89e3\u91ca\u6027\u8bed\u8a00":33,"\u4f46\u662f\u8fd9\u79cd\u65b9\u6cd5\u5728\u6bcf\u5c42\u53ea\u4fdd\u5b58\u9884\u8bbe\u6570\u91cf\u7684\u6700\u4f18\u72b6\u6001":89,"\u4f46\u662f\u9690\u85cf\u5c42\u4e2d\u7684\u6bcf\u4e2a\u666e\u901a\u8282\u70b9\u88ab\u4e00\u4e2a\u8bb0\u5fc6\u5355\u5143\u66ff\u6362":88,"\u4f46\u662fbatch":46,"\u4f46\u6709\u503c\u7684\u5730\u65b9\u5fc5\u987b\u4e3a1":[2,51],"\u4f46\u6709\u503c\u7684\u90e8\u5206\u53ef\u4ee5\u662f\u4efb\u4f55\u6d6e\u70b9\u6570":[2,51],"\u4f46\u8fd9\u4e2a\u5173\u7cfb\u53ef\u80fd\u4e0d\u6b63\u786e":2,"\u4f46\u9700\u6ce8\u610f\u53cd\u5411op\u6ca1\u6709":63,"\u4f46eigen":64,"\u4f4d\u7f6e":56,"\u4f4f":56,"\u4f53\u88c1\u5b57\u5178":86,"\u4f53\u88c1\u5b57\u6bb5":86,"\u4f59\u5f26\u76f8\u4f3c\u5ea6\u56de\u5f52":86,"\u4f59\u5f26\u76f8\u4f3c\u5ea6\u5c42":86,"\u4f5c\u4e3a\u4e0b\u4e00\u4e2a\u5b50\u53e5memory\u7684\u521d\u59cb\u72b6\u6001":56,"\u4f5c\u4e3a\u4f8b\u5b50\u6f14\u793a\u5982\u4f55\u914d\u7f6e\u590d\u6742\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u6a21\u578b":59,"\u4f5c\u4e3a\u53c2\u6570\u5c5e\u6027":63,"\u4f5c\u4e3a\u53c2\u6570\u7684id":46,"\u4f5c\u4e3a\u5b58\u50a8\u7cfb\u7edf":21,"\u4f5c\u4e3a\u5f53\u524d\u65f6\u523b\u8f93\u5165":58,"\u4f5c\u4e3a\u793a\u4f8b\u6570\u636e":85,"\u4f5c\u4e3a\u7c7b\u53e5\u67c4":33,"\u4f5c\u4e3a\u7ebf\u6027\u56de\u5f52\u7684\u8f93\u5165":47,"\u4f5c\u4e3a\u8f93\u51fa":59,"\u4f5c\u4e3a\u96c6\u7fa4\u8bad\u7ec3\u7684\u5de5\u4f5c\u7a7a\u95f4":68,"\u4f5c\u4e3aboot_layer\u4f20\u7ed9\u4e0b\u4e00\u4e2a\u5b50\u53e5\u7684memori":56,"\u4f5c\u5bb6":85,"\u4f5c\u7528":55,"\u4f60\u4e5f\u53ef\u4ee5\u4f7f\u7528\u8fd9\u4e09\u4e2a\u503c":82,"\u4f60\u4e5f\u53ef\u4ee5\u5148\u8df3\u8fc7\u672c\u6587\u7684\u89e3\u91ca\u73af\u8282":84,"\u4f60\u4e5f\u53ef\u4ee5\u7b80\u5355\u7684\u8fd0\u884c\u4ee5\u4e0b\u7684\u547d\u4ee4":80,"\u4f60\u4e5f\u53ef\u4ee5\u901a\u8fc7\u5728\u547d\u4ee4\u884c\u53c2\u6570\u4e2d\u589e\u52a0\u4e00\u4e2a\u53c2\u6570\u5982":82,"\u4f60\u4e5f\u8bb8\u53ef\u4ee5\u5c1d\u8bd5\u66f4\u8001\u7684\u65b9\u6cd5":49,"\u4f60\u53ea\u9700\u5b8c\u6210":68,"\u4f60\u53ea\u9700\u8981\u5728\u547d\u4ee4\u884c\u8f93\u5165\u4ee5\u4e0b\u547d\u4ee4":84,"\u4f60\u53ea\u9700\u8981\u6309\u7167\u5982\u4e0b\u65b9\u5f0f\u7ec4\u7ec7\u6570\u636e":89,"\u4f60\u53ef\u4ee5\u4f7f\u7528":82,"\u4f60\u53ef\u4ee5\u4f7f\u7528\u4e0b\u9762\u7684\u811a\u672c\u4e0b\u8f7d":88,"\u4f60\u53ef\u4ee5\u4f7f\u7528\u5b83\u6765\u8bbe\u7f6e":68,"\u4f60\u53ef\u4ee5\u4f7f\u7528\u672c\u5730\u8bad\u7ec3\u4e2d\u7684\u76f8\u540c\u6a21\u578b\u6587\u4ef6\u8fdb\u884c\u96c6\u7fa4\u8bad\u7ec3":68,"\u4f60\u53ef\u4ee5\u5728\u4efb\u4f55\u65f6\u5019\u7528":86,"\u4f60\u53ef\u4ee5\u5bf9\u6a21\u578b\u8fdb\u884c\u8bc4\u4f30":86,"\u4f60\u53ef\u4ee5\u5c06\u7f51\u7edc\u914d\u7f6e\u6210\u67d0\u4e9b\u5c42\u4f7f\u7528gpu\u8ba1\u7b97":72,"\u4f60\u53ef\u4ee5\u6267\u884c\u4e0a\u8ff0\u547d\u4ee4\u6765\u4e0b\u8f7d\u6240\u6709\u7684\u6a21\u578b\u548c\u5747\u503c\u6587\u4ef6":82,"\u4f60\u53ef\u4ee5\u901a\u8fc7\u6267\u884c\u4e0b\u9762\u7684\u547d\u4ee4\u6765\u5f97\u5230resnet\u7f51\u7edc\u7684\u7ed3\u6784\u53ef\u89c6\u5316\u56fe":82,"\u4f60\u53ef\u4ee5\u9884\u6d4b\u4efb\u4f55\u7528\u6237\u5bf9\u4e8e\u4efb\u4f55\u4e00\u90e8\u7535\u5f71\u7684\u8bc4\u4ef7":86,"\u4f60\u5c06\u4f1a\u770b\u5230\u4ee5\u4e0b\u7684\u6a21\u578b\u7ed3\u6784":80,"\u4f60\u5c06\u4f1a\u770b\u5230\u5982\u4e0b\u6d88\u606f":89,"\u4f60\u5c06\u4f1a\u770b\u5230\u5982\u4e0b\u7ed3\u679c":82,"\u4f60\u5c06\u4f1a\u770b\u5230\u7279\u5f81\u5b58\u50a8\u5728":82,"\u4f60\u5c06\u4f1a\u770b\u5230\u8fd9\u6837\u7684\u6d88\u606f":89,"\u4f60\u5c06\u5728\u76ee\u5f55":88,"\u4f60\u5c06\u770b\u5230\u5982\u4e0b\u7684\u4fe1\u606f":86,"\u4f60\u7684\u5de5\u4f5c\u7a7a\u95f4\u5e94\u5982\u4e0b\u6240\u793a":68,"\u4f60\u8fd8\u53ef\u4ee5\u5c06\u7528\u6237\u548c":68,"\u4f60\u9700\u8981\u4e00\u4e9b\u66f4\u590d\u6742\u7684\u5355\u5143\u6d4b\u8bd5\u6765\u4fdd\u8bc1\u4f60\u5b9e\u73b0\u7684\u7f51\u7edc\u5c42\u662f\u6b63\u786e\u7684":62,"\u4f60\u9700\u8981\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u6307\u5b9a\u8bbe\u5907\u7684id\u53f7":72,"\u4f60\u9700\u8981\u5728\u914d\u7f6ecmake\u65f6\u5c06":62,"\u4f60\u9700\u8981\u5b89\u88c5python\u7684\u7b2c\u4e09\u65b9\u5e93":86,"\u4f60\u9700\u8981\u628a\u8be5\u6587\u4ef6\u52a0\u5165":62,"\u4f7f\u5176\u8f6c\u53d8\u4e3a\u7ef4\u5ea6\u4e3ahidden_dim\u7684\u65b0\u5411\u91cf":84,"\u4f7f\u5f97":47,"\u4f7f\u5f97\u4e24\u4e2a\u5b57\u5178\u6709\u76f8\u540c\u7684\u4e0a\u4e0b\u6587":89,"\u4f7f\u5f97\u5355\u5143\u6d4b\u8bd5\u6709\u4e00\u4e2a\u5e72\u51c0\u7684\u73af\u5883":46,"\u4f7f\u5f97\u642d\u6a21\u578b\u65f6\u66f4\u65b9\u4fbf":62,"\u4f7f\u5f97\u6700\u7ec8\u5f97\u5230\u7684\u6a21\u578b\u51e0\u4e4e\u4e0e\u771f\u5b9e\u6a21\u578b\u4e00\u81f4":47,"\u4f7f\u7528":[34,42,46,49,56,58,59,62,67,70,84,87,88],"\u4f7f\u75280\u53f7\u548c1\u53f7gpu\u8ba1\u7b97fc2\u5c42":72,"\u4f7f\u75280\u53f7gpu\u8ba1\u7b97fc2\u5c42":72,"\u4f7f\u752810\u4e2a\u88c1\u526a\u56fe\u50cf\u5757":82,"\u4f7f\u75281\u53f7gpu\u8ba1\u7b97fc3\u5c42":72,"\u4f7f\u75282\u53f7\u548c3\u53f7gpu\u8ba1\u7b97fc3\u5c42":72,"\u4f7f\u7528\u4e00\u4e2a\u5c3a\u5ea6\u4e3a":62,"\u4f7f\u7528\u4e00\u4e2a\u8bcd\u524d\u4e24\u4e2a\u8bcd\u548c\u540e\u4e24\u4e2a\u8bcd":46,"\u4f7f\u7528\u4e0a\u6587\u521b\u5efa\u7684yaml\u6587\u4ef6\u521b\u5efakubernet":74,"\u4f7f\u7528\u4e0b\u9762\u547d\u4ee4":21,"\u4f7f\u7528\u4e86\u540c\u6837\u7684parameter\u548cbia":46,"\u4f7f\u7528\u4e86\u57fa\u4e8e\u53e5\u6cd5\u7ed3\u6784\u7684\u9884\u5b9a\u4e49\u7279\u5f81\u6a21\u677f":87,"\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\u8fdb\u884c\u6a21\u578b\u8bad\u7ec3":80,"\u4f7f\u7528\u5177\u6709softmax\u6fc0\u6d3b\u7684\u5168\u8fde\u63a5\u524d\u9988\u5c42\u6765\u6267\u884c\u5206\u7c7b\u4efb\u52a1":88,"\u4f7f\u7528\u52a8\u6001\u5e93":33,"\u4f7f\u7528\u540c\u6837\u7684\u8bad\u7ec3\u6570\u636eblock":20,"\u4f7f\u7528\u591a\u5757\u663e\u5361\u8bad\u7ec3":46,"\u4f7f\u7528\u591a\u7ebf\u7a0b\u8bad\u7ec3":46,"\u4f7f\u7528\u5982\u4e0b\u53c2\u6570":81,"\u4f7f\u7528\u5982\u4e0b\u547d\u4ee4":80,"\u4f7f\u7528\u5b66\u4e60\u5b8c\u6210\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u6a21\u578b\u751f\u6210\u5e8f\u5217":59,"\u4f7f\u7528\u5bb9\u5668\u65b9\u5f0f\u8fd0\u884c\u8bad\u7ec3\u4efb\u52a1\u7684kubernet":75,"\u4f7f\u7528\u6211\u4eec\u4e4b\u524d\u6784\u9020\u7684\u955c\u50cf":74,"\u4f7f\u7528\u6570\u503c\u6cd5\u68c0\u6d4b\u68af\u5ea6\u6b63\u786e\u6027\u548c\u7a33\u5b9a\u6027":63,"\u4f7f\u7528\u6587\u6863":63,"\u4f7f\u7528\u663e\u5361\u8bad\u7ec3":46,"\u4f7f\u7528\u667a\u80fd\u6307\u9488\u7684\u539f\u56e0\u662f":34,"\u4f7f\u7528\u6848\u4f8b":71,"\u4f7f\u7528\u7684\u7f16\u8bd1\u5668\u4e3aarm":53,"\u4f7f\u7528\u76f8\u5bf9\u8def\u5f84\u7684\u5f15\u7528\u65b9\u5f0f":34,"\u4f7f\u7528\u8005\u4e0d\u9700\u8981\u5173\u5fc3":70,"\u4f7f\u7528\u8005\u53ea\u9700\u8981\u5173\u6ce8\u4e8e\u8bbe\u8ba1rnn\u5728\u4e00\u4e2a\u65f6\u95f4\u6b65\u4e4b\u5185\u5b8c\u6210\u7684\u8ba1\u7b97":58,"\u4f7f\u7528\u8005\u53ef\u4ee5\u4f7f\u7528\u4e0b\u9762\u7684python\u811a\u672c\u6765\u8bfb\u53d6\u53c2\u6570\u503c":82,"\u4f7f\u7528\u8005\u65e0\u9700\u5173\u5fc3\u8fd9\u4e2a\u53c2\u6570":70,"\u4f7f\u7528\u8005\u901a\u5e38\u65e0\u9700\u5173\u5fc3":70,"\u4f7f\u7528\u81ea\u52a8\u7684\u66ff\u8865\u6765\u66ff\u4ee3\u7ecf\u9a8c\u4e30\u5bcc\u7684\u4eba\u5de5\u8bc4\u5224":89,"\u4f7f\u7528\u8be5learning_rate_schedule\u65f6":46,"\u4f7f\u7528\u8c13\u8bcd\u4e0a\u4e0b\u6587":87,"\u4f7f\u7528\u8fd9\u4e2a\u795e\u7ecf\u7f51\u7edc\u53ef\u4ee5\u5b8c\u6210\u5bf9\u65b0\u6570\u636e\u7684\u9884\u6d4b":20,"\u4f7f\u7528\u8fd9\u4e2a\u811a\u672c\u524d\u8bf7\u786e\u8ba4\u5df2\u7ecf\u5b89\u88c5\u4e86pillow\u53ca\u76f8\u5173\u4f9d\u8d56\u6a21\u5757":81,"\u4f7f\u7528\u8fd9\u79cd\u65b9\u5f0f":56,"\u4f7f\u7528\u8fdc\u7a0b\u7a00\u758f\u65b9\u5f0f\u8bad\u7ec3\u65f6":62,"\u4f7f\u7528\u968f\u673a\u68af\u5ea6\u4e0b\u964d":88,"\u4f7f\u7528\u9759\u6001\u5e93\u548c\u52a8\u6001\u5e93\u96be\u5ea6\u5dee\u4e0d\u591a":33,"\u4f7f\u7528\u9884\u8bad\u7ec3\u7684\u6807\u51c6\u683c\u5f0f\u8bcd\u5411\u91cf\u6a21\u578b":80,"\u4f7f\u7528args\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u8bbe\u7f6e":2,"\u4f7f\u7528c":34,"\u4f7f\u7528c99\u505a\u63a5\u53e3":33,"\u4f7f\u7528c99\u800c\u4e0d\u4f7f\u7528c11\u7684\u539f\u56e0\u662f":33,"\u4f7f\u7528c99\u800c\u4e0d\u4f7f\u7528c89":33,"\u4f7f\u7528checkgrad\u6a21\u5f0f\u65f6\u7684\u53c2\u6570\u53d8\u5316\u5927\u5c0f":70,"\u4f7f\u7528cpu\u4e24\u7ebf\u7a0b\u8ba1\u7b97fc4\u5c42":72,"\u4f7f\u7528cpu\u8ba1\u7b97fc4\u5c42":72,"\u4f7f\u7528cpu\u8bad\u7ec3":88,"\u4f7f\u7528docker\u5c31\u4e0d\u7528\u914d\u7f6e\u4ea4\u53c9\u7f16\u8bd1\u73af\u5883\u4e86":60,"\u4f7f\u7528docker\u6784\u5efapaddlepaddle\u7684\u6587\u6863":65,"\u4f7f\u7528init":72,"\u4f7f\u7528lstm\u4f5c\u4e3aencod":56,"\u4f7f\u7528max":81,"\u4f7f\u7528memory\u7684rnn\u5b9e\u73b0\u4fbf\u5982\u4e0b\u56fe\u6240\u793a":56,"\u4f7f\u7528model":72,"\u4f7f\u7528paddlepaddl":84,"\u4f7f\u7528python\u6570\u636e\u63d0\u4f9b\u5668":81,"\u4f7f\u7528rdma\u8fd8\u662ftcp\u4f20\u8f93\u534f\u8bae":70,"\u4f7f\u7528regress":42,"\u4f7f\u7528swig\u53ea\u652f\u6301cpython\u89e3\u91ca\u5668":33,"\u4f7f\u7528swig\u9700\u8981\u591a\u8bed\u8a00\u7ed1\u5b9a\u7684\u5f00\u53d1\u4eba\u5458\u719f\u7ec3\u638c\u63e1swig\u914d\u7f6e":33,"\u4f7f\u7528void":33,"\u4f7f\u8be5\u5c42\u7684\u53c2\u6570\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u4fdd\u6301\u4e0d\u53d8":46,"\u4f7f\u8f93\u5165\u5c42\u5230\u9690\u85cf\u5c42\u7684\u795e\u7ecf\u5143\u662f\u5168\u90e8\u8fde\u63a5\u7684":81,"\u4f86":56,"\u4f8b\u5982":[2,21,33,34,42,46,48,51,56,59,62,67,68,69,70,72,75,82,84,86,88],"\u4f8b\u5982\u4e0a\u6587\u7684pod":73,"\u4f8b\u5982\u4e0a\u9762\u7684":47,"\u4f8b\u5982\u4ee5\u592a\u7f51\u7684":68,"\u4f8b\u5982\u4f7f\u7528":46,"\u4f8b\u5982\u5bf9\u4e8ejava\u6216\u8005python":33,"\u4f8b\u5982\u5bf9\u4e8ejava\u6765\u8bf4":33,"\u4f8b\u5982\u5bf9\u4e8epython":33,"\u4f8b\u5982\u5c06\u7b2c\u4e00\u6761\u6570\u636e\u8f6c\u5316\u4e3a":56,"\u4f8b\u5982\u6587\u672c\u5206\u7c7b\u4e2d":56,"\u4f8b\u5982\u672c\u4f8b\u4e2d\u7684\u4e24\u4e2a\u7279\u5f81":56,"\u4f8b\u5982\u673a\u5668\u4e0a\u67094\u5757gpu":46,"\u4f8b\u5982\u7b2c300\u4e2apass\u7684\u6a21\u578b\u4f1a\u88ab\u4fdd\u5b58\u5728":81,"\u4f8b\u5982c":33,"\u4f8b\u5982hostpath":73,"\u4f8b\u5982java\u4e0epython\u7684\u9519\u8bef\u5904\u7406\u662f\u76f4\u63a5\u6254\u51fa\u6765except":33,"\u4f8b\u5982output\u76ee\u5f55\u4e0b\u5c31\u5b58\u653e\u4e86\u8f93\u51fa\u7ed3\u679c":75,"\u4f8b\u5982python\u53ef\u4ee5\u4f7f\u7528":33,"\u4f8b\u5982python\u7684":33,"\u4f8b\u5982sigmoid":62,"\u4f8b\u5982sigmoid\u53d8\u6362":84,"\u4f8b\u5b50\u4e2d\u662f":62,"\u4f8b\u5b50\u4e2d\u662f0":62,"\u4f8b\u5b50\u4e2d\u662f100":62,"\u4f8b\u5b50\u4e2d\u662f4096":62,"\u4f8b\u5b50\u4e2d\u662f8192":62,"\u4f8b\u5b50\u4e2d\u662ffc":62,"\u4f8b\u5b50\u4e2d\u662fsoftmax":62,"\u4f8b\u5b50\u4f7f\u7528":73,"\u4f9bpaddlepaddle\u52a0\u8f7d":70,"\u4f9d\u636e\u5206\u7c7b\u9519\u8bef\u7387\u83b7\u5f97\u6700\u4f73\u6a21\u578b\u8fdb\u884c\u6d4b\u8bd5":88,"\u4f9d\u636e\u662f\u5426\u5305\u542bkernel":63,"\u4f9d\u6b21\u7c7b\u63a8":42,"\u4f9d\u8d56\u4e8epython\u7684":81,"\u4fbf\u4e8e\u5b58\u50a8\u8d44\u6e90\u7ba1\u7406\u548cpod\u5f15\u7528":73,"\u4fbf\u4e8e\u672c\u5730\u9a8c\u8bc1\u548c\u6d4b\u8bd5":73,"\u4fbf\u4e8e\u7528\u6237\u6d4f\u89c8c":49,"\u4fbf\u5229":56,"\u4fbf\u548c\u5355\u5c42rnn\u914d\u7f6e\u4e2d\u7684":56,"\u4fbf\u5b9c":56,"\u4fbf\u662f\u5c06\u9759\u6001\u5e93\u52a0\u5165jvm\u4e2d":33,"\u4fdd\u5b58\u6a21\u578b\u53c2\u6570\u7684\u76ee\u5f55":70,"\u4fdd\u5b58\u751f\u6210\u7ed3\u679c\u7684\u6587\u4ef6":89,"\u4fdd\u5b58\u7684\u53c2\u6570\u4e5f\u662ffloat\u7c7b\u578b":46,"\u4fdd\u5b58\u7f51\u7edc\u5c42\u8f93\u51fa\u7ed3\u679c\u7684\u76ee\u5f55":70,"\u4fdd\u5b58\u9884\u6d4b\u7ed3\u679c\u7684\u6587\u4ef6\u540d":70,"\u4fdd\u6301\u5bbd\u9ad8\u6bd4\u7f29\u653e\u5230\u77ed\u8fb9\u4e3a256":82,"\u4fdd\u6301\u5c3d\u91cf\u5c11\u7684commit":61,"\u4fe1\u53f7\u6765\u81ea\u52a8\u7ec8\u6b62\u5b83\u542f\u52a8\u7684\u6240\u6709\u8fdb\u7a0b":68,"\u4fee\u590d\u6240\u6709bug\u540e":42,"\u4fee\u590ddocker\u7f16\u8bd1\u955c\u50cf\u95ee\u9898":42,"\u4fee\u590dubuntu":42,"\u4fee\u6539":[73,74],"\u4fee\u6539\u542f\u52a8\u811a\u672c\u540e":74,"\u4fee\u6539\u6210":42,"\u4fee\u6539\u6210\u66f4\u5feb\u7684\u7248\u672c":67,"\u4fee\u6539\u6587\u6863":66,"\u503c\u5f97\u6ce8\u610f\u7684\u662f":[56,61],"\u503c\u5f97\u6df1\u5165\u5206\u6790":67,"\u503c\u7c7b\u578b":72,"\u5047\u5982\u6211\u4eec\u662f\u4e09\u5206\u7c7b\u95ee\u9898":46,"\u5047\u8bbe":62,"\u5047\u8bbe\u53d8\u91cf":47,"\u5047\u8bbe\u60a8\u5df2\u7ecf\u5b8c\u6210\u4e86\u4e00\u4e2aai\u8bad\u7ec3\u7684python\u7a0b\u5e8f":49,"\u5047\u8bbe\u635f\u5931\u51fd\u6570\u662f":62,"\u5047\u8bbe\u7b2c\u4e00\u4e2alayer\u7684\u8f93\u51faa\u662f\u4e00\u4e2a":46,"\u5047\u8bbe\u8bcd\u5411\u91cf\u7ef4\u5ea6\u4e3a32":80,"\u504f\u7f6e\u53c2\u6570":82,"\u504f\u7f6e\u53c2\u6570\u7684\u5927\u5c0f":62,"\u505a\u4e00\u4e2a\u4ecb\u7ecd":64,"\u505a\u53ea\u8bfb\u6302\u8f7d":21,"\u505a\u5982\u4e0b\u51e0\u4e2a\u64cd\u4f5c":42,"\u505a\u63a5\u53e3":33,"\u505a\u68af\u5ea6\u68c0\u6d4b":63,"\u505a\u68c0\u67e5":63,"\u505c\u6b62\u4fdd\u5b58\u68c0\u67e5\u70b9\u7684\u7ebf\u7a0b":20,"\u505c\u6b62\u52a0\u8f7d\u6570\u636e":70,"\u505c\u7535":56,"\u50cf":49,"\u513f\u7ae5\u7247":85,"\u5143\u7d20":55,"\u5143\u7d20\u4e4b\u95f4\u7684\u987a\u5e8f\u662f\u91cd\u8981\u7684\u8f93\u5165\u4fe1\u606f":55,"\u5148\u5b9e\u73b0\u6a21\u578b\u63a8\u65ad\u7684api":34,"\u5148\u627e\u51fa\u53c2\u6570":46,"\u5148\u67e5\u770b\u4e00\u4e0b\u662f\u5426\u66fe\u7ecf\u5b89\u88c5\u8fc7paddl":46,"\u5148\u8c03\u7528initializer\u51fd\u6570":84,"\u5168\u5bb6":56,"\u5168\u8fde\u63a5\u5c42":[47,80,81,86],"\u5168\u8fde\u63a5\u5c42\u4ee5\u4e00\u4e2a\u7ef4\u5ea6\u4e3a":62,"\u5168\u8fde\u63a5\u5c42\u5c06\u7535\u5f71\u7684\u6bcf\u4e2a\u7279\u5f81\u7ed3\u5408\u6210\u4e00\u4e2a\u7535\u5f71\u7279\u5f81":86,"\u5168\u8fde\u63a5\u5c42\u6743\u91cd":82,"\u5168\u8fde\u63a5\u5c42\u6ca1\u6709\u7f51\u7edc\u5c42\u914d\u7f6e\u7684\u8d85\u53c2\u6570":62,"\u5168\u8fde\u63a5\u5c42\u7684\u5b9e\u73b0\u4f4d\u4e8e":62,"\u5168\u8fde\u63a5\u5c42\u7684\u6bcf\u4e2a\u8f93\u51fa\u90fd\u8fde\u63a5\u5230\u4e0a\u4e00\u5c42\u7684\u6240\u6709\u7684\u795e\u7ecf\u5143\u4e0a":62,"\u5168\u8fde\u63a5\u5c42python\u5c01\u88c5\u7684\u4f8b\u5b50\u4e2d\u5305\u542b\u4e0b\u9762\u51e0\u6b65":62,"\u516b\u4e2a\u7279\u5f81\u5206\u522b\u8f6c\u6362\u4e3a\u5411\u91cf":87,"\u516c\u5f0f":49,"\u516c\u94a5\u5199\u5165":68,"\u516d\u4e2a\u7279\u5f81\u548c\u6807\u7b7e\u90fd\u662f\u7d22\u5f15\u69fd":87,"\u5171\u4eab\u4e00\u4e2aop\u5b9a\u4e49":63,"\u5171\u4eab\u4efb\u52a1\u4e2d\u8bbe\u7f6e\u7684\u6570\u636e\u8fdb\u884c\u8bad\u7ec3\u548c\u6d4b\u8bd5":87,"\u5171\u4eab\u540c\u4e00\u4e2akernel\u65f6":63,"\u5171\u4eab\u5b58\u50a8\u6302\u5728\u7684\u8def\u5f84":75,"\u5171\u670932":80,"\u5173\u4e8e\u5728paddlepaddle\u4e2d\u5982\u4f55\u4f7f\u7528eigen\u5e93":63,"\u5173\u4e8e\u5982\u4f55\u5b9a\u4e49\u7f51\u7edc\u4e2d\u7684\u5c42":81,"\u5173\u4e8e\u65f6\u95f4\u5e8f\u5217":56,"\u5173\u4e8e\u6784\u5efa\u548c\u6d4b\u8bd5\u7684\u66f4\u591a\u4fe1\u606f":61,"\u5173\u4e8eeigen":64,"\u5173\u4e8elstm":46,"\u5173\u4e8epaddlepaddle\u7684\u5206\u5e03\u5f0f\u8bad\u7ec3":75,"\u5173\u4e8epaddlepaddle\u7684\u66f4\u591a\u4f7f\u7528\u65b9\u6cd5\u8bf7\u53c2\u8003":51,"\u5173\u4e8eunbound":58,"\u5173\u4e8evgg\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u7684\u63cf\u8ff0\u53ef\u4ee5\u53c2\u8003":81,"\u5173\u952e\u8bcd\u5305\u62ec":61,"\u5176\u4e0b\u5b50\u6587\u4ef6\u5939\u7684\u7ed3\u6784\u5982\u4e0b":81,"\u5176\u4e2d":[2,33,42,46,47,51,53,54,59,62,80,81,82],"\u5176\u4e2d156\u548c285\u662f\u8fd9\u4e9b\u56fe\u50cf\u7684\u5206\u7c7b\u6807\u7b7e":82,"\u5176\u4e2d50000\u5f20\u56fe\u7247\u4f5c\u4e3a\u8bad\u7ec3\u96c6":81,"\u5176\u4e2d\u5206\u522b\u5305\u542b\u4e86cifar":81,"\u5176\u4e2d\u5305\u542b6":85,"\u5176\u4e2d\u5305\u542b\u4e86200\u79cd\u9e1f\u7c7b\u7684\u7167\u7247":81,"\u5176\u4e2d\u5305\u542b\u7b97\u6cd5\u548c\u7f51\u7edc\u914d\u7f6e":88,"\u5176\u4e2d\u5305\u62ec\u51fd\u6570":87,"\u5176\u4e2d\u5b9a\u4e49\u4e86\u6a21\u578b\u67b6\u6784\u548csolver\u914d\u7f6e":89,"\u5176\u4e2d\u6587\u672c\u8f93\u5165\u7c7b\u578b\u5b9a\u4e49\u4e3a\u6574\u6570\u65f6\u5e8f\u7c7b\u578binteger_value_sequ":84,"\u5176\u4e2d\u6bcf\u4e00\u884c\u5bf9\u5e94\u4e00\u4e2a\u6570\u636e\u6587\u4ef6\u5730\u5740":1,"\u5176\u4e2d\u6bcf\u4e2a\u5143\u7d20\u662f\u53cc\u5c42\u5e8f\u5217\u4e2d\u6bcf\u4e2asubseq\u6700\u540e\u4e00\u4e2a":55,"\u5176\u4e2d\u6bcf\u4e2a\u5411\u91cf\u5bf9\u5e94\u8f93\u5165\u8bed\u53e5\u4e2d\u7684\u4e00\u4e2a\u5143\u7d20":89,"\u5176\u4e2d\u6bcf\u6761pass\u82b1\u8d39\u4e867\u4e2a\u5c0f\u65f6":89,"\u5176\u4e2d\u6bcf\u884c\u6570\u636e\u4ee3\u8868\u4e00\u5f20\u56fe\u7247":2,"\u5176\u4e2d\u8be6\u7ec6\u8bf4\u660e\u4e86\u6a21\u578b\u67b6\u6784":89,"\u5176\u4e2d\u8f93\u5165\u56fe\u50cf\u7684\u989c\u8272\u901a\u9053\u987a\u5e8f\u4e3a":82,"\u5176\u4e2dbeam":89,"\u5176\u4e2dcheckgrad\u4e3b\u8981\u4e3a\u5f00\u53d1\u8005\u4f7f\u7528":70,"\u5176\u4e2dmean\u548cstd\u662f\u8bad\u7ec3\u914d\u7f6e\u4e2d\u7684\u53c2\u6570":70,"\u5176\u4e2dvalue\u5373\u4e3asoftmax\u5c42\u7684\u8f93\u51fa":4,"\u5176\u4e2dx\u8868\u793a\u8f93\u5165\u6570\u636e\u662f\u4e00\u4e2a\u7ef4\u5ea6\u4e3a2\u7684\u7a20\u5bc6\u5411\u91cf":51,"\u5176\u4e3b\u8981\u63a5\u53e3\u5982\u4e0b":64,"\u5176\u4ed6":85,"\u5176\u4ed6\u4eba\u53ef\u4ee5\u590d\u73b0\u95ee\u9898\u4ee5\u4fbf\u5e2e\u52a9":60,"\u5176\u4ed6\u4eba\u5f88\u5bb9\u6613\u770b\u61c2\u955c\u50cf\u751f\u6210\u8fc7\u7a0b":49,"\u5176\u4ed6\u516d\u884c\u5217\u51fa\u4e86\u96c6\u675f\u641c\u7d22\u7684\u7ed3\u679c":89,"\u5176\u4ed6\u5185\u5b58\u6742\u9879":46,"\u5176\u4ed6\u5185\u5b58\u6742\u9879\u662f\u6307paddlepaddle\u672c\u8eab\u6240\u7528\u7684\u4e00\u4e9b\u5185\u5b58":46,"\u5176\u4ed6\u51fd\u6570\u5747\u8fd4\u56de":34,"\u5176\u4ed6\u53c2\u6570\u4f7f\u7528":2,"\u5176\u4ed6\u53c2\u6570\u8bf7\u53c2\u8003":84,"\u5176\u4ed6\u6240\u6709\u5c42\u90fd\u4f1a\u4f7f\u7528gpu\u8ba1\u7b97":72,"\u5176\u4ed6\u7528\u6237\u7684fork\u7248\u672c\u5e93\u5e76\u4e0d\u9700\u8981\u4e25\u683c\u9075\u5b88":42,"\u5176\u4ed6\u914d\u7f6e\u53c2\u6570":[53,54],"\u5176\u4ed6\u9ad8\u7ea7\u529f\u80fd\u5305\u62ec\u5b9a\u4e49\u591a\u4e2amemori":59,"\u5176\u4f1a\u81ea\u52a8\u88ab\u52a0\u5165\u7f16\u8bd1\u5217\u8868":62,"\u5176\u4f59\u884c\u662f":80,"\u5176\u4f5c\u7528\u662f\u5c06\u6570\u636e\u4f20\u5165\u5185\u5b58\u6216\u663e\u5b58":1,"\u5176\u5177\u4f53\u8bf4\u660e\u4e86\u5b57\u6bb5\u7c7b\u578b\u548c\u6587\u4ef6\u540d\u79f0":86,"\u5176\u5185\u90e8\u7684\u6587\u4ef6\u4e5f\u4f1a\u968f\u4e4b\u6d88\u5931":73,"\u5176\u5305\u62ec\u4e24\u4e2a\u51fd\u6570":84,"\u5176\u53c2\u6570\u5982\u4e0b":2,"\u5176\u5b83\u90e8\u5206\u548c\u903b\u8f91\u56de\u5f52\u7f51\u7edc\u7ed3\u6784\u4e00\u81f4":84,"\u5176\u5b83layer\u7684\u8f93\u51fa":58,"\u5176\u5b9e\u4e5f\u662f\u548c\u6bcf\u4e2amini":46,"\u5176\u63d0\u4f9b\u5e94\u7528\u90e8\u7f72":73,"\u5176\u6b21":[2,56,84],"\u5176\u76ee\u7684\u662f\u5728\u7ed9\u5b9a\u7684\u8f93\u5165\u53e5\u5b50\u4e2d\u53d1\u73b0\u6bcf\u4e2a\u8c13\u8bcd\u7684\u8c13\u8bcd\u8bba\u5143\u7ed3\u6784":87,"\u5176\u8bf4\u660e\u5982\u4e0b":56,"\u5176\u8f93\u5165\u53c2\u6570\u5982\u4e0b":89,"\u5176\u8f93\u51fa\u88ab\u7528\u4f5cmemory\u7684\u521d\u59cb\u503c":59,"\u5176name\u7531\u53c2\u6570":46,"\u5177\u4f53\u4f7f\u7528\u65b9\u6cd5\u4e3a":[34,46],"\u5177\u4f53\u505a\u6cd5\u8bf7\u53c2\u8003":60,"\u5177\u4f53\u539f\u56e0\u53c2\u8003":34,"\u5177\u4f53\u53ef\u4ee5\u53c2\u8003":[2,46,62],"\u5177\u4f53\u53ef\u53c2\u8003\u6587\u6863":58,"\u5177\u4f53\u5982\u4e0b":49,"\u5177\u4f53\u60c5\u51b5\u56e0\u4eba\u800c\u5f02":67,"\u5177\u4f53\u64cd\u4f5c\u5982\u4e0b":46,"\u5177\u4f53\u6d41\u7a0b\u5982\u4e0b":84,"\u5177\u4f53\u7684\u683c\u5f0f\u8bf4\u660e":2,"\u5177\u4f53\u7684\u89e3\u51b3\u65b9\u6cd5\u662f":46,"\u5177\u4f53\u8bf7\u53c2\u7167\u793a\u4f8b":82,"\u5177\u4f53\u8bf7\u53c2\u8003":[2,34,61],"\u5177\u4f53\u8bf7\u89c1":61,"\u5177\u6709\u76f8\u540c\u7684\u7ed3\u679c\u4e86":56,"\u5177\u6709\u81ea\u5faa\u73af\u8fde\u63a5\u7684\u795e\u7ecf\u5143":88,"\u5185":59,"\u5185\u5b58":67,"\u5185\u5b58\u4e0d\u8db3":46,"\u5185\u5b58\u5bb9\u9650\u9608\u503c":70,"\u5185\u5bb9":[63,84],"\u5185\u5bb9\u5982\u4e0b":74,"\u5185\u5c42inner_step\u7684recurrent_group\u548c\u5355\u5c42\u5e8f\u5217\u7684\u51e0\u4e4e\u4e00\u6837":56,"\u5185\u5df2\u7ecf\u5305\u542bpaddlepaddle\u7684\u6267\u884c\u7a0b\u5e8f\u4f46\u662f\u8fd8\u6ca1\u4e0a\u8ff0\u529f\u80fd":75,"\u5185\u90e8":75,"\u5185\u90e8\u9a71\u52a8python\u89e3\u91ca\u5668\u8fdb\u884c\u6a21\u578b\u914d\u7f6e\u89e3\u6790\u548c\u6570\u636e\u8bfb\u53d6":33,"\u518d\u4e3apaddle\u7684\u8bad\u7ec3\u8fc7\u7a0b\u63d0\u4f9b\u6587\u4ef6\u5217\u8868":86,"\u518d\u4ee5":63,"\u518d\u5199\u5165\u7f51\u7edc\u53c2\u6570":46,"\u518d\u5728\u6bcf\u4e00\u4e2aapi\u4e2d\u81ea\u5df1\u68c0\u67e5\u7c7b\u578b":33,"\u518d\u57fa\u4e8e":42,"\u518d\u5b89\u88c5":46,"\u518d\u5bf9\u6bcf\u4e00\u4e2a\u5355\u5c42\u65f6\u95f4\u5e8f\u5217\u8fdb\u884c\u5904\u7406":56,"\u518d\u5bf9\u6bcf\u4e00\u53e5\u8bdd\u7684\u7f16\u7801\u5411\u91cf\u7528lstm\u7f16\u7801\u6210\u4e00\u4e2a\u6bb5\u843d\u7684\u5411\u91cf":56,"\u518d\u5bf9\u8fd9\u4e2a\u6bb5\u843d\u5411\u91cf\u8fdb\u884c\u5206\u7c7b":56,"\u518d\u6307\u5b9a":48,"\u518d\u6b21\u5bf9\u4ee3\u7801\u8fdb\u884c\u6027\u80fd\u5206\u6790":67,"\u518d\u7528\u8fd9\u4e2a\u68af\u5ea6\u53bb\u548c":62,"\u518d\u901a\u8fc7\u51fd\u6570":75,"\u5192\u9669\u7247":85,"\u5197\u4f59\u7b49\u529f\u80fd":73,"\u5199\u4ee3\u7801":33,"\u5199\u5165\u5feb\u7167\u6570\u636e":20,"\u5199\u5b8c\u6a21\u578b\u914d\u7f6e\u4e4b\u540e":89,"\u5199\u68af\u5ea6\u68c0\u67e5\u5355\u5143\u6d4b\u8bd5\u662f\u4e00\u4e2a\u9a8c\u8bc1\u65b0\u5b9e\u73b0\u7684\u5c42\u662f\u5426\u6b63\u786e\u7684\u76f8\u5bf9\u7b80\u5355\u7684\u529e\u6cd5":62,"\u519c\u6c11":85,"\u51c6\u5907":56,"\u51c6\u5907\u597d\u6570\u636e":86,"\u51c6\u5907\u9884\u6d4b\u6570\u636e":4,"\u51cf\u5c0f\u5e8f\u5217\u7684\u957f\u5ea6":46,"\u51cf\u5c0f\u8fd9\u4e2a\u5185\u5b58\u6c60\u5373\u53ef\u51cf\u5c0f\u5185\u5b58\u5360\u7528":46,"\u51cf\u5c0fbatch":46,"\u51e0\u4e2a\u57fa\u7840\u7684\u6982\u5ff5\u5e2e\u52a9\u7406\u89e3\u548c\u4f7f\u7528docker":49,"\u51fa\u53bb\u73a9":56,"\u51fa\u5dee":56,"\u51fa\u6765":56,"\u51fa\u73b0\u4ee5\u4e0b\u9519\u8bef":46,"\u51fa\u73b0\u8be5\u9519\u8bef\u7684\u539f\u56e0\u4e00\u822c\u662f\u7528\u6237\u5bf9\u4e0d\u540clayer\u7684\u53c2\u6570":46,"\u51fa\u73b0\u8fd9\u4e2a\u95ee\u9898\u7684\u4e3b\u8981\u539f\u56e0\u662f":46,"\u51fd\u6570":[2,47,59,62,67,87,88],"\u51fd\u6570\u4e2d\u4f7f\u7528":2,"\u51fd\u6570\u4e2d\u64cd\u4f5c\u7684\u91cd\u8981\u53d8\u91cf\u7684\u8be6\u7ec6\u89e3\u91ca":63,"\u51fd\u6570\u4e2d\u8bbe\u7f6e\u7684":68,"\u51fd\u6570\u5047\u8bbe":59,"\u51fd\u6570\u52a0\u5230\u4ee3\u7801\u4e2d":67,"\u51fd\u6570\u5373\u53ef\u5b8c\u6210\u8f6c\u6362":21,"\u51fd\u6570\u53ea\u5173\u6ce8\u4e8ernn\u4e00\u4e2a\u65f6\u95f4\u6b65\u4e4b\u5185\u7684\u8ba1\u7b97":58,"\u51fd\u6570\u540d\u4e3a":34,"\u51fd\u6570\u547d\u540d":33,"\u51fd\u6570\u5b9a\u4e49\u8f93\u5165":63,"\u51fd\u6570\u5bf9\u6bd4cpu":63,"\u51fd\u6570\u5c31\u662f\u6839\u636e\u8be5\u673a\u5236\u914d\u7f6e\u7684":2,"\u51fd\u6570\u5f97\u5230\u7684\u68af\u5ea6\u53bb\u5bf9\u6bd4":62,"\u51fd\u6570\u5fc5\u987b\u5148\u8c03\u7528\u57fa\u7c7b\u4e2d\u7684\u51fd\u6570":62,"\u51fd\u6570\u5fc5\u987b\u8fd4\u56de\u4e00\u4e2a\u6216\u591a\u4e2alayer\u7684\u8f93\u51fa":58,"\u51fd\u6570\u6307\u51fa\u4e86\u5728\u8bad\u7ec3\u65f6\u9700\u8981\u4ece\u53c2\u6570\u670d\u52a1\u5668\u53d6\u51fa\u7684\u884c":62,"\u51fd\u6570\u6765\u5c06\u4fe1\u606f\u8f93\u51fa\u5230\u754c\u9762\u4e2d":67,"\u51fd\u6570\u67e5\u8be2\u8f6f\u4ef6\u5305\u76f8\u5173api\u8bf4\u660e":4,"\u51fd\u6570\u7684":2,"\u51fd\u6570\u7684\u5b9e\u73b0\u662f\u6b63\u786e\u7684":62,"\u51fd\u6570\u7684\u5f00\u5934\u5fc5\u987b\u8c03\u7528":62,"\u51fd\u6570\u91cc\u5b9e\u73b0":63,"\u5206\u4e3a\u597d\u8bc4":84,"\u5206\u522b\u4e3a":80,"\u5206\u522b\u4e3atrain":89,"\u5206\u522b\u4ece\u8bcd\u8bed\u548c\u53e5\u5b50\u7ea7\u522b\u7f16\u7801\u8f93\u5165\u6570\u636e":58,"\u5206\u522b\u4f7f\u7528\u5355\u53cc\u5c42rnn\u4f5c\u4e3a\u7f51\u7edc\u914d\u7f6e\u7684\u6a21\u578b":56,"\u5206\u522b\u5305\u542b\u4e86\u6cd5\u8bed\u5230\u82f1\u8bed\u7684\u5e73\u884c\u8bed\u6599\u5e93\u7684\u8bad\u7ec3\u6570\u636e":89,"\u5206\u522b\u5b9a\u4e49\u5b50\u53e5\u7ea7\u522b\u548c\u8bcd\u8bed\u7ea7\u522b\u4e0a\u9700\u8981\u5b8c\u6210\u7684\u8fd0\u7b97":58,"\u5206\u522b\u5bf9\u5e94\u4e8e\u53d8\u91cf":47,"\u5206\u522b\u662f":55,"\u5206\u522b\u662frnn\u72b6\u6001\u548c\u8f93\u5165\u7684\u53d8\u6362\u77e9\u9635":59,"\u5206\u522b\u662fsentences\u548clabel":56,"\u5206\u522b\u662fwords\u548clabel":56,"\u5206\u522b\u8ba1\u7b97\u6bcf\u4e2a\u53c2\u6570\u7684\u68af\u5ea6":62,"\u5206\u522b\u8fdb\u884c\u5e8f\u5217\u64cd\u4f5c":56,"\u5206\u5272":[85,87],"\u5206\u5272\u6587\u4ef6\u7684\u65b9\u6cd5\u662f":86,"\u5206\u5e03\u5f0f\u5b58\u50a8\u670d\u52a1":20,"\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf":73,"\u5206\u6210\u4e24\u90e8\u5206":2,"\u5206\u652f":[42,61],"\u5206\u652f\u4e00\u65e6\u5efa\u7acb":42,"\u5206\u652f\u4e0a":61,"\u5206\u652f\u4e0a\u521b\u5efa\u65b0\u5206\u652f":61,"\u5206\u652f\u4e2d":42,"\u5206\u652f\u4e3a\u5f00\u53d1":42,"\u5206\u652f\u4e3a\u6bcf\u4e00\u6b21release\u65f6\u5efa\u7acb\u7684\u4e34\u65f6\u5206\u652f":42,"\u5206\u652f\u4e3a\u7a33\u5b9a":42,"\u5206\u652f\u529f\u80fd\u7684\u5c01\u95ed":42,"\u5206\u652f\u5408\u5165":42,"\u5206\u652f\u5408\u5165master\u5206\u652f":42,"\u5206\u652f\u540c\u6b65\u4e3b\u7248\u672c\u5e93\u7684":42,"\u5206\u652f\u540d":61,"\u5206\u652f\u540d\u4e3a":42,"\u5206\u652f\u5b58\u5728\u7684\u65f6\u5019":42,"\u5206\u652f\u6d3e\u751f\u51fa\u65b0\u7684\u5206\u652f":42,"\u5206\u652f\u7528\u6765\u6d4b\u8bd5\u53ea\u9700\u8981\u8ba1\u7b97\u4e00\u4e2a\u8f93\u5165\u68af\u5ea6\u7684\u60c5\u51b5":63,"\u5206\u652f\u7684\u7248\u672c\u90fd\u662f\u7ecf\u8fc7\u5355\u5143\u6d4b\u8bd5\u548c\u56de\u5f52\u6d4b\u8bd5\u7684\u7248\u672c":42,"\u5206\u652f\u7684\u7248\u672c\u90fd\u7ecf\u8fc7\u5355\u5143\u6d4b\u8bd5":42,"\u5206\u652f\u7684\u76ee\u6807\u673a\u7248openblas\u5e93":53,"\u5206\u652f\u89c4\u8303":61,"\u5206\u6790\u5f97\u5230\u7684\u4fe1\u606f\u7528\u4e8e\u534f\u52a9\u8fdb\u884c\u7a0b\u5e8f\u7684\u4f18\u5316":67,"\u5206\u7247":20,"\u5206\u7c7b\u6210\u6b63\u9762\u60c5\u7eea\u548c\u8d1f\u9762\u60c5\u7eea\u4e24\u7c7b":2,"\u5206\u7c7b\u8bef\u5dee\u662f0":88,"\u5206\u7c7b\u9519\u8bef\u7387\u548c\u6a21\u578b\u5927\u5c0f\u7531\u4e0b\u8868\u7ed9\u51fa":82,"\u5206\u8bcd\u5e8f\u5217\u7684\u5f00\u59cb":80,"\u5206\u8bcd\u5e8f\u5217\u7684\u7ed3\u675f":80,"\u5206\u8bcd\u98ce\u683c\u5982\u4e0b":80,"\u5206\u914d\u5230\u5f53\u524d\u6570\u636e\u5757\u6837\u672c\u6570\u7684\u56db\u5206\u4e4b\u4e00":70,"\u5206\u9694":[80,86],"\u5206\u9694\u7b26\u4e3a":85,"\u5207\u6362\u5230":61,"\u5207\u6362\u5230\u6240\u5efa\u5206\u652f":61,"\u5217\u8868":86,"\u5217\u8868\u5982\u4e0b":[2,51],"\u5219\u4e0d\u5728\u4e4e\u5185\u5b58\u6682\u5b58\u591a\u5c11\u6761\u6570\u636e":2,"\u5219\u4e0d\u9700\u8981\u91cd\u5199\u8be5\u51fd\u6570":62,"\u5219\u4f1a\u9884\u5148\u8bfb\u53d6\u5168\u90e8\u6570\u636e\u5230\u5185\u5b58\u4e2d":2,"\u5219\u4f7f\u7528\u533a\u57df\u6807\u8bb0":87,"\u5219\u4f7f\u7528\u540c\u6b65\u8bad\u7ec3":70,"\u5219\u4f7f\u7528\u542f\u52a8\u53c2\u6570\u5b9a\u4e49\u7684\u521d\u59cb\u5316\u65b9\u6cd5\u521d\u59cb\u5316\u53c2\u6570":20,"\u5219\u4f7f\u7528\u8be5\u53c2\u6570\u4f5c\u4e3a\u9ed8\u8ba4\u503c":70,"\u5219\u5148\u505a\u5d4c\u5165":86,"\u5219\u53ef\u4ee5\u50cf":68,"\u5219\u53ef\u8bbe\u7f6e":[53,54],"\u5219\u5b57\u4e0e\u5b57\u4e4b\u95f4\u7528\u7a7a\u683c\u5206\u9694":84,"\u5219\u5ffd\u7565":20,"\u5219\u603b\u4f1a\u663e\u793a\u963b\u9694\u6458\u8981\u4fe1\u606f":70,"\u5219\u628a\u53e6\u4e00\u4e2a\u6162\u901f\u7684kill\u6389":20,"\u5219\u63a8\u8350\u5927\u4e8e\u8bad\u7ec3\u65f6batch":2,"\u5219\u662f\u5e26gui\u7684nvidia\u53ef\u89c6\u5316\u6027\u80fd\u5206\u6790\u5de5\u5177":67,"\u5219\u663e\u793a\u963b\u9694\u6027\u80fd\u7684\u6458\u8981\u4fe1\u606f":70,"\u5219\u76f4\u63a5\u5f15\u5165\u53e6\u4e00\u79cd\u7c7b\u578b\u7684\u5934\u6587\u4ef6":34,"\u5219\u8bbe\u7f6e\u6210":[53,54],"\u5219\u9700\u8981\u4f7f\u7528\u7b49\u4e8e\u6743\u91cd\u53c2\u6570\u89c4\u6a21\u5927\u7ea65\u500d\u7684\u5185\u5b58":46,"\u5219\u9700\u8981\u56de\u6eda\u5230\u4e0a\u4e00\u4e2a\u68c0\u67e5\u70b9":20,"\u5219\u9700\u8981\u914d\u7f6e":73,"\u5219\u9700\u9009\u7528":54,"\u521b\u5efa\u4e00\u4e2akubernet":75,"\u521b\u5efa\u53cd\u5411op\u5bf9\u5e94\u7684\u524d\u5411op":63,"\u521b\u5efa\u548c\u53d1\u5e03\u81ea\u5df1\u7684ai\u7a0b\u5e8f\u955c\u50cf":49,"\u521b\u5efa\u5e76\u5207\u6362\u5230\u65b0\u5206\u652f":61,"\u521b\u5efa\u6210\u529f\u540e":75,"\u521b\u5efa\u8bad\u7ec3\u6570\u636e\u7684":89,"\u521b\u5efagener":2,"\u521d\u59cb\u5316\u4e4b\u540e":4,"\u521d\u59cb\u5316\u504f\u7f6e\u5411\u91cf":62,"\u521d\u59cb\u5316\u65f6\u8c03\u7528\u7684\u51fd\u6570":2,"\u521d\u59cb\u5316\u6743\u91cd\u8868":62,"\u521d\u59cb\u5316\u6a21\u578b\u7684\u8def\u5f84":70,"\u521d\u59cb\u5316\u6a21\u578b\u7684\u8def\u5f84\u914d\u7f6e\u4e3a":80,"\u521d\u59cb\u5316\u7236\u7c7b":62,"\u521d\u59cb\u5316biases_":62,"\u521d\u59cb\u5316paddlepaddle\u73af\u5883":4,"\u521d\u59cb\u72b6\u6001":58,"\u5220\u9664":61,"\u5220\u9664\u78c1\u76d8\u76ee\u5f55\u4e2d\u4e0d\u662f\u5f53\u524duuid\u7684\u5feb\u7167\u6587\u4ef6":20,"\u5229\u7528\u5206\u5e03\u5f0f\u8bad\u7ec3\u9a7e\u9a6d\u66f4\u591a\u7684\u8ba1\u7b97\u8d44\u6e90":46,"\u5229\u7528\u5355\u8bcdid\u67e5\u627e\u8be5\u5355\u8bcd\u5bf9\u5e94\u7684\u8fde\u7eed\u5411\u91cf":84,"\u5229\u7528\u66f4\u591a\u7684\u8ba1\u7b97\u8d44\u6e90\u53ef\u4ee5\u5206\u4e3a\u4e00\u4e0b\u51e0\u4e2a\u65b9\u5f0f\u6765\u8fdb\u884c":46,"\u5229\u7528\u6b63\u5219\u8868\u8fbe\u5f0f\u6765\u89e3\u6790\u8be5\u7279\u5f81":86,"\u5229\u7528\u8bad\u7ec3\u96c6\u751f\u6210\u7684\u5b57\u5178":88,"\u5229\u7528\u8fd9\u79cd\u7279\u6027":58,"\u5229\u7528\u903b\u8f91\u56de\u5f52\u6a21\u578b\u5bf9\u8be5\u5411\u91cf\u8fdb\u884c\u5206\u7c7b":84,"\u5229\u7528kubernetes\u80fd\u65b9\u4fbf\u5730\u7ba1\u7406\u8de8\u673a\u5668\u8fd0\u884c\u5bb9\u5668\u5316\u7684\u5e94\u7528":73,"\u5229\u843d":56,"\u522b\u4eba\u5e2e\u4e86\u5fd9":61,"\u522b\u5fd8\u4e86":60,"\u5230":[20,46,59],"\u5230\u6240\u6709\u8282\u70b9\u800c\u4e0d\u7528\u5bc6\u7801":68,"\u5230\u672c\u5730":61,"\u5230\u6b64":[49,63],"\u5230\u76ee\u524d\u4e3a\u6b62":87,"\u5236\u4f5c\u65b0\u955c\u50cf\u6765\u5b8c\u6210\u4ee5\u4e0a\u7684\u5de5\u4f5c":75,"\u5236\u4f5cpaddlepaddle\u5f00\u53d1\u955c\u50cf":49,"\u5236\u4f5cpaddlepaddle\u751f\u4ea7\u955c\u50cf":49,"\u5236\u4f5cpaddlepaddle\u955c\u50cf":75,"\u5237\u7259":56,"\u524d\u4e00\u7bc7\u6587\u7ae0\u4ecb\u7ecd\u4e86\u5982\u4f55\u5728kubernetes\u96c6\u7fa4\u4e0a\u542f\u52a8\u4e00\u4e2a\u5355\u673apaddlepaddle\u8bad\u7ec3\u4f5c\u4e1a":75,"\u524d\u4e09\u884cimport\u4e86\u5b9a\u4e49network":89,"\u524d\u53f0":56,"\u524d\u5411\u4f20\u64ad":62,"\u524d\u5411\u4f20\u64ad\u7ed9\u5b9a\u8f93\u5165":62,"\u524d\u5411\u548c\u540e\u5411":62,"\u524d\u5411op":63,"\u524d\u5411op\u5355\u5143\u6d4b\u8bd5\u7ee7\u627f\u81ea":63,"\u524d\u5411op\u5b9e\u73b0\u5b8c\u6210":63,"\u524d\u5411operator\u5355\u6d4b":63,"\u524d\u8005\u5728":46,"\u524d\u8005\u5b58\u50a8op\u7684\u8f93\u5165\u8f93\u51fa\u548c\u53c2\u6570\u5c5e\u6027":63,"\u524d\u8005\u622a\u65ad\u53ef\u5b66\u4e60\u53c2\u6570\u7684\u68af\u5ea6":46,"\u524d\u8005op\u7684\u5b9a\u4e49\u7ee7\u627f\u81ea":63,"\u524d\u81ea\u52a8\u68c0\u67e5\u4e00\u4e9b\u57fa\u672c\u4e8b\u5b9c":61,"\u5269\u4e0b\u7684pass\u4f1a\u76f4\u63a5\u4ece\u5185\u5b58\u91cc":2,"\u529f\u80fd":28,"\u529f\u80fd\u7684\u6b63\u786e\u6027\u5305\u62ec\u9a8c\u8bc1paddlepaddle\u76ee\u524d\u7684":42,"\u52a0\u4e0a\u504f\u7f6e\u5411\u91cf":62,"\u52a0\u4e86l2\u6b63\u5219\u548c\u68af\u5ea6\u622a\u65ad":84,"\u52a0\u5165":67,"\u52a0\u6743\u548c\u7528\u6765\u751f\u6210":59,"\u52a0\u6743\u7f16\u7801\u5411\u91cf":59,"\u52a0\u8f7d\u5177\u4f53\u7f51\u7edc\u53c2\u6570":46,"\u52a0\u8f7d\u6570\u636e":87,"\u52a0\u8f7d\u6a21\u578b":87,"\u52a0\u8f7d\u6a21\u578b\u53c2\u6570":89,"\u52a0\u8f7d\u9884\u8bad\u7ec3\u53c2\u6570":46,"\u52a0\u8f7dtest":70,"\u52a0\u901fpaddlepaddle\u8bad\u7ec3\u53ef\u4ee5\u8003\u8651\u4ece\u4ee5\u4e0b\u51e0\u4e2a\u65b9\u9762":46,"\u52a8\u4f5c\u7247":85,"\u52a8\u6001\u5e93":33,"\u52a8\u753b\u7247":85,"\u52a8\u8bcd":87,"\u52a9\u624b":62,"\u5305\u542b12":88,"\u5305\u542b20\u4e2a\u8bad\u7ec3\u6837\u4f8b":80,"\u5305\u542b3\u4e2a\u5c5e\u6027":80,"\u5305\u542b50":88,"\u5305\u542b\u4e86\u67d0\u79cd\u7c7b\u578b\u7684\u7c7b\u578b\u5b9a\u4e49\u548c\u66b4\u9732\u7684\u5168\u90e8\u51fd\u6570":34,"\u5305\u542b\u6240\u6709\u548c\u524d\u7aef\u7f51\u9875\u8bbe\u8ba1\u76f8\u5173\u7684\u6587\u4ef6":65,"\u5305\u542b\u7684\u6240\u6709\u4f9d\u8d56\u5047\u8bbe\u90fd\u53ef\u4ee5\u5728paddle\u7684\u8fd0\u884c\u5bb9\u5668\u4e2d":49,"\u5305\u542bkernel\u7684op\u548c\u4e0d\u5305\u542bkernel\u7684op":63,"\u5305\u62ec":[21,70,84,87,89],"\u5305\u62ec\u4ee5\u4e0b\u4e24\u79cd":2,"\u5305\u62ec\u53d1\u884c\u65f6\u95f4":85,"\u5305\u62ec\u5b57\u7b26\u4e32\u5206\u914d":46,"\u5305\u62ec\u6570\u636e\u8f93\u5165":47,"\u5305\u62ec\u6743\u91cdw\u548c\u504f\u7f6eb":20,"\u5305\u62ec\u751f\u6210cpu":48,"\u5305\u62ec\u795e\u7ecf\u7f51\u7edc\u62d3\u6251\u7ed3\u6784":51,"\u5305\u62ec\u7b80\u5355\u7684":84,"\u5305\u62ecbool":72,"\u5305\u62eclinux":49,"\u5305\u7684\u65b9\u6cd5\u662f":46,"\u533a\u522b\u662f\u540c\u65f6\u5904\u7406\u4e86\u4e24\u4e2a\u8f93\u5165":56,"\u533a\u522b\u662frnn\u4f7f\u7528\u4e24\u5c42\u5e8f\u5217\u6a21\u578b":56,"\u533b\u751f":85,"\u533b\u7597\u4fdd\u5065":85,"\u5341\u4e00":56,"\u5347\u5e8f\u6392\u5217":89,"\u534e\u6da6\u4e07\u5bb6":56,"\u534f\u540c\u5b8c\u6210releas":42,"\u5355\u4e2a\u503c":21,"\u5355\u4f4d\u662fmb":70,"\u5355\u5143\u6d4b\u8bd5":[60,64],"\u5355\u5143\u6d4b\u8bd5\u4f1a\u5f15\u7528site":46,"\u5355\u5143\u6d4b\u8bd5\u4f1a\u88ab\u81ea\u52a8\u52a0\u5165\u5de5\u7a0b\u8fdb\u884c\u7f16\u8bd1":63,"\u5355\u5143\u6d4b\u8bd5\u5728\u5185\u7684\u6240\u6709\u5355\u5143\u6d4b\u8bd5":60,"\u5355\u5143\u6d4b\u8bd5checkgrad_ep":69,"\u5355\u53cc\u5c42\u5e8f\u5217\u7684\u53e5\u5b50\u662f\u4e00\u6837\u7684":56,"\u5355\u53cc\u5c42rnn":57,"\u5355\u53d8\u91cf\u7684\u7ebf\u6027\u56de\u5f52":47,"\u5355\u5c42":58,"\u5355\u5c42\u4e0d\u7b49\u957frnn":56,"\u5355\u5c42\u548c\u53cc\u5c42\u5e8f\u5217\u7684\u4f7f\u7528\u548c\u793a\u4f8b2\u4e2d\u7684\u793a\u4f8b\u7c7b\u4f3c":56,"\u5355\u5c42\u5e8f\u5217":55,"\u5355\u5c42\u5e8f\u5217\u7684\u6bcf\u4e2a\u5143\u7d20":55,"\u5355\u5c42\u5e8f\u5217\u7b2ci\u4e2a\u5143\u7d20":55,"\u5355\u5c42\u6216\u53cc\u5c42":55,"\u5355\u5c42\u65f6\u95f4\u5e8f\u5217":56,"\u5355\u5c42rnn":[56,58],"\u5355\u5c42rnn\u548c\u53cc\u5c42rnn\u7684\u7f51\u7edc\u914d\u7f6e":56,"\u5355\u673acpu\u8bad\u7ec3":46,"\u5355\u673agpu\u8bad\u7ec3":46,"\u5355\u6b65\u51fd\u6570":59,"\u5355\u6b65\u51fd\u6570\u548c\u8f93\u51fa\u51fd\u6570\u5728":59,"\u5355\u6b65\u51fd\u6570\u548c\u8f93\u51fa\u51fd\u6570\u90fd\u975e\u5e38\u7b80\u5355":59,"\u5355\u6b65\u51fd\u6570\u7684\u5b9e\u73b0\u5982\u4e0b\u6240\u793a":59,"\u5355\u6d4b\u5305\u62ec\u5bf9\u6bd4\u524d\u5411op\u4e0d\u540c\u8bbe\u5907":63,"\u5355\u70b9\u6545\u969c":20,"\u5355\u8fdb\u5355\u51fa":58,"\u5360\u4f4d\u7b26":80,"\u536b\u751f":56,"\u5373":[34,46,47,63,65,75,84,88],"\u5373\u4e00\u4e2a\u5c06\u5355\u8bcd\u5b57\u7b26\u4e32\u6620\u5c04\u5230\u5355\u8bcdid\u7684\u5b57\u5178":2,"\u5373\u4e0a\u8ff0\u4ee3\u7801\u4e2d\u7684\u7b2c19\u884c":56,"\u5373\u4e0d\u5141\u8bb8\u5728":63,"\u5373\u4e0d\u8981\u5c06\u6bcf\u4e00\u4e2a\u6837\u672c\u90fd\u653e\u5165train":2,"\u5373\u4e0d\u9700\u8981\u4f7f\u7528memori":56,"\u5373\u4e3a\u4e00\u4e2a\u65f6\u95f4\u6b65":56,"\u5373\u4e3a\u5355\u5c42rnn\u5e8f\u5217\u7684\u4f7f\u7528\u4ee3\u7801":56,"\u5373\u4e3a\u65f6\u95f4\u5e8f\u5217\u7684\u8f93\u5165":56,"\u5373\u4e3a\u8fd9\u4e2a\u53cc\u5c42rnn\u7684\u7f51\u7edc\u7ed3\u6784":56,"\u5373\u4e3a\u8fd9\u4e2a\u6570\u636e\u6587\u4ef6\u7684\u540d\u5b57":2,"\u5373\u4e8c\u7ef4\u6570\u7ec4":56,"\u5373\u4f7f\u7528":[34,46],"\u5373\u4f7f\u7528\u6237\u76f4\u63a5\u5f15\u7528\u67d0\u79cd\u7c7b\u578b\u7684\u5934\u6587\u4ef6":34,"\u5373\u4f7f\u95f4\u9694\u5f88\u5c0f":70,"\u5373\u4f7fc":34,"\u5373\u4f7fprocess\u51fd\u6570\u91cc\u9762\u53ea\u6709\u4e00\u4e2ayield":2,"\u5373\u4f8b\u5982":34,"\u5373\u4fbf\u662f":60,"\u5373\u4fbf\u8bbe\u7f6e":46,"\u5373\u4fbfpaddl":34,"\u5373\u521d\u59cb\u72b6\u6001\u4e3a0":58,"\u5373\u5305\u542b\u65f6\u95f4\u6b65\u4fe1\u606f":2,"\u5373\u5355\u65f6\u95f4\u6b65\u6267\u884c\u7684\u51fd\u6570":59,"\u5373\u53cc\u5411lstm\u548c\u4e09\u5c42\u5806\u53e0lstm":88,"\u5373\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217":56,"\u5373\u53cc\u5c42rnn\u7684\u6bcf\u4e2a\u72b6\u6001":58,"\u5373\u53ef":[47,61],"\u5373\u53ef\u4ee5\u6781\u5927\u7684\u52a0\u901f\u6570\u636e\u8f7d\u5165\u6d41\u7a0b":46,"\u5373\u5728\u53cc\u5c42\u5e8f\u5217\u7684\u539f\u59cb\u6570\u636e\u4e2d":56,"\u5373\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d":46,"\u5373\u5927\u90e8\u5206\u503c\u4e3a0":[2,51],"\u5373\u5b8c\u6210\u67d0\u4e00\u4e2a\u4efb\u52a1\u7684\u6700\u5c11\u51fd\u6570":34,"\u5373\u5bf9\u7b2c\u4e09\u6b65\u8fdb\u884c\u66ff\u6362":84,"\u5373\u5c06\u4e00\u6bb5\u82f1\u6587\u6587\u672c\u6570\u636e":2,"\u5373\u5c06\u4e00\u6bb5\u8bdd\u8fdb\u884c\u5206\u7c7b":56,"\u5373\u5f53\u524d\u65f6\u95f4\u6b65\u4e0b\u7684\u795e\u7ecf\u7f51\u7edc\u4f9d\u8d56\u524d\u4e00\u4e2a\u65f6\u95f4\u6b65\u795e\u7ecf\u7f51\u7edc\u4e2d\u67d0\u4e00\u4e2a\u795e\u7ecf\u5143\u8f93\u51fa":56,"\u5373\u6211\u4eec\u53ef\u4ee5\u5148\u5b9a\u4e49\u4e00\u4e2atensor":64,"\u5373\u6211\u4eec\u7684\u8bad\u7ec3\u76ee\u6807":47,"\u5373\u628a\u5355\u5c42rnn\u751f\u6210\u540e\u7684subseq\u7ed9\u62fc\u63a5\u6210\u4e00\u4e2a\u65b0\u7684\u53cc\u5c42seq":58,"\u5373\u6574\u4e2a\u53cc\u5c42group\u662f\u5c06\u524d\u4e00\u4e2a\u5b50\u53e5\u7684\u6700\u540e\u4e00\u4e2a\u5411\u91cf":56,"\u5373\u6574\u4e2a\u8f93\u5165\u5e8f\u5217":55,"\u5373\u6574\u6570\u6570\u7ec4":56,"\u5373\u65f6\u95f4\u9012\u5f52\u795e\u7ecf\u7f51\u7edc":56,"\u5373\u662f\u8de8\u8d8a\u65f6\u95f4\u6b65\u7684\u7f51\u7edc\u8fde\u63a5":56,"\u5373\u66b4\u9732":34,"\u5373\u6b63\u9762\u548c\u8d1f\u9762":88,"\u5373\u6b63\u9762\u8bc4\u4ef7\u6807\u7b7e\u548c\u8d1f\u9762\u8bc4\u4ef7\u6807\u7b7e":88,"\u5373\u7279\u5f81\u7684\u6570\u7ec4":56,"\u5373\u7f51\u5361\u540d":75,"\u5373\u82e5\u5e72\u6570\u636e\u6587\u4ef6\u8def\u5f84\u7684\u67d0\u4e00\u4e2a":2,"\u5373\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u51fa\u73b0nan\u6216\u8005inf":46,"\u5373\u8bbe\u7f6e":46,"\u5373\u8fd9\u4e2a\u52a8\u6001\u5e93\u662f\u4e0d\u4f9d\u8d56\u4e8e\u5176\u4ed6\u4efb\u4f55\u6587\u4ef6\u7684":33,"\u5373define_py_data_sources2\u5e94\u6539\u4e3a":46,"\u5373input":58,"\u5373rnn\u4e4b\u95f4\u6709\u4e00\u6b21\u5d4c\u5957\u5173\u7cfb":56,"\u5377\u79ef\u5c42":81,"\u5377\u79ef\u5c42\u6743\u91cd":82,"\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u53ef\u4ee5\u8fa8\u8bc6\u56fe\u7247\u4e2d\u7684\u4e3b\u4f53":81,"\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u5728\u56fe\u7247\u5206\u7c7b\u4e0a\u6709\u7740\u60ca\u4eba\u7684\u6027\u80fd":81,"\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u662f\u4e00\u79cd\u4f7f\u7528\u5377\u79ef\u5c42\u7684\u524d\u5411\u795e\u7ecf\u7f51\u7edc":81,"\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u80fd\u591f\u5f88\u597d\u7684\u8868\u793a\u8fd9\u4e24\u7c7b\u4fe1\u606f":81,"\u5377\u79ef\u7f51\u7edc\u662f\u4e00\u79cd\u7279\u6b8a\u7684\u4ece\u8bcd\u5411\u91cf\u8868\u793a\u5230\u53e5\u5b50\u8868\u793a\u7684\u65b9\u6cd5":84,"\u5378\u8f7dpaddlepaddle\u5305":46,"\u538b\u6241\u6210\u4e3aeigen\u7684\u4e00\u7ef4tensor":64,"\u538b\u7f29\u6210\u4e00\u4e2a\u5411\u91cf":56,"\u539f\u56e0":61,"\u539f\u56e0\u5728\u4e8e\u6ca1\u6709\u628a\u673a\u5668\u4e0acuda\u76f8\u5173\u7684\u9a71\u52a8\u548c\u5e93\u6620\u5c04\u5230\u5bb9\u5668\u5185\u90e8":46,"\u539f\u56e0\u662f\u6bcf\u4e2a\u56de\u590d\u90fd\u4f1a\u53d1\u9001\u4e00\u5c01\u90ae\u4ef6":61,"\u53bb\u8fc7":56,"\u53c2\u6570":[2,6,7,8,9,10,11,12,14,33,46,60,62,69,75,80,82,88],"\u53c2\u6570\u4e3a":63,"\u53c2\u6570\u5171\u4eab\u7684\u914d\u7f6e\u793a\u4f8b\u4e3a":46,"\u53c2\u6570\u521d\u59cb\u5316\u8def\u5f84":87,"\u53c2\u6570\u5373\u53ef":88,"\u53c2\u6570\u540d":82,"\u53c2\u6570\u6570\u91cf":84,"\u53c2\u6570\u670d\u52a1\u5668":69,"\u53c2\u6570\u670d\u52a1\u5668\u7684\u53c2\u6570\u5206\u5757\u5927\u5c0f":70,"\u53c2\u6570\u670d\u52a1\u5668\u7684\u76d1\u542c\u7aef\u53e3":70,"\u53c2\u6570\u670d\u52a1\u5668\u7684\u7f51\u7edc\u8bbe\u5907\u540d\u79f0":70,"\u53c2\u6570\u670d\u52a1\u5668\u7684ip\u5730\u5740":70,"\u53c2\u6570\u670d\u52a1\u5668\u7a00\u758f\u66f4\u65b0\u7684\u53c2\u6570\u5206\u5757\u5927\u5c0f":70,"\u53c2\u6570\u6765\u63a7\u5236\u7f13\u5b58\u65b9\u6cd5":46,"\u53c2\u6570\u6982\u8ff0":71,"\u53c2\u6570\u7684\u89e3\u6790":75,"\u53c2\u6570\u7ef4\u5ea6":80,"\u53c2\u6570\u884c":80,"\u53c2\u6570\u8bbe\u7f6e\u4e86\u5916\u5c42":56,"\u53c2\u6570\u8f93\u5165":46,"\u53c2\u6570\u9700\u8981\u5b9e\u73b0":59,"\u53c2\u7167\u4e0a\u8ff0\u6b65\u9aa4\u66f4\u65b0":61,"\u53c2\u8003":[28,33,73],"\u53c2\u8003\u5f3a\u8c03\u90e8\u5206":67,"\u53c2\u8003\u6587\u732e":89,"\u53c2\u8003\u65f6\u95f4\u5e8f\u5217":56,"\u53c2\u8003\u955c\u50cf\u7684":75,"\u53c8":56,"\u53c8\u662f\u4e00\u4e2a\u5355\u5c42\u7684\u5e8f\u5217":55,"\u53c8\u8981\u4fdd\u8bc1\u6570\u636e\u662f\u968f\u673a\u7684":46,"\u53ca":62,"\u53cc\u5411\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u9690\u85cf\u72b6\u6001":59,"\u53cc\u5411\u9a8c\u8bc1":28,"\u53cc\u5c42":58,"\u53cc\u5c42\u4e0d\u7b49\u957frnn":56,"\u53cc\u5c42\u5e8f\u5217":55,"\u53cc\u5c42\u5e8f\u5217\u6216\u5355\u5c42\u5e8f\u5217":55,"\u53cc\u5c42\u5e8f\u5217\u6570\u636e\u4e00\u5171\u67094\u4e2a\u6837\u672c":56,"\u53cc\u5c42\u5e8f\u5217\u662f\u4e00\u4e2a\u5d4c\u5957\u7684\u5e8f\u5217":55,"\u53cc\u5c42\u5e8f\u5217\u662fpaddlepaddle\u652f\u6301\u7684\u4e00\u79cd\u975e\u5e38\u7075\u6d3b\u7684\u6570\u636e\u7ec4\u7ec7\u65b9\u5f0f":58,"\u53cc\u5c42\u5e8f\u5217\u6bcf\u4e2asubseq\u4e2d\u6bcf\u4e2a\u5143\u7d20":55,"\u53cc\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u53d8\u6210\u4e00\u4e2a0\u5c42\u5e8f\u5217":55,"\u53cc\u5c42\u6216\u8005\u5355\u5c42":55,"\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217\u7684dataprovider\u7684\u4ee3\u7801":56,"\u53cc\u5c42rnn":58,"\u53cc\u5c42rnn\u6570\u636e\u968f\u610f\u52a0\u4e86\u4e00\u4e9b\u9694\u65ad":56,"\u53cc\u5c42rnn\u987e\u540d\u601d\u4e49":56,"\u53cc\u8fdb\u5355\u51fa":58,"\u53cc\u8fdb\u53cc\u51fa":58,"\u53cd\u4e4b\u5219":87,"\u53cd\u5411\u4f20\u64ad":62,"\u53cd\u5411\u4f20\u64ad\u6839\u636e\u8f93\u51fa\u7684\u68af\u5ea6":62,"\u53cd\u5411\u5355\u5143\u6d4b\u8bd5\u51fd\u6570\u9700\u8981\u4ee5":63,"\u53cd\u5411op\u5355\u5143\u6d4b\u8bd5\u7ee7\u627f\u81ea":63,"\u53cd\u5411op\u7684\u68af\u5ea6\u6d4b\u8bd5":63,"\u53cd\u5411op\u7c7b":63,"\u53cd\u5411op\u7c7b\u7684\u5b9a\u4e49":63,"\u53cd\u5411operator\u5355\u6d4b":63,"\u53cd\u5411opkernel\u7684\u5b9a\u4e49\u4e0e\u524d\u5411op\u7c7b\u4f3c":63,"\u53d1\u5e03":49,"\u53d1\u5e03\u5230dockerhub":42,"\u53d1\u5e03\u5230github":42,"\u53d1\u6563\u5230\u4e86\u4e00\u4e2a\u6570\u503c\u7279\u522b\u5927\u7684\u5730\u65b9":46,"\u53d1\u884c\u548c\u7ef4\u62a4":61,"\u53d1\u9001\u53c2\u6570\u7684\u7aef\u53e3\u53f7":70,"\u53d6\u503c\u76f8\u540c\u7684layer":46,"\u53d6\u51b3\u4e8e":63,"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230cuda\u5de5\u5177\u94fe":48,"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230gtest":48,"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230swig":48,"\u53d8\u6362\u77e9\u9635":62,"\u53d8\u91cf\u6765\u8bbe\u7f6e\u5185\u5b58\u4e2d\u6682\u5b58\u7684\u6570\u636e\u6761":2,"\u53e3\u5934":56,"\u53e3\u7edf\u8ba1\u5b66\u4fe1\u606f\u7684\u7528\u6237\u624d\u88ab\u5305\u542b\u5728\u6570\u636e\u96c6\u4e2d":85,"\u53e5\u5b50":88,"\u53e5\u5b50\u4e2d\u7684\u7ec4\u5757\u5c06\u4f1a\u626e\u6f14\u67d0\u4e9b\u8bed\u4e49\u89d2\u8272":87,"\u53e5\u5b50\u8868\u793a\u7684\u8ba1\u7b97\u66f4\u65b0\u4e3a\u4e24\u6b65":84,"\u53e6\u4e00\u4e2a\u4f8b\u5b50\u662f\u901a\u8fc7\u5206\u6790\u6bcf\u65e5twitter\u535a\u5ba2\u7684\u6587\u672c\u5185\u5bb9\u6765\u9884\u6d4b\u80a1\u7968\u53d8\u52a8":88,"\u53e6\u4e00\u4e2a\u597d\u5904\u662f\u6211\u4eec\u53ef\u4ee5\u628apaddlepaddle\u5bb9\u5668\u8fd0\u884c\u5728\u8fdc\u7a0b\u670d\u52a1\u5668\u4e0a":49,"\u53e6\u4e00\u4e2a\u662f\u5185\u5b58\u64cd\u4f5c\u91cf":67,"\u53e6\u4e00\u4e2a\u662f\u521b\u5efadockerfile\u5e76\u8fd0\u884cdock":49,"\u53e6\u4e00\u4e2a\u662f\u6bcf\u6761\u5e8f\u5217":46,"\u53e6\u4e00\u4e2a\u7ec8\u7aef\u8fd0\u884cpython":49,"\u53e6\u4e00\u65b9\u9762":88,"\u53e6\u4e00\u79cd\u65b9\u5f0f\u662f\u5c06\u7f51\u7edc\u5c42\u5212\u5206\u5230\u4e0d\u540c\u7684gpu\u4e0a\u53bb\u8ba1\u7b97":72,"\u53e6\u5916":[56,60],"\u53e6\u5916\u4e24\u4e2a\u5206\u522b\u662f\u6ed1\u52a8\u5747\u503c\u548c\u65b9\u5dee":82,"\u53ea\u4f5c\u4e3aread":58,"\u53ea\u4fdd\u5b58\u6700\u540e\u4e00\u8f6e\u7684\u53c2\u6570":70,"\u53ea\u5141\u8bb8\u6574\u6570\u7684\u661f\u7ea7":85,"\u53ea\u5728\u7b2c\u4e00\u6b21cmake\u7684\u65f6\u5019\u6709\u6548":48,"\u53ea\u5bf9\u7279\u6b8a\u5728\u7ebf\u7cfb\u7edf\u8003\u8651\u4e24\u53f0\u4ee5\u4e0a\u540c\u65f6\u6545\u969c\u7684\u5bb9\u707e":20,"\u53ea\u622a\u53d6\u4e2d\u5fc3\u65b9\u5f62\u7684\u56fe\u50cf\u533a\u57df":82,"\u53ea\u662f\u53cc\u5c42\u5e8f\u5217\u5c06\u5176\u53c8\u505a\u4e86\u5b50\u5e8f\u5217\u5212\u5206":56,"\u53ea\u662f\u5c06\u53e5\u5b50\u7528\u8fde\u7eed\u5411\u91cf\u8868\u793a\u66ff\u6362\u4e3a\u7528\u7a00\u758f\u5411\u91cf\u8868\u793a":84,"\u53ea\u662f\u8bf4\u660e\u6570\u636e\u7684\u987a\u5e8f\u662f\u91cd\u8981\u7684":2,"\u53ea\u662f\u8bf7\u4e0d\u8981\u5fd8\u8bb0\u63d0\u524d\u5728\u7269\u7406\u673a\u4e0a\u5b89\u88c5gpu\u6700\u65b0\u9a71\u52a8":49,"\u53ea\u66b4\u9732\u6982\u5ff5\u7684\u63a5\u53e3":34,"\u53ea\u6709":56,"\u53ea\u67092\u4e2a\u914d\u7f6e\u4e0d\u4e00\u6837":80,"\u53ea\u6709\u542b\u6709\u4eba":85,"\u53ea\u6709\u5f53\u8bbe\u7f6e\u4e86spars":70,"\u53ea\u7528\u4e8e\u5728\u5e8f\u5217\u751f\u6210\u4efb\u52a1\u4e2d\u6307\u5b9a\u8f93\u5165\u6570\u636e":58,"\u53ea\u80fd\u5728recurrent_group\u4e2d\u4f5c\u4e3astep":46,"\u53ea\u80fd\u6d4b\u8bd5\u5355\u4e2a\u6a21\u578b":72,"\u53ea\u80fd\u8bbf\u95ee\u5b83\u4eec\u7684\u8f93\u51fa\u503c":46,"\u53ea\u80fd\u8c03\u7528paddle\u7684\u52a8\u6001\u5e93":33,"\u53ea\u8981\u4e00\u7cfb\u5217\u7279\u5f81\u6570\u636e\u4e2d\u7684":56,"\u53ea\u8981\u51fa\u73b0\u6d6e\u70b9\u6570\u5f02\u5e38":46,"\u53ea\u8981\u5728docker\u91cc\u542f\u52a8paddlepaddle\u7684\u65f6\u5019\u7ed9\u5b83\u4e00\u4e2a\u540d\u5b57":49,"\u53ea\u8bfbmemory\u8f93\u5165":58,"\u53ea\u9488\u5bf9\u5185\u5b58":46,"\u53ea\u9700\u4e2d\u65ad":68,"\u53ea\u9700\u4f7f\u7528":68,"\u53ea\u9700\u5220\u9664\u6700\u540e\u4e00\u884c\u4e2d\u7684\u6ce8\u91ca\u5e76\u628a":88,"\u53ea\u9700\u5728linux\u4e0b\u8fd0\u884c\u5982\u4e0b\u547d\u4ee4":89,"\u53ea\u9700\u7528\u4f60\u5b9a\u4e49\u7684\u76ee\u5f55\u4fee\u6539":68,"\u53ea\u9700\u77e5\u9053\u8fd9\u662f\u4e00\u4e2a\u6807\u8bb0\u5c5e\u6027\u7684\u65b9\u6cd5\u5c31\u53ef\u4ee5\u4e86":2,"\u53ea\u9700\u8981":59,"\u53ea\u9700\u8981\u4e00\u884c\u4ee3\u7801\u5c31\u53ef\u4ee5\u8c03\u7528\u8fd9\u4e2apydataprovider2":2,"\u53ea\u9700\u8981\u5728\u51fd\u6570\u4e2d\u8c03\u7528\u591a\u6b21yield\u5373\u53ef":2,"\u53ea\u9700\u8981\u6062\u590d\u8fd9\u53f0\u8282\u70b9":20,"\u53ea\u9700\u8981\u7b80\u5355\u5730\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4":89,"\u53ea\u9700\u8981\u7b80\u5355\u7684\u8fd0\u884c\u4e0b\u9762\u7684\u547d\u4ee4\u5373\u53ef":86,"\u53ea\u9700\u8981\u8fd0\u884c":86,"\u53ef\u4ee5":[56,61,68],"\u53ef\u4ee5\u4ee5\u540e\u53f0\u8fdb\u7a0b\u65b9\u5f0f\u8fd0\u884c\u5bb9\u5668":49,"\u53ef\u4ee5\u4f20\u5165\u4e00\u4e2a\u51fd\u6570":2,"\u53ef\u4ee5\u4f30\u8ba1\u51fa\u5982\u679c\u6a21\u578b\u91c7\u7528\u4e0d\u53d8\u7684\u8f93\u51fa\u6700\u5c0f\u7684cost0\u662f\u591a\u5c11":46,"\u53ef\u4ee5\u4f7f\u7528":46,"\u53ef\u4ee5\u4f7f\u7528\u5982\u4e0b\u4ee3\u7801":46,"\u53ef\u4ee5\u4f7f\u7528\u76f8\u5e94\u6570\u636e\u7c7b\u578b\u7684":46,"\u53ef\u4ee5\u4f7f\u7528\u8be5\u53c2\u6570":70,"\u53ef\u4ee5\u4f7f\u7528kubernetes\u7684\u547d\u4ee4\u884c\u5de5\u5177\u521b\u5efajob":75,"\u53ef\u4ee5\u4f7f\u7528python\u7684":4,"\u53ef\u4ee5\u51cf\u5c0f\u7cfb\u7edf\u590d\u6742\u6027":20,"\u53ef\u4ee5\u51cf\u5c11\u7f13\u5b58\u6c60\u7684\u5927\u5c0f":46,"\u53ef\u4ee5\u521b\u5efa\u4e00\u4e2a":74,"\u53ef\u4ee5\u521b\u5efa\u975e":63,"\u53ef\u4ee5\u53c2\u7167\u4e0b\u9762\u7684\u547d\u4ee4\u8fdb\u884c\u5b89\u88c5":81,"\u53ef\u4ee5\u53c2\u8003":[56,59,60,61,73,75,89],"\u53ef\u4ee5\u53c2\u8003\u4e0b\u9762\u7684\u6b65\u9aa4\u6392\u67e5":46,"\u53ef\u4ee5\u53c2\u8003\u4fdd\u5b58\u5728":80,"\u53ef\u4ee5\u53c2\u8003paddlepaddl":51,"\u53ef\u4ee5\u540c\u65f6\u5728cpu":64,"\u53ef\u4ee5\u544a\u8bc9\u60a8\u67d0\u4e2a\u64cd\u4f5c\u5230\u5e95\u82b1\u4e86\u591a\u957f\u65f6\u95f4":67,"\u53ef\u4ee5\u5728":49,"\u53ef\u4ee5\u5728\u4efb\u4f55\u673a\u5668\u4e0a\u6267\u884c\u7684":33,"\u53ef\u4ee5\u5728\u5171\u4eab\u5b58\u50a8\u4e0a\u67e5\u770b\u8f93\u51fa\u7684\u65e5\u5fd7\u548c\u6a21\u578b":75,"\u53ef\u4ee5\u5728\u5f88\u5927\u7a0b\u5ea6\u4e0a\u6d88\u9664\u6b67\u4e49":87,"\u53ef\u4ee5\u5728\u7f51\u7ad9\u4e0a\u627e\u5230":87,"\u53ef\u4ee5\u5728\u8fd9\u4e2a":61,"\u53ef\u4ee5\u5728kubernetes\u4e2d\u6309\u7167":73,"\u53ef\u4ee5\u5b9e\u73b0\u4ecepaddl":64,"\u53ef\u4ee5\u5c06\u67d0\u4e00\u4e2a\u51fd\u6570\u6807\u8bb0\u6210\u4e00\u4e2apydataprovider2":2,"\u53ef\u4ee5\u5c06\u78c1\u76d8\u4e0a\u67d0\u4e2a\u76ee\u5f55\u5171\u4eab\u7ed9\u7f51\u7edc\u4e2d\u5176\u4ed6\u673a\u5668\u8bbf\u95ee":73,"\u53ef\u4ee5\u5c06memory\u7406\u89e3\u4e3a\u4e00\u4e2a\u65f6\u5ef6\u64cd\u4f5c":58,"\u53ef\u4ee5\u5c06op\u5206\u4e3a\u4e24\u79cd":63,"\u53ef\u4ee5\u5e2e\u60a8\u63d0\u4f9b\u4e00\u4e9b\u5b9a\u4f4d\u6027\u80fd\u74f6\u9888\u7684\u5efa\u8bae":67,"\u53ef\u4ee5\u5e76\u884c\u7f16\u8bd1\u5417":60,"\u53ef\u4ee5\u5feb\u901f\u8fed\u4ee3":49,"\u53ef\u4ee5\u6267\u884c":[46,49],"\u53ef\u4ee5\u6267\u884c\u4ee5\u4e0b\u547d\u4ee4\u7f16\u8bd1\u751f\u6210\u6587\u6863":65,"\u53ef\u4ee5\u628a\u5b83\u60f3\u8c61\u4e3a\u4e00\u4e2a\u7c7b\u4f3c":60,"\u53ef\u4ee5\u628a\u672c\u5730\u7684\u6570\u636e\u4e0a\u4f20\u5230\u5b58\u50a8\u96c6\u7fa4\u4e2d":21,"\u53ef\u4ee5\u6307\u5b9a\u54ea\u4e00\u4e2a\u8f93\u5165\u548c\u8f93\u51fa\u5e8f\u5217\u4fe1\u606f\u4e00\u81f4":56,"\u53ef\u4ee5\u6309\u5982\u4e0b\u7684\u7ed3\u6784\u6765\u51c6\u5907\u6570\u6910":88,"\u53ef\u4ee5\u662f\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":[55,58],"\u53ef\u4ee5\u662f\u4e00\u4e2a\u975e\u5e8f\u5217":58,"\u53ef\u4ee5\u662f\u4ee5\u4e0b\u51e0\u79cd":62,"\u53ef\u4ee5\u663e\u793a\u5730\u6307\u5b9a\u4e00\u4e2alayer\u7684\u8f93\u51fa\u7528\u4e8e\u521d\u59cb\u5316memori":58,"\u53ef\u4ee5\u6709\u4ee5\u4e0b\u4e24\u79cd":58,"\u53ef\u4ee5\u6709\u6548\u51cf\u5c0f\u7f51\u7edc\u7684\u963b\u585e":70,"\u53ef\u4ee5\u6709\u6548\u7684\u907f\u514dparamet":20,"\u53ef\u4ee5\u67e5\u770b":75,"\u53ef\u4ee5\u67e5\u770b\u6b64pod\u8fd0\u884c\u7684\u5bbf\u4e3b\u673a":74,"\u53ef\u4ee5\u6d4b\u8bd5\u591a\u4e2a\u6a21\u578b":72,"\u53ef\u4ee5\u7528":[28,60],"\u53ef\u4ee5\u7528\u4e8e\u4ece\u5b98\u65b9\u7f51\u7ad9\u4e0a\u4e0b\u8f7dcifar":81,"\u53ef\u4ee5\u7528\u4e8e\u5c0f\u91cf\u6570\u636e\u7684\u9a8c\u8bc1":73,"\u53ef\u4ee5\u7528\u4e8e\u63a5\u6536\u548cpydataprovider2\u4e00\u6837\u7684\u8f93\u5165\u6570\u636e\u5e76\u8f6c\u6362\u6210\u9884\u6d4b\u63a5\u53e3\u6240\u9700\u7684\u6570\u636e\u7c7b\u578b":4,"\u53ef\u4ee5\u7528\u4ee5\u4e0b\u6307\u4ee4":21,"\u53ef\u4ee5\u7528\u5982\u4e0b\u547d\u4ee4":61,"\u53ef\u4ee5\u7528\u6765\u8ba1\u7b97cpu\u51fd\u6570\u6216cuda\u5185\u6838\u7684\u65f6\u95f4\u6d88\u8017":67,"\u53ef\u4ee5\u770b\u4f5c\u662f\u4e00\u4e2a\u975e\u5e8f\u5217\u8f93\u5165":55,"\u53ef\u4ee5\u7cbe\u786e\u8bf4\u660e\u4e00\u4e2a\u957f\u8017\u65f6\u64cd\u4f5c\u7684\u5177\u4f53\u539f\u56e0":67,"\u53ef\u4ee5\u7ee7\u7eed\u5728\u81ea\u5df1\u7684\u529f\u80fd\u5206\u652f\u63d0\u4ea4\u4ee3\u7801":42,"\u53ef\u4ee5\u7f16\u5199":49,"\u53ef\u4ee5\u8003\u8651\u4f7f\u7528\u4e00\u4e9b\u4f18\u5316\u7b97\u6cd5":46,"\u53ef\u4ee5\u8054\u7cfbop":46,"\u53ef\u4ee5\u8054\u7cfbop\u662f\u5426\u53ef\u4ee5\u66f4\u6362\u96c6\u7fa4\u6216\u5347\u7ea7\u5f53\u524d\u96c6\u7fa4":46,"\u53ef\u4ee5\u88c5\u7684\u662f":60,"\u53ef\u4ee5\u8bbe\u7f6e":[53,54,81],"\u53ef\u4ee5\u8bbf\u95ee\u7531recurr":46,"\u53ef\u4ee5\u8c03\u7528resize\u63a5\u53e3\u8fdb\u884c\u6539\u53d8":64,"\u53ef\u4ee5\u8f7b\u677e\u5730\u5b8c\u6210\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e":51,"\u53ef\u4ee5\u8fd0\u884c\u4e0b\u9762\u7684\u547d\u4ee4\u6765\u751f\u6210":86,"\u53ef\u4ee5\u8fd0\u884c\u811a\u672ctrain":81,"\u53ef\u4ee5\u9009\u62e9\u662f\u5426\u4f7f\u7528\u53c2\u6570":72,"\u53ef\u4ee5\u901a\u8fc7":[49,61,73],"\u53ef\u4ee5\u901a\u8fc7\u4fee\u6539\u8fd9\u4e24\u4e2a\u51fd\u6570\u6765\u5b9e\u73b0\u590d\u6742\u7684\u7f51\u7edc\u914d\u7f6e":59,"\u53ef\u4ee5\u901a\u8fc7\u8c03\u7528":4,"\u53ef\u4ee5\u901a\u8fc7\u9636\u6bb5\u6027\u7684\u4fdd\u5b58\u6bcf\u4e2aparamet":20,"\u53ef\u4ee5\u901a\u8fc7show_parameter_stats_period\u8bbe\u7f6e\u6253\u5370\u53c2\u6570\u4fe1\u606f\u7b49":84,"\u53ef\u4ee5\u91cd\u547d\u540d\u8fd9\u4e2awhl\u5305\u4e3a":46,"\u53ef\u7528\u4e8e\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u89e3\u6790\u8fd9\u4e9b\u53c2\u6570":72,"\u53ef\u7528\u5728\u6d4b\u8bd5\u6216\u8bad\u7ec3\u65f6\u6307\u5b9a\u521d\u59cb\u5316\u6a21\u578b":84,"\u53ef\u80fd\u4f1a\u5bfc\u81f4\u51fa\u9519":75,"\u53ef\u80fd\u4f1a\u9020\u6210\u7f51\u7edc\u62e5\u585e":20,"\u53ef\u80fd\u7684\u4ee3\u7801\u4e3a":46,"\u53ef\u80fd\u7684\u539f\u56e0\u662f":46,"\u53ef\u80fd\u7684\u60c5\u51b5\u4e0b":67,"\u53ef\u80fd\u9700\u8981\u6ce8\u610f\u7ed9\u8fd9\u4e2a\u865a\u62df\u673a\u591a\u5206\u914d\u4e00\u4e9b":60,"\u53ef\u89c1\u8be5\u8ba1\u7b97\u7531\u4e24\u4e2a\u8f93\u5165":63,"\u53ef\u8bbe\u7f6e":53,"\u53ef\u9009":[2,62],"\u53ef\u91c7\u7528\u7b2c\u4e8c\u79cd\u65b9\u5f0f":46,"\u53f3\u8fb9\u662f":82,"\u5403":56,"\u5403\u996d":56,"\u5404\u65b9\u9762":56,"\u5404\u9879\u53c2\u6570\u7684\u8be6\u7ec6\u8bf4\u660e\u53ef\u4ee5\u5728\u547d\u4ee4\u884c\u53c2\u6570\u76f8\u5173\u6587\u6863\u4e2d\u627e\u5230":81,"\u5404\u9879\u66f4\u52a0\u5177\u4f53\u7684\u5355\u5143\u6d4b\u8bd5\u5728":63,"\u5408":56,"\u5408\u5e76":89,"\u5408\u5e76\u6bcf\u4e2a":89,"\u5408\u7406":56,"\u540c\u65f6":[46,67],"\u540c\u65f6\u4e5f\u4f1a\u8bfb\u53d6\u76f8\u5173\u8def\u5f84\u53d8\u91cf\u6765\u8fdb\u884c\u641c\u7d22":48,"\u540c\u65f6\u4e5f\u53ef\u4ee5\u52a0\u901f\u5f00\u59cb\u8bad\u7ec3\u524d\u6570\u636e\u8f7d\u5165\u7684\u8fc7\u7a0b":46,"\u540c\u65f6\u4e5f\u53ef\u4ee5\u901a\u8fc7":61,"\u540c\u65f6\u4e5f\u80fd\u591f\u5f15\u5165\u66f4\u52a0\u590d\u6742\u7684\u8bb0\u5fc6\u673a\u5236":58,"\u540c\u65f6\u4f1a\u8ba1\u7b97\u5206\u7c7b\u51c6\u786e\u7387":84,"\u540c\u65f6\u4f60\u53ef\u4ee5\u4f7f\u7528":82,"\u540c\u65f6\u4f7f\u7528\u4e86l2\u6b63\u5219":84,"\u540c\u65f6\u5176\u5185\u90e8\u5b9e\u73b0\u53ef\u4ee5\u907f\u514d\u7eafcpu\u7248\u672cpaddlepaddle\u5728\u6267\u884c\u672c\u8bed\u53e5\u65f6\u53d1\u751f\u5d29\u6e83":67,"\u540c\u65f6\u518d\u5c06":42,"\u540c\u65f6\u53ef\u4ee5\u4f7f\u7528\u6237\u53ea\u5173\u6ce8\u5982\u4f55\u4ece\u6587\u4ef6\u4e2d\u8bfb\u53d6\u6bcf\u4e00\u6761\u6570\u636e":2,"\u540c\u65f6\u5728\u5185\u5b58\u91cc\u76f4\u63a5\u968f\u5373\u9009\u53d6\u6570\u636e\u6765\u505ashuffl":46,"\u540c\u65f6\u5c06\u53c2\u6570\u521d\u59cb\u5316\u4e3a":46,"\u540c\u65f6\u628a\u5f53\u524d\u76ee\u5f55":60,"\u540c\u65f6\u63d0\u8d77":42,"\u540c\u65f6\u7528\u6237\u9700\u8981\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u6307\u5b9a":72,"\u540c\u65f6\u8bbe\u7f6e\u5185\u5b58\u7f13\u5b58\u529f\u80fd":46,"\u540c\u65f6\u8bbe\u7f6e\u5b83\u7684input_types\u5c5e\u6027":2,"\u540c\u65f6\u8f93\u51fa\u5e8f\u5217\u5c42\u548c\u975e\u5e8f\u5217\u5c42":46,"\u540c\u65f6\u9884\u6d4b\u7f51\u7edc\u901a\u5e38\u76f4\u63a5\u8f93\u51fa\u6700\u540e\u4e00\u5c42\u7684\u7ed3\u679c\u800c\u4e0d\u662f\u50cf\u8bad\u7ec3\u7f51\u7edc\u4e00\u6837\u518d\u63a5\u4e00\u5c42cost":4,"\u540c\u6837":51,"\u540c\u6837\u4e5f\u53ef\u4ee5\u5728\u6d4b\u8bd5\u6a21\u5f0f\u4e2d\u6307\u5b9a\u6a21\u578b\u8def\u5f84":70,"\u540c\u6837\u53ef\u4ee5\u6267\u884c":49,"\u540c\u6837\u53ef\u4ee5\u6269\u5c55\u5230\u53cc\u5c42\u5e8f\u5217\u7684\u5904\u7406\u4e0a":58,"\u540c\u6837\u53ef\u83b7\u53d6\u5230\u8f93\u5165\u8f93\u51fa\u548c\u5c5e\u6027\u53c2\u6570":63,"\u540c\u6b65\u6267\u884c\u64cd\u4f5c\u7684\u7ebf\u7a0b\u6570":70,"\u540c\u7406":63,"\u540d\u5b57\u4fee\u9970":33,"\u540d\u79f0":84,"\u540e":[46,48,53,54,61,75,88],"\u540e\u5411\u4f20\u64ad":62,"\u540e\u5411\u4f20\u64ad\u7ed9\u5b9a\u8f93\u51fa\u7684\u68af\u5ea6":62,"\u540e\u8005\u5728\u6fc0\u6d3b\u51fd\u6570\u53cd\u5411\u8ba1\u7b97\u65f6\u88ab\u8c03\u7528":46,"\u540e\u8005\u622a\u65ad\u56de\u4f20\u7ed9\u524d\u5c42\u7684\u68af\u5ea6":46,"\u540e\u8005\u7528\u4e8e\u68c0\u67e5\u53c2\u6570\u5c5e\u6027\u7684\u5408\u6cd5\u6027":63,"\u540e\u8005\u7ee7\u627f\u81ea":63,"\u540e\u9762\u8fde\u5168\u8fde\u63a5\u5c42\u548csoftmax\u5c42":88,"\u5411\u6307\u5b9a\u7684\u76ee\u5f55\u4e2d\u4e00\u4e2a\u65b0\u7684\u6587\u4ef6":20,"\u5411\u91cfenable_parallel_vector":69,"\u5411paddlepaddle\u7684\u4e3b\u7248\u672c\u5e93\u63d0\u4ea4":42,"\u5417":60,"\u5426":48,"\u5426\u5219":[1,53,54,63,68,86],"\u5426\u5219\u4f1a\u628a":61,"\u5426\u5219\u4f60\u9700\u8981\u81ea\u5df1\u4e0b\u8f7d":89,"\u5426\u5219\u4f7f\u7528\u591a\u673a\u8bad\u7ec3":70,"\u5426\u5219\u4f7f\u7528cpu\u6a21\u5f0f":70,"\u5426\u5219\u4f7f\u7528gpu":72,"\u5426\u5219\u5b83\u4ee5\u4e00\u4e2a\u5e8f\u5217\u8f93\u5165":59,"\u5426\u5219\u5f97\u628apaddle\u9759\u6001\u5e93\u94fe\u63a5\u5230\u89e3\u91ca\u5668\u91cc":33,"\u5426\u5219\u9891\u7e41\u7684\u591a\u8282\u70b9\u5de5\u4f5c\u7a7a\u95f4\u90e8\u7f72\u53ef\u80fd\u4f1a\u5f88\u9ebb\u70e6":68,"\u5426\u5b9a":87,"\u542b\u4e49":[82,88],"\u542b\u6709":85,"\u542b\u6709\u5e8f\u5217\u4fe1\u606f\u548c\u5b50\u5e8f\u5217\u4fe1\u606f\u7684\u7a20\u5bc6\u5411\u91cf":62,"\u542b\u6709\u5e8f\u5217\u4fe1\u606f\u7684\u6574\u6570":62,"\u542b\u6709\u5e8f\u5217\u4fe1\u606f\u7684\u7a20\u5bc6\u5411\u91cf":62,"\u542f\u52a8\u4e00\u4e2a\u65b0\u7684\u7ebf\u7a0b\u5f00\u59cb\u4fdd\u5b58\u68c0\u67e5\u70b9":20,"\u542f\u52a8\u5bb9\u5668\u5f00\u59cb\u8bad\u7ec3":75,"\u542f\u52a8\u5e76\u884c\u5411\u91cf\u7684\u9608\u503c":70,"\u542f\u52a8\u5feb\u901f\u5e94\u7b54":70,"\u542f\u7528\u68af\u5ea6\u53c2\u6570\u7684\u9608\u503c":70,"\u5440":56,"\u544a\u8bc9paddle\u54ea\u4e2a\u6587\u4ef6\u662f\u795e\u7ecf\u7f51\u7edc\u7684\u914d\u7f6e\u6587\u4ef6":86,"\u544a\u8bc9paddle\u5c06\u6a21\u578b\u4fdd\u5b58\u5728":86,"\u5468\u56f4":56,"\u547d\u4ee4":68,"\u547d\u4ee4\u4e3a":[46,74],"\u547d\u4ee4\u521b\u5efa\u65b0\u955c\u50cf":74,"\u547d\u4ee4\u5220\u9664":[53,54],"\u547d\u4ee4\u53ef\u4ee5\u8bbe\u7f6e":48,"\u547d\u4ee4\u6307\u5b9a\u7684\u53c2\u6570\u4f1a\u4f20\u5165\u7f51\u7edc\u914d\u7f6e\u4e2d":84,"\u547d\u4ee4\u6709\u65f6\u5019\u4f1a\u4ea7\u751f\u4e00\u4e9b\u4e2d\u95f4\u7ed3\u679c":60,"\u547d\u4ee4\u770b\u5230\u505c\u6b62\u540e\u4f46\u662f\u6ca1\u6709\u5220\u9664\u7684":60,"\u547d\u4ee4\u7f16\u8bd1\u6e90\u7801\u5373\u53ef":60,"\u547d\u4ee4\u884c\u53c2\u6570\u6587\u6863":84,"\u547d\u4ee4\u8bbe\u7f6e\u8be5\u7c7b\u7f16\u8bd1\u9009\u9879":48,"\u547d\u4ee4\u9009\u9879\u5e76\u4e14":68,"\u547d\u4ee4\u91cc\u90fd\u7528\u4e86":60,"\u547d\u540d\u4e3a":61,"\u547d\u540d\u7a7a\u95f4":73,"\u547d\u540d\u7a7a\u95f4\u4e3b\u8981\u4e3a\u4e86\u5bf9\u8c61\u8fdb\u884c\u903b\u8f91\u4e0a\u7684\u5206\u7ec4\u4fbf\u4e8e\u7ba1\u7406":73,"\u547d\u540d\u89c4\u8303":63,"\u547d\u540d\u8bf7\u9075\u5b88":63,"\u548c":[21,33,34,42,46,47,48,53,54,56,59,60,61,62,63,64,65,67,68,72,73,80,81,84,86,89],"\u548c\u4e00\u4e2a\u5df2\u7ecf\u5206\u8bcd\u540e\u7684\u53e5\u5b50":56,"\u548c\u4e09\u79cd\u5e8f\u5217\u6a21\u5f0f":[2,51],"\u548c\u4e0b\u9762\u5c06\u8981\u4ecb\u7ecd\u7684\u6ce8\u518c\u51fd\u6570\u4e00\u8d77\u653e\u5728":63,"\u548c\u4e2d\u6587\u6587\u6863":65,"\u548c\u4e4b\u524d\u51cf\u5c0f\u901a\u8fc7\u51cf\u5c0f\u7f13\u5b58\u6c60\u6765\u51cf\u5c0f\u5185\u5b58\u5360\u7528\u7684\u539f\u7406\u4e00\u81f4":46,"\u548c\u504f\u7f6e\u5411\u91cf":62,"\u548c\u5185\u5b58":60,"\u548c\u533a\u57df\u6807\u8bb0":87,"\u548c\u53cc\u5c42\u5e8f\u5217\u542b\u6709subseq":55,"\u548c\u56fd\u5185\u955c\u50cf":49,"\u548c\u5728":2,"\u548c\u5bf9\u8c61\u5b58\u50a8api":73,"\u548c\u5dee\u8bc4":84,"\u548c\u5e8f\u5217\u4e2d\u542b\u6709\u5143\u7d20\u7684\u6570\u76ee\u540c":55,"\u548c\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u8f93\u5165":59,"\u548c\u64cd\u4f5c\u7cfb\u7edf\u4e0a\u76f4\u63a5\u8fd0\u884c\u7684":60,"\u548c\u68af\u5ea6\u622a\u65ad":84,"\u548c\u6a21\u578b\u8def\u5f84":88,"\u548c\u771f\u5b9e":47,"\u548c\u793a\u4f8b2\u4e2d\u7684\u914d\u7f6e\u7c7b\u4f3c":56,"\u548c\u79bb\u7ebf\u6570\u636e\u7684\u65b9\u5f0f":21,"\u548c\u7b2c6\u884c\u7684":89,"\u548c\u90e8\u5206layer":58,"\u548cadam\u5b66\u4e60\u65b9\u6cd5":89,"\u548cargument":87,"\u548cpool":55,"\u548cpython\u63a5\u53e3\u6765\u63d0\u53d6\u7279\u5f81":82,"\u54c1\u8d28":56,"\u54ea\u4e2atrainer\u5148\u5b8c\u6210block\u7684\u8bad\u7ec3":20,"\u54ea\u4e9b\u4e0d\u662f":56,"\u5546\u52a1":56,"\u554a":56,"\u559c\u5267\u7247":85,"\u5668":84,"\u56db\u79cd\u6570\u636e\u7c7b\u578b":[2,51],"\u56de\u5f52\u8bef\u5dee\u4ee3\u4ef7\u5c42":47,"\u56e0\u4e3a\u5168\u8fde\u63a5\u5c42\u7684\u6fc0\u6d3b\u53ef\u4ee5\u662fsoftmax":62,"\u56e0\u4e3a\u5355\u4e2a\u8c13\u8bcd\u4e0d\u80fd\u7cbe\u786e\u5730\u63cf\u8ff0\u8c13\u8bcd\u4fe1\u606f":87,"\u56e0\u4e3a\u53c2\u6570":72,"\u56e0\u4e3a\u5b83\u4eec\u7684\u8ba1\u7b97\u6548\u7387\u6bd4":59,"\u56e0\u4e3a\u5b83\u6bd4":59,"\u56e0\u4e3a\u5b98\u65b9\u955c\u50cf":75,"\u56e0\u4e3a\u5bb9\u5668\u5185\u7684\u6587\u4ef6\u90fd\u662f\u6682\u65f6\u5b58\u5728\u7684":73,"\u56e0\u4e3a\u6211\u4eec\u4f1a\u628a\u6240\u6709\u7f16\u8bd1\u5de5\u5177\u90fd\u5b89\u88c5\u8fdb\u4e00\u4e2a":60,"\u56e0\u4e3a\u6e90\u7801\u5c31\u5728\u672c\u673a\u4e0a":60,"\u56e0\u4e3a\u8be5\u6587\u4ef6\u53ef\u9002\u7528\u4e8e\u9884\u6d4b":81,"\u56e0\u4e3a\u8fd9\u6837\u505a\u4e5f\u6ca1\u6cd5\u4fdd\u8bc1\u6d88\u9664\u968f\u673a\u6027":20,"\u56e0\u4e3adocker\u80fd\u5728\u6240\u6709\u4e3b\u8981\u64cd\u4f5c\u7cfb\u7edf":49,"\u56e0\u4e3apython\u7684\u641c\u7d22\u8def\u5f84\u662f\u4f18\u5148\u5df2\u7ecf\u5b89\u88c5\u7684python\u5305":46,"\u56e0\u4e3aswig\u5728\u7b2c\u4e09\u65b9\u8bed\u8a00\u4e2d\u66b4\u9732\u7684\u51fd\u6570\u540d":33,"\u56e0\u6b64":[1,2,56,58,62,63],"\u56e0\u6b64\u4f7f\u7528":2,"\u56e0\u6b64\u53cc\u5c42\u5e8f\u5217\u7684\u914d\u7f6e\u4e2d":56,"\u56e0\u6b64\u53ef\u4ee5\u4f7f\u7528\u8be5\u9009\u9879":80,"\u56e0\u6b64\u53ef\u80fd\u4f1a\u6709\u4e00\u4e9b\u9519\u8bef\u548c\u4e0d\u4e00\u81f4\u53d1\u751f":85,"\u56e0\u6b64\u53ef\u80fd\u6d4b\u8bd5\u4e0d\u591f\u5b8c\u5907":64,"\u56e0\u6b64\u5728\u8f6c\u6362\u65f6\u9700\u8981\u663e\u793a\u7684\u6307\u5b9a":64,"\u56e0\u6b64\u5982\u679c\u8fd9\u4e2a\u811a\u672c\u8fd0\u884c\u5931\u8d25":81,"\u56e0\u6b64\u5b83\u662finteger_value_sub_sequ":56,"\u56e0\u6b64\u6211\u4eec\u91c7\u7528\u8f93\u51fa\u7684\u52a0\u6743\u548c":62,"\u56e0\u6b64\u6709\u4e24\u79cd\u89e3\u51b3\u65b9\u6848":2,"\u56e0\u6b64\u7528\u6237\u5e76\u4e0d\u9700\u8981\u5173\u5fc3\u5b83\u4eec":69,"\u56e0\u6b64\u8be5\u5c42\u4e2d\u6ca1\u6709\u504f\u7f6e":82,"\u56e0\u6b64\u9519\u8bef\u7684\u4f7f\u7528\u4e8c\u8fdb\u5236\u53d1\u884c\u7248\u53ef\u80fd\u4f1a\u5bfc\u81f4\u8fd9\u79cd\u9519\u8bef":46,"\u56e0\u6b64init_hook\u5c3d\u91cf\u4f7f\u7528":2,"\u56e2\u8d2d\u7f51\u7ad9":88,"\u56fe":[82,88],"\u56fe2\u662f\u53cc\u5411lstm\u7f51\u7edc":88,"\u56fe3\u662f\u4e09\u5c42lstm\u7ed3\u6784":88,"\u56fe\u50cf\u5206\u7c7b":[42,83],"\u56fe\u50cf\u5927\u5c0f\u4e3a3":82,"\u56fe\u50cf\u63cf\u8ff0":89,"\u56fe\u7247\u5206\u4e3a10\u7c7b":81,"\u56fe\u7684\u5e95\u90e8\u662fword":88,"\u56fe\u8868":[49,88],"\u5728":[2,34,42,55,56,59,60,61,63,68,82,84,85,87],"\u57280":49,"\u5728\u4e00\u4e2a\u4e0d\u53ef\u4e2d\u65ad\u5e76\u7f3a\u5c11\u5907\u4efd\u7684\u8bad\u7ec3\u4efb\u52a1\u4e2d":20,"\u5728\u4e00\u4e2a\u529f\u80fd\u9f50\u5168\u7684kubernetes\u673a\u7fa4\u91cc":74,"\u5728\u4e00\u4e2a\u53c2\u6570\u7684\u68af\u5ea6\u88ab\u66f4\u65b0\u540e":62,"\u5728\u4e00\u4e2a\u5468\u671f\u5185\u6d4b\u8bd5\u6240\u6709\u6570\u636e":87,"\u5728\u4e00\u8f6e\u4e2d\u6bcfsave":70,"\u5728\u4e0a\u56fe\u4e2d\u663e\u793a\u4e86\u5728\u4e00\u4e2a\u5b9e\u9645\u751f\u4ea7\u73af\u5883\u4e2d\u7684\u5e94\u7528":21,"\u5728\u4e0a\u9762\u4ee3\u7801\u4e2d":56,"\u5728\u4e0a\u9762\u7684\u4ee3\u7801\u4e2d":63,"\u5728\u4e0b\u4e00\u7bc7\u4e2d":74,"\u5728\u4e0b\u9762\u4f8b\u5b50\u91cc":84,"\u5728\u4e0b\u9762\u7684\u4f8b\u5b50\u4e2d":81,"\u5728\u4e0d\u540c\u64cd\u4f5c\u7cfb\u7edf":73,"\u5728\u4e4b\u540e\u7684":46,"\u5728\u4efb\u610f\u957f\u5ea6\u8bed\u53e5\u7ffb\u8bd1\u7684\u573a\u666f\u4e0b\u90fd\u53ef\u4ee5\u89c2\u5bdf\u5230\u5176\u6548\u679c\u7684\u63d0\u5347":89,"\u5728\u4f7f\u7528\u5b83\u4e4b\u524d\u8bf7\u5b89\u88c5paddlepaddle\u7684python":88,"\u5728\u4f7f\u7528\u8be5\u6587\u6863\u4e4b\u524d":51,"\u5728\u4f7f\u7528twine\u4e0a\u4f20\u4e4b\u524d":42,"\u5728\u5168\u8fde\u63a5\u5c42\u4e2d":62,"\u5728\u5177\u4f53\u7684\u8ba1\u7b97\u4e2d":64,"\u5728\u51c6\u5907\u53d1\u8d77":61,"\u5728\u51fa\u73b0\u5355\u70b9\u6545\u969c\u65f6":20,"\u5728\u51fd\u6570":75,"\u5728\u5206\u5e03\u5f0f\u73af\u5883\u4e2d\u6d4b\u8bd5":70,"\u5728\u5206\u5e03\u5f0f\u8bad\u7ec3\u4e2d":70,"\u5728\u521b\u5efaparameters\u540e":46,"\u5728\u5355\u5c42\u6570\u636e\u7684\u57fa\u7840\u4e0a":56,"\u5728\u53c2\u6570\u670d\u52a1\u5668\u4e0a\u52a0\u8f7d\u548c\u4fdd\u5b58\u53c2\u6570":70,"\u5728\u53c2\u6570\u670d\u52a1\u5668\u7ec8\u7aef\u6bcflog":70,"\u5728\u53cc\u5c42rnn\u4e2d\u7684\u7ecf\u5178\u60c5\u51b5\u662f\u5c06\u5185\u5c42\u7684\u6bcf\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217\u6570\u636e":56,"\u5728\u53cd\u5411\u4f20\u9012\u7684\u65f6\u5019":46,"\u5728\u53d8\u6362\u65f6\u9700\u8981\u5c06\u8f93\u5165\u5e8f\u5217\u4f20\u5165":56,"\u5728\u540c\u4e00\u4e2a\u547d\u540d\u7a7a\u95f4\u4e2d":73,"\u5728\u542f\u52a8job\u4e4b\u524d":75,"\u5728\u56de\u590d\u8bc4\u5ba1\u4eba\u610f\u89c1\u65f6":61,"\u5728\u58f0\u660edataprovider\u7684\u65f6\u5019\u4f20\u5165dictionary\u4f5c\u4e3a\u53c2\u6570":2,"\u5728\u591acpu\u8bad\u7ec3\u65f6\u5171\u4eab\u8be5\u53c2\u6570":70,"\u5728\u5b8c\u6210\u795e\u7ecf\u7f51\u7edc\u7684\u642d\u5efa\u4e4b\u540e":51,"\u5728\u5b9a\u4e49\u8f93\u5165layer\u4e4b\u540e":51,"\u5728\u5b9e\u73b0\u8fc7\u7a0b\u4e2d":34,"\u5728\u5b9e\u9645\u5e94\u7528\u4e2d":46,"\u5728\u5bb9\u5668\u521b\u5efa\u540e":75,"\u5728\u5bf9\u5bb9\u5668\u7684\u63cf\u8ff0":75,"\u5728\u5c42\u4e2d\u6307\u5b9a":72,"\u5728\u5e8f\u5217\u751f\u6210\u4efb\u52a1\u4e2d":58,"\u5728\u5f00\u59cb\u8bad\u7ec3\u4e4b\u524d":[21,81],"\u5728\u5f02\u6784\u96c6\u7fa4\u4e2d":20,"\u5728\u5f15\u5165\u5176\u4ed6\u7c7b\u578b\u7684\u5934\u6587\u4ef6\u65f6":34,"\u5728\u5f53\u524d\u7684\u5b9e\u73b0\u65b9\u5f0f\u4e0b":62,"\u5728\u5f97\u5230":75,"\u5728\u5feb\u7167\u5199\u5165\u5b8c\u6210\u540e":20,"\u5728\u60a8\u7684\u5b9e\u9645\u73af\u5883\u4e2d":20,"\u5728\u6211\u4eec\u7684\u4f8b\u5b50\u4e2d":59,"\u5728\u6211\u4eec\u7684\u6d4b\u8bd5\u4e2d":88,"\u5728\u6267\u884c\u65f6":64,"\u5728\u63d0\u4ea4":61,"\u5728\u642d\u5efa\u795e\u7ecf\u7f51\u7edc\u7684\u8fc7\u7a0b\u4e2d":51,"\u5728\u6570\u636e\u52a0\u8f7d\u548c\u7f51\u7edc\u914d\u7f6e\u5b8c\u6210\u4e4b\u540e":84,"\u5728\u6587\u4ef6":86,"\u5728\u6709\u65b0\u7684\u5355\u8bcd\u6765\u4e34\u7684\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u9aa4\u5185":88,"\u5728\u672c\u4f8b\u4e2d":[56,61,72],"\u5728\u672c\u4f8b\u4e2d\u6ca1\u6709\u4f7f\u7528":2,"\u5728\u672c\u6559\u7a0b\u4e2d":[59,81],"\u5728\u672c\u6587\u4e2d":68,"\u5728\u672c\u6587\u4e2d\u4f7f\u7528\u7684":68,"\u5728\u672c\u6587\u6863\u4e2d":28,"\u5728\u672c\u6f14\u793a\u4e2d":88,"\u5728\u672c\u793a\u4f8b\u4e2d":[56,88],"\u5728\u672c\u8282\u4e2d":59,"\u5728\u673a\u7fa4\u4e0a\u8fd0\u884c\u8f6c\u6362\u7a0b\u5e8f":21,"\u5728\u6811\u7684\u6bcf\u4e00\u5c42\u4e0a":70,"\u5728\u6837\u4f8b\u4e2d":34,"\u5728\u6a21\u578b\u6587\u4ef6\u7684":68,"\u5728\u6a21\u578b\u914d\u7f6e\u4e2d\u901a\u8fc7":84,"\u5728\u6b64":[69,72],"\u5728\u6b64\u4e3a\u65b9\u4fbf\u5bf9\u6bd4\u4e0d\u540c\u7f51\u7edc\u7ed3\u6784":84,"\u5728\u6b64\u611f\u8c22":80,"\u5728\u6bcf\u4e2a\u65f6\u95f4\u6b65\u4e2d":59,"\u5728\u6bcf\u4e2a\u65f6\u95f4\u6b65\u7684\u5b50\u5e8f\u5217\u957f\u5ea6\u53ef\u4ee5\u4e0d\u76f8\u7b49":56,"\u5728\u6bcf\u4e2a\u65f6\u95f4\u6b65\u957f":59,"\u5728\u6bcf\u4e2apod\u4e0a\u90fd\u901a\u8fc7volume\u65b9\u5f0f\u6302\u8f7d\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u7684\u4e00\u4e2a\u76ee\u5f55\u7528\u4e8e\u4fdd\u5b58\u8bad\u7ec3\u6570\u636e\u548c\u8f93\u51fa\u7ed3\u679c":75,"\u5728\u6bcf\u8bad\u7ec3":86,"\u5728\u6d4b\u8bd5\u9636\u6bb5":70,"\u5728\u6d4b\u8bd5\u9636\u6bb5\u5b83\u4eec\u5c06\u4f1a\u88ab\u52a0\u8f7d\u5230\u6a21\u578b\u4e2d":82,"\u5728\u6e90\u7801\u76ee\u5f55\u6811\u7684\u6839\u76ee\u5f55\u4e2d\u8fd0\u884c":61,"\u5728\u6f14\u793a\u4e2d":87,"\u5728\u7269\u7406\u673a\u4e0a\u624b\u52a8\u90e8\u7f72":73,"\u5728\u751f\u6210\u65f6":59,"\u5728\u751f\u6210\u8fc7\u7a0b\u4e2d":89,"\u5728\u751f\u6210\u8fc7\u7a0b\u4e2d\u6211\u4eec\u4f7f\u7528sgd\u8bad\u7ec3\u7b97\u6cd5":89,"\u5728\u7528\u6237\u4f7f\u7528c":34,"\u5728\u7528\u6237\u6587\u4ef6user":86,"\u5728\u7528\u6237\u719f\u6089\u7684\u5f00\u53d1\u5e73\u53f0":53,"\u5728\u7535\u5f71\u6587\u4ef6movi":86,"\u5728\u76f8\u5e94\u7684\u4f18\u5316\u7b97\u6cd5\u91cc\u8bbe\u7f6elearning_rate_schedule\u53ca\u76f8\u5173\u53c2\u6570":46,"\u5728\u76f8\u5e94layer\u7684":46,"\u5728\u793a\u4f8b\u4e2d\u6211\u4eec\u4f7f\u7528attention\u7248\u672c\u7684gru\u7f16\u89e3\u7801\u7f51\u7edc":89,"\u5728\u793a\u4f8b\u4e2d\u6211\u4eec\u4f7f\u7528sgd\u8bad\u7ec3\u7b97\u6cd5":89,"\u5728\u793a\u4f8b\u4e2d\u6211\u4eec\u5b9a\u4e49\u4e86\u4e00\u4e2a\u5e8f\u5217\u5230\u5e8f\u5217\u7684\u751f\u6210\u6570\u636e":89,"\u5728\u793a\u4f8b\u4e2d\u6211\u4eec\u5b9a\u4e49\u4e86\u4e00\u4e2a\u5e8f\u5217\u5230\u5e8f\u5217\u7684\u8bad\u7ec3\u548c\u6d4b\u8bd5\u6570\u636e":89,"\u5728\u7a0b\u5e8f\u5f00\u59cb\u9636\u6bb5":4,"\u5728\u7b2c\u4e00\u884c\u4e2d\u6211\u4eec\u8f7d\u5165\u7528\u4e8e\u5b9a\u4e49\u7f51\u7edc\u7684\u51fd\u6570":81,"\u5728\u7ebf\u6a21\u578b\u9884\u6d4b\u670d\u52a1":21,"\u5728\u7ec4\u5408\u65f6":51,"\u5728\u7f16\u8bd1\u5bbf\u4e3b\u673a\u7248protoc\u53ef\u6267\u884c\u6587\u4ef6\u548c\u76ee\u6807\u673a\u7248openblas\u5e93\u65f6\u9700\u8981\u7528\u5230":[53,54],"\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d":62,"\u5728\u7f51\u7edc\u914d\u7f6e\u91cc":2,"\u5728\u7ffb\u8bd1\u6cd5\u8bed\u53e5\u5b50\u4e4b\u524d":89,"\u5728\u811a\u672c":86,"\u5728\u81ea\u7136\u8bed\u8a00\u5904\u7406\u4efb\u52a1\u4e2d":55,"\u5728\u8bad\u7ec3\u4e2d":59,"\u5728\u8bad\u7ec3\u4e4b\u524d":75,"\u5728\u8bad\u7ec3\u4e86":86,"\u5728\u8bad\u7ec3\u4e86\u51e0\u4e2a\u8f6e\u6b21\u4ee5\u540e":86,"\u5728\u8bad\u7ec3\u5b8c\u6210\u540e":81,"\u5728\u8bad\u7ec3\u6570\u96c6\u4e0a\u8bad\u7ec3\u751f\u6210\u8bcd\u5411\u91cf\u5b57\u5178":80,"\u5728\u8bad\u7ec3\u65f6":74,"\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d":[75,89],"\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u6bcfshow":70,"\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u8fdb\u884c\u6d4b\u8bd5":1,"\u5728\u8bbe\u7f6e":[53,54],"\u5728\u8bc4\u5ba1\u8fc7\u7a0b\u4e2d":42,"\u5728\u8be5\u793a\u4f8b\u4e2d":46,"\u5728\u8be5\u914d\u7f6e\u76847":56,"\u5728\u8bed\u8a00\u751f\u6210\u9886\u57df\u4e2d":89,"\u5728\u8d2d\u7269\u7f51\u7ad9\u4e0a":84,"\u5728\u8f6f\u4ef6\u5de5\u7a0b\u7684\u8303\u7574\u91cc":67,"\u5728\u8f93\u51fa\u7684\u8fc7\u7a0b\u4e2d":58,"\u5728\u8fd0\u884c":88,"\u5728\u8fd0\u884c\u795e\u7ecf\u7f51\u7edc\u8ba1\u7b97\u56fe\u65f6":64,"\u5728\u8fd9\u4e2a":42,"\u5728\u8fd9\u4e2a\u4efb\u52a1\u4e2d":89,"\u5728\u8fd9\u4e2a\u4f8b\u5b50\u4e2d":[47,88],"\u5728\u8fd9\u4e2a\u4f8b\u5b50\u91cc":[62,74],"\u5728\u8fd9\u4e2a\u51fd\u6570\u4e2d":56,"\u5728\u8fd9\u4e2a\u52a8\u6001\u5e93\u4e2d\u4e0d\u5d4c\u5165\u4efb\u4f55\u5176\u4ed6\u8bed\u8a00\u7684\u89e3\u91ca\u5668":33,"\u5728\u8fd9\u4e2a\u6559\u7a0b\u4e2d":67,"\u5728\u8fd9\u4e2a\u6a21\u578b\u4e2d":59,"\u5728\u8fd9\u4e2a\u793a\u4f8b\u4e2d":87,"\u5728\u8fd9\u4e2a\u9636\u6bb5\u7684\u4ee3\u7801\u6b63\u5728\u7ecf\u5386\u56de\u5f52\u6d4b\u8bd5":42,"\u5728\u8fd9\u4e9b\u5934\u6587\u4ef6\u4e2d":34,"\u5728\u8fd9\u4e9b\u6587\u4ef6\u4e2d":34,"\u5728\u8fd9\u4e9b\u7f51\u7edc\u4e2d":86,"\u5728\u8fd9\u4e9blayer\u4e2d":56,"\u5728\u8fd9\u65f6\u771f\u6b63\u7684\u5206\u914d\u5185\u5b58":64,"\u5728\u8fd9\u6b65\u4efb\u52a1\u4e2d":88,"\u5728\u8fd9\u6bb5\u4ee3\u7801\u4e2d":64,"\u5728\u8fd9\u79cd\u60c5\u51b5\u4e0b":[59,62],"\u5728\u8fd9\u79cd\u7ed3\u6784\u4e2d":58,"\u5728\u8fd9\u7bc7\u6587\u6863\u91cc":74,"\u5728\u8fd9\u7bc7\u6587\u7ae0\u91cc":75,"\u5728\u8fd9\u91cc":58,"\u5728\u8fd9\u91cc\u6211\u4eec\u4f7f\u7528\u5168\u8fde\u63a5\u5c42\u4f5c\u4e3a\u4f8b\u5b50\u6765\u5c55\u793a\u5b9e\u73b0\u65b0\u7f51\u7edc\u5c42\u6240\u9700\u8981\u7684\u56db\u4e2a\u6b65\u9aa4":62,"\u5728\u8fd9\u91cc\u7528eigenvector\u6765\u8868\u793a":64,"\u5728\u8fd9\u91cc\u9700\u8981\u6ce8\u610f\u7684\u662f":64,"\u5728\u8fdb\u884c\u7f51\u7edc\u914d\u7f6e\u4e4b\u524d":51,"\u5728\u914d\u7f6e\u4e2d\u9700\u8981\u8bfb\u53d6\u5916\u90e8\u5b57\u5178":2,"\u5728\u914d\u7f6e\u6587\u4ef6\u4e2d\u7684":82,"\u5728\u91c7\u7528sgd":46,"\u5728\u96c6\u7fa4\u4e0a\u8bad\u7ec3\u4e00\u4e2a\u7a00\u758f\u6a21\u578b\u9700\u8981\u52a0\u4e0a\u4e0b\u9762\u7684\u53c2\u6570":72,"\u5728\u975e\u5e8f\u5217\u8f93\u5165\u65f6":46,"\u5728\u9884\u5904\u7406\u542b\u6709\u591a\u884c\u6570\u6910\u7684\u6587\u672c\u6587\u4ef6\u65f6\u53c2\u6570\u8bbe\u7f6e\u7a0d\u6709\u4e0d\u540c":88,"\u5728\u9884\u6d4b\u5e8f\u5217\u6216\u6bb5\u843d\u7684\u60c5\u611f\u4e2d\u8d77\u4e3b\u8981\u4f5c\u7528":88,"\u5728android\u5e73\u53f0\u4e0a\u4e0d\u652f\u6301\u901a\u8fc7swig\u8c03\u7528\u6765\u8bad\u7ec3\u6216\u8005\u9884\u6d4b":53,"\u5728android\u5e73\u53f0\u4e0a\u53ea\u652f\u6301\u4f7f\u7528c":53,"\u5728aws\u4e0a\u5feb\u901f\u90e8\u7f72\u96c6\u7fa4":73,"\u5728build\u76ee\u5f55\u4e0b\u6267\u884c":46,"\u5728c":33,"\u5728c\u7684\u5934\u6587\u4ef6":33,"\u5728cub":81,"\u5728eigen\u4e2d":64,"\u5728generator\u7684\u4e0a\u4e0b\u6587\u4e2d\u5c3d\u91cf\u7559\u4e0b\u975e\u5e38\u5c11\u7684\u53d8\u91cf\u5f15\u7528":2,"\u5728kubernetes\u4e2d\u521b\u5efa\u7684\u6240\u6709\u8d44\u6e90\u5bf9\u8c61":73,"\u5728linux\u4e0b":89,"\u5728meta\u6587\u4ef6\u4e2d\u6709\u4e24\u79cd\u7279\u5f81":86,"\u5728movielen":86,"\u5728paddl":75,"\u5728paddle\u4e2d":72,"\u5728paddle\u4e4b\u4e0a\u8fd0\u884c\u7684\u6df1\u5ea6\u5b66\u4e60\u8bad\u7ec3\u8f93\u51fa\u7684\u6a21\u578b\u4f1a\u63d0\u4f9b\u7ed9\u5728\u7ebf\u4eba\u8138\u8bc6\u522b\u7684\u5e94\u7528\u4f7f\u7528":21,"\u5728paddlepaddle\u4e2d":[51,58],"\u5728paddlepaddle\u4e2d\u4f7f\u7528dropout\u6709\u4e24\u79cd\u65b9\u5f0f":46,"\u5728paddlepaddle\u4e2d\u5305\u542b\u4ee5\u4e0b":46,"\u5728paddlepaddle\u7684\u6587\u6863\u4e2d":56,"\u5728paddlepaddle\u91cc":47,"\u5728paramet":20,"\u5728python\u811a\u672c\u4e2d\u5b9e\u73b0\u4e0e\u524d\u5411operator\u76f8\u540c\u7684\u8ba1\u7b97\u903b\u8f91":63,"\u5728step\u51fd\u6570\u4e2d\u5b9a\u4e49":58,"\u5728step\u51fd\u6570\u4e2d\u5b9a\u4e49memori":58,"\u5728trainer":72,"\u5730\u5740\u4e5f\u53ef\u4ee5\u4e3ahdfs\u6587\u4ef6\u8def\u5f84":1,"\u5730\u6bb5":56,"\u5730\u7406\u4f4d\u7f6e":56,"\u5730\u94c1\u7ad9":56,"\u5747\u4f1a\u88ab\u5b89\u88c5\u5230includ":34,"\u5747\u503c\u56fe\u50cf\u6587\u4ef6":82,"\u5747\u5300\u5206\u5e03":46,"\u5747\u5300\u5206\u5e03\u7684\u8303\u56f4\u662f":70,"\u5747\u662f\u5728":34,"\u5747\u6709\u4e09\u4e2a\u5b50\u5e8f\u5217":56,"\u5747\u6709\u4e24\u7ec4\u7279\u5f81":56,"\u57fa\u4e8e\u53cc\u5c42\u5e8f\u5217\u8f93\u5165":58,"\u57fa\u4e8e\u5b57\u6bcd\u7684\u8bcd\u5d4c\u5165\u7279\u5f81":86,"\u57fa\u4e8e\u7c98\u6027\u4f1a\u8bdd\u7684\u8d1f\u8f7d\u5747\u8861\u529f\u80fd":28,"\u57fa\u4e8epython\u7684\u6a21\u578b\u9884\u6d4b":4,"\u57fa\u4e8epython\u7684\u9884\u6d4b":[3,84],"\u57fa\u672c\u4e0a\u548cmnist\u6837\u4f8b\u4e00\u81f4":2,"\u57fa\u672c\u4f7f\u7528\u6982\u5ff5":52,"\u57fa\u672c\u76f8\u540c":80,"\u57fa\u7c7b":63,"\u586b\u5199":61,"\u589e\u52a0":63,"\u589e\u52a0\u4e86\u4e00\u6761cd\u547d\u4ee4":74,"\u589e\u52a0\u4e86\u8bbe\u5907\u7c7b\u578b":63,"\u589e\u52a0\u5982\u4e0b\u53c2\u6570":72,"\u589e\u52a0\u68af\u5ea6\u68c0\u6d4b\u7684\u5355\u5143\u6d4b\u8bd5":62,"\u58f0\u660epython\u6570\u636e\u6e90":86,"\u5904\u7406\u5668\u6709\u4e24\u4e2a\u5173\u952e\u6027\u80fd\u9650\u5236":67,"\u5904\u7406\u6570\u636e\u7684python\u811a\u672c\u6587\u4ef6":84,"\u5904\u7406\u7684\u8f93\u5165\u5e8f\u5217\u4e3b\u8981\u5206\u4e3a\u4ee5\u4e0b\u4e09\u79cd\u7c7b\u578b":58,"\u5904\u7406\u76f8\u4f3c\u5ea6\u56de\u5f52":86,"\u5904\u7406\u8fc7\u7a0b\u4e2d\u6570\u636e\u5b58\u50a8\u683c\u5f0f":81,"\u5907\u6ce8":67,"\u590d\u6742\u5ea6\u6216\u65f6\u95f4\u590d\u6742\u5ea6":67,"\u5916\u5c42memory\u662f\u4e00\u4e2a\u5143\u7d20":56,"\u5916\u5c42outer_step\u4e2d":56,"\u591a\u4e2a\u503c":21,"\u591a\u4e2a\u5c42\u7684\u8f93\u51fa\u77e9\u9635\u7684\u9ad8\u5ea6\u4e0d\u4e00\u81f4\u5bfc\u81f4\u62fc\u63a5\u5931\u8d25":46,"\u591a\u4e2a\u8f93\u51fa\u5c42\u5904\u7406\u591a\u4e2a\u4e0d\u540c\u957f\u5ea6\u7684\u5e8f\u5217":46,"\u591a\u4e2ainput\u4ee5list\u65b9\u5f0f\u8f93\u5165":84,"\u591a\u4e2aparamet":20,"\u591a\u53e5\u8bdd\u8fdb\u4e00\u6b65\u6784\u6210\u4e86\u6bb5\u843d":58,"\u591a\u673a\u8bad\u7ec3":46,"\u591a\u7ebf\u7a0b\u7684\u6570\u636e\u8bfb\u53d6":2,"\u591a\u8f6e\u5bf9\u8bdd\u7b49\u66f4\u4e3a\u590d\u6742\u7684\u8bed\u8a00\u6570\u636e":58,"\u5927\u578b\u7535\u5f71\u8bc4\u8bba\u6570\u636e\u96c6":88,"\u5927\u591a\u6570\u5c42\u4e0d\u9700\u8981\u8fdc\u7a0b\u7a00\u758f\u8bad\u7ec3\u51fd\u6570":62,"\u5927\u591a\u6570\u5c42\u9700\u8981\u8bbe\u7f6e\u4e3a":62,"\u5927\u591a\u6570\u6210\u529f\u7684srl\u7cfb\u7edf\u662f\u5efa\u7acb\u5728\u67d0\u79cd\u5f62\u5f0f\u7684\u53e5\u6cd5\u5206\u6790\u7ed3\u679c\u4e4b\u4e0a\u7684":87,"\u5927\u591a\u6570\u7f51\u7edc\u5c42\u4e0d\u9700\u8981\u652f\u6301\u8fdc\u7a0b\u7a00\u758f\u66f4\u65b0":62,"\u5927\u591a\u6570\u8bed\u8a00\u90fd\u652f\u6301\u4f7f\u7528c\u8bed\u8a00api":33,"\u5927\u5b66\u751f":85,"\u5927\u5bb6\u53ef\u4ee5\u7528\u628a\u5f00\u53d1\u5de5\u5177\u5b89\u88c5\u8fdb\u5165":60,"\u5927\u5bb6\u53ef\u4ee5\u901a\u8fc7\u5b83\u5236\u4f5c\u548c\u5206\u4eab\u5e26\u6709\u4ee3\u7801":49,"\u5927\u5c0f":68,"\u5927\u5c0f\u4e0d\u4e00\u6837\u65f6":46,"\u5927\u6982\u82b1\u5341\u5206\u949f\u770b\u4e00\u4e0b":60,"\u5929":56,"\u5929\u4e00\u5e7f\u573a":56,"\u5929\u4e00\u9601":56,"\u5929\u732b":88,"\u5934\u4fe1\u606f\u4e2d":46,"\u5934\u6587\u4ef6\u4e2d\u628a\u53c2\u6570\u5b9a\u4e49\u4e3a\u7c7b\u7684\u6210\u5458\u53d8\u91cf":62,"\u5934\u6587\u4ef6\u5982\u4e0b":62,"\u5947\u5e7b\u7247":85,"\u597d":56,"\u597d\u5403":56,"\u597d\u8bc4":84,"\u5982":[2,59,61,63,68,72],"\u59822":68,"\u5982\u4e0a\u4e00\u5c0f\u8282\u6240\u793a":64,"\u5982\u4e0b":[2,86,88],"\u5982\u4e0b\u56fe\u6240\u793a":[56,67,81],"\u5982\u4e0b\u6240\u793a":[72,82,86],"\u5982\u4e0b\u662f\u4e00\u6bb5\u4f7f\u7528mnist":4,"\u5982\u4e0b\u8868\u683c":84,"\u5982\u4f55":86,"\u5982\u4f55\u5b58\u50a8\u7b49\u7b49":2,"\u5982\u4f55\u89e3\u6790\u8be5\u5730\u5740\u4e5f\u662f\u7528\u6237\u81ea\u5b9a\u4e49dataprovider\u65f6\u9700\u8981\u8003\u8651\u7684\u5730\u65b9":1,"\u5982\u4f55\u8d21\u732e":66,"\u5982\u4f55\u8d21\u732e\u4ee3\u7801":66,"\u5982\u4f55\u8fdb\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3":84,"\u5982\u4fe1\u606f\u63d0\u53d6":87,"\u5982\u56fe2\u6240\u793a":88,"\u5982\u56fe\u4e2dtrainer":20,"\u5982\u5f62\u5bb9\u8bcd\u548c\u526f\u8bcd":88,"\u5982\u60f3\u4e86\u89e3\u66f4\u591a\u8be6\u7ec6\u7684\u89e3\u91ca":89,"\u5982\u6709":63,"\u5982\u672c\u4f8b\u4e2d":2,"\u5982\u672c\u4f8b\u7684":2,"\u5982\u679c\u4e00\u4e2a\u7f51\u7edc\u5c42\u9700\u8981\u914d\u7f6e\u7684\u8bdd":62,"\u5982\u679c\u4e0a\u9762\u4e24\u6b65\u51fa\u73b0\u9519\u8bef":20,"\u5982\u679c\u4e0b\u8f7d\u6210\u529f":82,"\u5982\u679c\u4e0d\u4e3a0":70,"\u5982\u679c\u4e0d\u4e86\u89e3":2,"\u5982\u679c\u4e0d\u5207\u8bcd":84,"\u5982\u679c\u4e0d\u6536\u655b":46,"\u5982\u679c\u4e0d\u662f\u5e8f\u5217":86,"\u5982\u679c\u4e3a":2,"\u5982\u679c\u4e3a0":70,"\u5982\u679c\u4e3afals":70,"\u5982\u679c\u4e3atrue":[2,70],"\u5982\u679c\u4e4b\u540e\u60f3\u8981\u91cd\u65b0\u8bbe\u7f6e":48,"\u5982\u679c\u4ec5\u4ec5\u4fee\u6539\u4e00\u4e2a\u6587\u4ef6\u4f46\u63d0\u4ea4\u4e86\u5341\u51e0\u4e2acommit":61,"\u5982\u679c\u4ed4\u7ec6\u8bbe\u7f6e\u7684\u8bdd":70,"\u5982\u679c\u4f60\u4e0d\u9700\u8981\u8fd9\u4e2a\u64cd\u4f5c":88,"\u5982\u679c\u4f60\u53ea\u9700\u8981\u4f7f\u7528\u7b80\u5355\u7684rnn":59,"\u5982\u679c\u4f60\u5b89\u88c5gpu\u7248\u672c\u7684paddlepaddl":88,"\u5982\u679c\u4f60\u60f3\u4f7f\u7528\u8fd9\u4e9b\u7279\u6027":72,"\u5982\u679c\u4f60\u60f3\u8981\u4fdd\u5b58\u67d0\u4e9b\u5c42\u7684\u7279\u5f81\u56fe":70,"\u5982\u679c\u4f60\u60f3\u8fdb\u884c\u8bf8\u5982\u8bed\u4e49\u8f6c\u8ff0":89,"\u5982\u679c\u4f60\u6267\u884c\u5176\u5b83\u7684\u7528\u60c5\u611f\u5206\u6790\u6765\u5206\u7c7b\u6587\u672c\u7684\u4efb\u52a1":88,"\u5982\u679c\u4f60\u66fe\u7ecf\u5728\u6e90\u7801\u76ee\u5f55\u4e0b\u7f16\u8bd1\u8fc7\u5176\u4ed6\u5e73\u53f0\u7684paddlepaddle\u5e93":[53,54],"\u5982\u679c\u4f60\u6b63\u5728\u5904\u7406\u5e8f\u5217\u6807\u8bb0\u4efb\u52a1":59,"\u5982\u679c\u4f60\u6ca1\u6709gpu\u73af\u5883":81,"\u5982\u679c\u4f60\u8981\u4e3a\u4e86\u6d4b\u8bd5\u800c\u589e\u52a0\u65b0\u7684\u6587\u4ef6":62,"\u5982\u679c\u4f7f\u7528":[68,80],"\u5982\u679c\u4f7f\u7528nvidia":49,"\u5982\u679c\u4f7f\u7528ssl\u8ba4\u8bc1":73,"\u5982\u679c\u4f7f\u7528swig\u6211\u4eec\u9700\u8981\u5c06\u5728interface\u6587\u4ef6\u91cc":33,"\u5982\u679c\u51fa\u73b0\u4ee5\u4e0bpython\u76f8\u5173\u7684\u5355\u5143\u6d4b\u8bd5\u90fd\u8fc7\u4e0d\u4e86\u7684\u60c5\u51b5":46,"\u5982\u679c\u53c2\u6570\u4fdd\u5b58\u4e0b\u6765\u7684\u6a21\u578b\u76ee\u5f55":46,"\u5982\u679c\u53c2\u6570\u6a21\u578b\u6587\u4ef6\u7f3a\u5931":80,"\u5982\u679c\u53d1\u73b0\u6700\u65e9\u7684\u62a5\u9519\u5c31\u662f\u7f51\u7edc\u901a\u4fe1\u7684\u95ee\u9898":46,"\u5982\u679c\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u672a\u8bbe\u7f6easync":70,"\u5982\u679c\u5728\u8bad\u7ec3\u671f\u95f4\u540c\u65f6\u53d1\u8d77\u53e6\u5916\u4e00\u4e2a\u8fdb\u7a0b\u8fdb\u884c\u6d4b\u8bd5":70,"\u5982\u679c\u5728\u8bad\u7ec3\u914d\u7f6e\u4e2d\u8bbe\u7f6ebatch":70,"\u5982\u679c\u5728\u8bad\u7ec3nlp\u76f8\u5173\u6a21\u578b\u65f6":46,"\u5982\u679c\u591a\u4e2aop\u4f9d\u8d56\u4e00\u4e9b\u5171\u7528\u7684\u51fd\u6570":63,"\u5982\u679c\u5931\u8d25":42,"\u5982\u679c\u5b58\u5728\u67d0\u4e9btrainer\u6267\u884c\u901f\u5ea6\u8fc7\u6162\u4f1a\u5f71\u54cd\u6574\u4f53\u96c6\u7fa4\u7684\u901f\u5ea6":20,"\u5982\u679c\u5b83\u4f4d\u4e8e\u8c13\u8bcd\u4e0a\u4e0b\u6587\u533a\u57df\u4e2d":87,"\u5982\u679c\u5c06\u8fd9\u4e2a\u5185\u5b58\u6c60\u51cf\u5c0f":46,"\u5982\u679c\u5df2\u5b89\u88c5":87,"\u5982\u679c\u5df2\u7ecf\u6709pod\u8fd0\u884c":75,"\u5982\u679c\u5df2\u7ecf\u6b63\u5728\u6267\u884c\u4fdd\u5b58\u68c0\u67e5\u70b9\u7684\u7ebf\u7a0b":20,"\u5982\u679c\u5f00\u542f\u4f1a\u5bfc\u81f4\u8fd0\u884c\u7565\u6162":48,"\u5982\u679c\u5f53\u524dmpi\u96c6\u7fa4\u5e76\u4e0d\u652f\u6301\u4efb\u52a1\u72ec\u5360\u6a21\u5f0f":46,"\u5982\u679c\u60a8\u5728\u56fd\u5185":49,"\u5982\u679c\u60a8\u60f3\u8981\u66f4\u6df1\u5165\u4e86\u89e3deep":49,"\u5982\u679c\u60a8\u6709\u597d\u7684\u5efa\u8bae\u6765":86,"\u5982\u679c\u60a8\u6ca1\u6709\u542c\u8bf4":60,"\u5982\u679c\u60a8\u7684gpu\u7406\u8bba\u53ef\u4ee5\u8fbe\u52306":67,"\u5982\u679c\u60f3\u4e3a\u4e00\u4e2a\u6570\u636e\u6587\u4ef6\u8fd4\u56de\u591a\u6761\u6837\u672c":2,"\u5982\u679c\u60f3\u4f7f\u7528\u53ef\u89c6\u5316\u7684\u5206\u6790\u5668":67,"\u5982\u679c\u60f3\u5f88\u597d\u7684\u7406\u89e3\u7a0b\u5e8f\u7684\u884c\u4e3a":67,"\u5982\u679c\u60f3\u6539\u53d8\u539f\u6709tensor\u7684shape\u4fe1\u606f":64,"\u5982\u679c\u60f3\u8981\u4e86\u89e3\u53cc\u5c42rnn\u5728\u5177\u4f53\u95ee\u9898\u4e2d\u7684\u4f7f\u7528":56,"\u5982\u679c\u60f3\u8981\u542f\u7528paddlepaddle\u7684\u5185\u7f6e\u5b9a\u65f6\u5668":67,"\u5982\u679c\u6211\u4eec\u53ea\u9700\u8981\u7f16\u8bd1\u4e00\u4e2a\u53ea\u652f\u6301":60,"\u5982\u679c\u6211\u77e5\u9053\u5185\u6838\u82b1\u4e8610ms\u6765\u79fb\u52a81gb\u6570\u636e":67,"\u5982\u679c\u6267\u884c\u5931\u8d25":73,"\u5982\u679c\u6267\u884c\u6210\u529f":82,"\u5982\u679c\u6307\u5b9a\u4e862\u4e2alayer\u4f5c\u4e3a\u8f93\u51fa\u5c42":46,"\u5982\u679c\u63d0\u793a\u6b63\u786e":65,"\u5982\u679c\u652f\u6301\u589e\u52a0\u6b64\u53c2\u6570\u63d0\u4ea4":46,"\u5982\u679c\u6570\u636e\u6587\u4ef6\u5b58\u4e8e\u672c\u5730\u78c1\u76d8":1,"\u5982\u679c\u6570\u6910\u83b7\u53d6\u6210\u529f":88,"\u5982\u679c\u662f\u4f7f\u7528\u975essl\u65b9\u5f0f\u8bbf\u95ee":73,"\u5982\u679c\u662f\u5176\u5b83\u7c7b\u578b":21,"\u5982\u679c\u662f\u5e8f\u5217":86,"\u5982\u679c\u6709\u591a\u4e2a\u8f93\u5165":58,"\u5982\u679c\u6709\u591a\u4e2a\u8f93\u5165\u5e8f\u5217":58,"\u5982\u679c\u6709\u66f4\u590d\u6742\u7684\u4f7f\u7528":1,"\u5982\u679c\u6709\u9700\u8981\u4fee\u6539\u7684\u5730\u65b9":61,"\u5982\u679c\u6709bugfix\u7684\u884c\u4e3a":42,"\u5982\u679c\u672a\u8bbe\u7f6e":70,"\u5982\u679c\u672a\u8bbe\u7f6egpu":72,"\u5982\u679c\u67d0\u4e00\u4e2a\u7c7b\u578b\u9700\u8981\u5f15\u7528\u53e6\u4e00\u4e2a\u7c7b\u578b":34,"\u5982\u679c\u67d0\u4e00\u4e2apaddl":34,"\u5982\u679c\u67d0\u4e00\u4e2apaddle\u6982\u5ff5\u5fc5\u987b\u8981\u66b4\u9732":34,"\u5982\u679c\u67d0\u4e00\u5757\u6839\u672c\u5c31\u4e0d\u600e\u4e48\u8017\u65f6":67,"\u5982\u679c\u68c0\u67e5\u5230\u5206\u914d\u5728\u4e0d\u540c\u53c2\u6570\u670d\u52a1\u5668\u4e0a\u7684\u53c2\u6570\u7684\u5206\u5e03\u4e0d\u5747\u5300\u6b21\u6570\u5927\u4e8echeck":70,"\u5982\u679c\u6ca1\u6709\u5b9a\u4e49memori":58,"\u5982\u679c\u6ca1\u6709\u8bbe\u7f6e":89,"\u5982\u679c\u6ca1\u6709\u8bbe\u7f6etest":1,"\u5982\u679c\u6ca1\u8fc7":61,"\u5982\u679c\u6d88\u606f\u6570\u636e\u592a\u5c0f":70,"\u5982\u679c\u6ee1\u8db3\u6761\u4ef6":20,"\u5982\u679c\u7528\u516c\u7528\u7684\u7535\u8111\u5f00\u53d1":60,"\u5982\u679c\u7528\u6237\u4e0d\u663e\u793a\u6307\u5b9a\u8fd4\u56de\u6570\u636e\u7684\u5bf9\u5e94\u5173\u7cfb":2,"\u5982\u679c\u7528\u6237\u4e0d\u9700\u8981\u8bbf\u95eelstm\u7684\u4e2d\u95f4\u53d8\u91cf":46,"\u5982\u679c\u7528\u6237\u60f3\u8981\u4e86\u89e3\u8be6\u7ec6\u7684\u6570\u636e\u96c6\u7684\u683c\u5f0f":80,"\u5982\u679c\u7528\u6237\u60f3\u8981\u81ea\u5b9a\u4e49\u521d\u59cb\u5316\u65b9\u5f0f":46,"\u5982\u679c\u7528\u6237\u8981\u628apaddle\u7684\u9759\u6001\u5e93":33,"\u5982\u679c\u7528\u81ea\u5df1\u7684\u7535\u8111\u5f00\u53d1":60,"\u5982\u679c\u771f\u60f3\u6316\u6398\u5185\u6838\u6df1\u5904\u7684\u67d0\u4e2a\u79d8\u5bc6":67,"\u5982\u679c\u7a0b\u5e8f\u5d29\u6e83\u4f60\u4e5f\u53ef\u4ee5\u624b\u52a8\u7ec8\u6b62":68,"\u5982\u679c\u7cfb\u7edf\u5b89\u88c5\u4e86\u591a\u4e2apython\u7248\u672c":46,"\u5982\u679c\u7cfb\u7edf\u652f\u6301":46,"\u5982\u679c\u7cfb\u7edf\u652f\u6301\u7684\u662f":46,"\u5982\u679c\u7f16\u8bd1\u7684\u65f6\u5019\u6211\u4eec\u7528\u4e86":60,"\u5982\u679c\u7f51\u7edc\u5c42\u4e0d\u9700\u8981\u8fdc\u7a0b\u7a00\u758f\u66f4\u65b0":62,"\u5982\u679c\u7f51\u7edc\u67b6\u6784\u7b80\u5355":59,"\u5982\u679c\u8981\u4e0a\u4f20gpu\u7248\u672c\u7684\u5305":42,"\u5982\u679c\u8981\u4f7f\u7528\u53cc\u5411lstm":88,"\u5982\u679c\u8981\u4f7f\u7528gpu":49,"\u5982\u679c\u8981\u542f\u7528gpu":68,"\u5982\u679c\u8981\u8fd0\u884c\u6240\u6709\u7684\u5355\u5143\u6d4b\u8bd5":61,"\u5982\u679c\u89e3\u51b3\u4e86\u67d0\u4e2aissue\u7684\u95ee\u9898":61,"\u5982\u679c\u8bad\u7ec3\u4e00\u4e2apass":46,"\u5982\u679c\u8bad\u7ec3\u8fc7\u7a0b\u542f\u52a8\u6210\u529f\u7684\u8bdd":86,"\u5982\u679c\u8bad\u7ec3\u8fc7\u7a0b\u7684\u7684cost\u660e\u663e\u9ad8\u4e8e\u8fd9\u4e2a\u5e38\u6570\u8f93\u51fa\u7684cost":46,"\u5982\u679c\u8bbe\u7f6e":2,"\u5982\u679c\u8bbe\u7f6e\u8be5\u53c2\u6570":70,"\u5982\u679c\u8bc4\u5ba1\u610f\u89c1\u6bd4\u8f83\u591a":61,"\u5982\u679c\u8bf4\u4e00\u4e2adocker\u955c\u50cf\u5c31\u662f\u4e00\u4e2a\u7a0b\u5e8f":49,"\u5982\u679c\u8c03\u7528\u9759\u6001\u5e93\u53ea\u80fd\u5c06\u9759\u6001\u5e93\u4e0e\u89e3\u91ca\u5668\u94fe\u63a5":33,"\u5982\u679c\u8f93\u51fa\u662fno":49,"\u5982\u679c\u8fd0\u884c":46,"\u5982\u679c\u8fd0\u884c\u6210\u529f":[82,88],"\u5982\u679c\u8fd8\u4e0d\u884c":46,"\u5982\u679c\u96c6\u7fa4\u8282\u70b9\u6570\u91cf\u5c11":68,"\u5982\u679c\u9700\u8981\u5305\u542b\u66f4\u591a\u7684\u4f9d\u8d56":49,"\u5982\u679c\u9700\u8981\u6269\u5927\u77e9\u9635":62,"\u5982\u679c\u9700\u8981\u7f29\u51cf\u77e9\u9635":62,"\u5982\u679cgpu":63,"\u5982\u679clearning_rate\u592a\u5927":46,"\u5982\u679clearning_rate\u592a\u5c0f":46,"\u5982\u679cop\u6ca1\u6709\u5b9e\u73b0gpu":63,"\u5982\u679cop\u7684\u67d0\u4e2a\u8f93\u5165\u4e0d\u53c2\u4e0e\u53cd\u5411\u68af\u5ea6\u7684\u8ba1\u7b97":63,"\u5982\u679cpaddlepaddle\u5305\u5df2\u7ecf\u5728python\u7684sit":46,"\u5982\u679cparamet":20,"\u5982\u6bcf\u4e2a\u6587\u4ef6\u53ea\u6709\u4e00\u4e2a":61,"\u5982\u795e\u7ecf\u5143\u6fc0\u6d3b\u503c\u7b49":46,"\u5982\u8981build\u8fd9\u4e2a\u5f00\u53d1\u955c\u50cf":61,"\u5982\u9ad8\u4eae\u90e8\u5206":67,"\u5b50":56,"\u5b50\u53e5":58,"\u5b50\u53e5\u7684\u5355\u8bcd\u6570\u548c\u6307\u5b9a\u7684\u4e00\u4e2a\u8f93\u5165\u5e8f\u5217\u4e00\u81f4":58,"\u5b50\u76ee\u5f55":60,"\u5b57\u5178":89,"\u5b57\u5178\u4f1a\u5305\u542b\u8f93\u5165\u6570\u636e\u96c6\u4e2d\u7684\u6240\u6709\u5355\u8bcd":89,"\u5b57\u5178\u5171\u5305\u542b":80,"\u5b57\u5178\u6587\u4ef6":[87,88],"\u5b57\u5178\u91c7\u7528utf8\u7f16\u7801":80,"\u5b57\u5178imdb":88,"\u5b57\u6bb5\u4e2d":75,"\u5b57\u6bb5\u4e3a\u4f8b":46,"\u5b57\u6bb5\u8868\u793a\u5bb9\u5668\u7684\u73af\u5883\u53d8\u91cf":75,"\u5b57\u6bb5\u8868\u793a\u8fd9\u4e2ajob\u4f1a\u540c\u65f6\u5f00\u542f3\u4e2apaddlepaddle\u8282\u70b9":75,"\u5b57\u7b26\u4e32":21,"\u5b58\u50a8":21,"\u5b58\u50a8\u5377":73,"\u5b58\u50a8\u5728\u8bb0\u5fc6\u5355\u5143\u533a\u5757\u7684\u5386\u53f2\u4fe1\u606f\u88ab\u66f4\u65b0\u7528\u6765\u8fed\u4ee3\u7684\u5b66\u4e60\u5355\u8bcd\u4ee5\u5408\u7406\u7684\u5e8f\u5217\u7a0b\u73b0":88,"\u5b58\u50a8\u6a21\u578b\u7684\u8def\u5f84":89,"\u5b58\u50a8\u7740\u7535\u5f71\u6216\u7528\u6237\u4fe1\u606f":86,"\u5b58\u5165settings\u5bf9\u8c61":2,"\u5b58\u5728\u6216\u66f4\u6539\u4e3a\u5176\u5b83\u6a21\u578b\u8def\u5f84":88,"\u5b66\u4e60":60,"\u5b66\u4e60\u6210\u672c\u9ad8":33,"\u5b66\u4e60\u7387\u4e3a":46,"\u5b66\u4e60\u7b97\u6cd5":47,"\u5b66\u672f":85,"\u5b81\u6ce2":56,"\u5b83\u4e0d\u4ec5\u80fd\u591f\u5904\u7406imdb\u6570\u636e":88,"\u5b83\u4eec\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4f5c\u4e3a\u7f51\u7edc\u7684\u51fa\u53e3":47,"\u5b83\u4eec\u7684\u6587\u4ef6\u540d\u662f":21,"\u5b83\u4eec\u8fd8\u53ef\u4ee5\u4f9b\u90a3\u4e9b\u8fd0\u884c\u66f4\u590d\u6742\u7684\u96c6\u7fa4\u7ba1\u7406\u7cfb\u7edf":68,"\u5b83\u4f1a\u5728dataprovider\u521b\u5efa\u7684\u65f6\u5019\u6267\u884c":2,"\u5b83\u4f7f\u752850\u5c42\u7684resnet\u6a21\u578b\u6765\u5bf9":82,"\u5b83\u5305\u542b\u4e86\u8fd9\u4e2a\u8f6f\u4ef6\u672c\u8eab\u548c\u5b83\u6240\u4f9d\u8d56\u7684\u8fd0\u884c\u73af\u5883":49,"\u5b83\u5305\u542b\u4ee5\u4e0b\u51e0\u6b65":62,"\u5b83\u5305\u542b\u4ee5\u4e0b\u53c2\u6570":62,"\u5b83\u5305\u542b\u7684\u5c5e\u6027\u53c2\u6570\u5982\u4e0b":2,"\u5b83\u5305\u62ec\u4e86\u4e00\u4e2a\u53cc\u5411\u7684gru\u4f5c\u4e3a\u7f16\u7801\u5668\u548c\u89e3\u7801\u5668":89,"\u5b83\u53eb\u505a":59,"\u5b83\u53ef\u4ee5\u5728\u53e5\u5b50\u7ea7\u522b\u5229\u7528\u53ef\u6269\u5c55\u7684\u4e0a\u4e0b\u6587":88,"\u5b83\u53ef\u4ee5\u5e2e\u52a9\u51cf\u5c11\u5206\u53d1\u5ef6\u8fdf":68,"\u5b83\u53ef\u4ee5\u5e2e\u52a9\u6211\u4eec\u683c\u5f0f\u5316\u6e90\u4ee3\u7801":61,"\u5b83\u53ef\u4ee5\u6307\u6d4b\u91cf\u4e00\u4e2a\u7a0b\u5e8f\u7684\u7a7a\u95f4":67,"\u5b83\u53ef\u4ee5\u88ab\u5e94\u7528\u4e8e\u8fdb\u884c\u673a\u5668\u7ffb\u8bd1":89,"\u5b83\u53ef\u80fd\u6709\u4e0d\u6b62\u4e00\u4e2a\u6743\u91cd":62,"\u5b83\u540c\u65f6\u5b66\u4e60\u6392\u5217":89,"\u5b83\u548c\u6570\u636e\u4f20\u5165\u51fd\u6570\u7684\u7b2c\u4e00\u4e2a\u53c2\u6570":2,"\u5b83\u5b58\u50a8\u5f53\u524d\u8282\u70b9\u6240\u6709\u8bad\u7ec3":68,"\u5b83\u5b9a\u4e49\u4e86":59,"\u5b83\u5b9a\u4e49\u4e86\u6a21\u578b\u53c2\u6570\u6539\u53d8\u7684\u89c4\u5219":47,"\u5b83\u5b9a\u4e49\u89e3\u7801\u7f51\u7edc\u7684":59,"\u5b83\u5c06\u88ab\u5206\u53d1\u5230":68,"\u5b83\u5c06\u8f93\u5165\u8bed\u53e5\u7f16\u7801\u4e3a\u5411\u91cf\u7684\u5e8f\u5217":89,"\u5b83\u5c06\u8fd4\u56de\u5982\u4e0b\u7684\u5b57\u5178":82,"\u5b83\u5c31\u4f1a\u5728\u6e90\u8bed\u53e5\u4e2d\u641c\u7d22\u51fa\u6700\u76f8\u5173\u4fe1\u606f\u7684\u4f4d\u7f6e\u7684\u96c6\u5408":89,"\u5b83\u5e76\u4e0d\u662f\u4e00\u4e2a\u5b8c\u6574\u7684recurr":46,"\u5b83\u652f\u6301\u591a\u7ebf\u7a0b\u66f4\u65b0":62,"\u5b83\u662finteger_value\u7c7b\u578b\u7684":56,"\u5b83\u662finteger_value_sequence\u7c7b\u578b\u7684":56,"\u5b83\u6709\u52a9\u4e8e\u5e2e\u52a9\u9891\u7e41\u4fee\u6539\u548c\u8bbf\u95ee\u5de5\u4f5c\u533a\u6587\u4ef6\u7684\u7528\u6237\u51cf\u5c11\u8d1f\u62c5":68,"\u5b83\u6a21\u62df\u4e86\u89e3\u7801\u7ffb\u8bd1\u8fc7\u7a0b\u4e2d\u5728\u6e90\u8bed\u53e5\u4e2d\u7684\u641c\u7d22":89,"\u5b83\u7684":59,"\u5b83\u7684\u529f\u80fd\u662f":63,"\u5b83\u7684\u6536\u655b\u901f\u5ea6\u6bd4":88,"\u5b83\u7684\u6bcf\u4e00\u4e2a\u5143\u7d20":55,"\u5b83\u7684\u76ee\u7684\u662f\u9884\u6d4b\u5728\u4e00\u4e2a\u5e8f\u5217\u4e2d\u8868\u8fbe\u7684\u60c5\u611f\u6001\u5ea6":88,"\u5b83\u7684\u8f93\u5165\u4e0e\u7ecf\u8fc7\u5b66\u4e60\u7684\u53c2\u6570\u505a\u5185\u79ef\u5e76\u52a0\u4e0a\u504f\u7f6e":62,"\u5b83\u76f4\u63a5\u5b66\u4e60\u6bb5\u843d\u8868\u793a":88,"\u5b83\u80fd\u591f\u4ece\u8bcd\u7ea7\u5230\u5177\u6709\u53ef\u53d8\u4e0a\u4e0b\u6587\u957f\u5ea6\u7684\u4e0a\u4e0b\u6587\u7ea7\u522b\u6765\u603b\u7ed3\u8868\u793a":88,"\u5b83\u8868\u793a":60,"\u5b83\u8bfb\u5165\u6570\u636e\u5e76\u5c06\u5b83\u4eec\u4f20\u8f93\u5230\u63a5\u4e0b\u6765\u7684\u7f51\u7edc\u5c42":47,"\u5b83\u8fd4\u56degen":89,"\u5b83\u8fd4\u56detrain":89,"\u5b83\u9700\u8981\u5728\u8fd9\u91cc\u6307\u5b9a":88,"\u5b83\u9996\u5148\u8c03\u7528\u57fa\u6784\u9020\u51fd\u6570":62,"\u5b89\u6392":56,"\u5b89\u88c5\u4e0e\u7f16\u8bd1":52,"\u5b89\u88c5\u540e\u7684\u76ee\u5f55\u7ed3\u6784\u4e3a":34,"\u5b89\u88c5\u597ddocker\u4e4b\u540e\u53ef\u4ee5\u4f7f\u7528\u6e90\u7801\u76ee\u5f55\u4e0b\u7684\u811a\u672c\u6784\u5efa\u6587\u6863":65,"\u5b89\u88c5\u5f00\u53d1\u5de5\u5177\u5230":60,"\u5b89\u88c5\u6587\u6863":51,"\u5b89\u88c5\u6d41\u7a0b":84,"\u5b89\u88c5\u8be5\u8f6f\u4ef6\u5305\u5c31\u53ef\u4ee5\u5728python\u73af\u5883\u4e0b\u5b9e\u73b0\u6a21\u578b\u9884\u6d4b":4,"\u5b89\u88c5paddlepaddl":84,"\u5b89\u88c5pillow":81,"\u5b89\u9759":56,"\u5b8c\u6210":61,"\u5b8c\u6210\u4e00\u4e2a\u4f20\u8f93\u52a8\u4f5c\u5b8c\u6210\u7684\u65f6\u95f4\u4e5f\u6bd4\u8f83\u77ed":28,"\u5b8c\u6210\u4efb\u610f\u7684\u8fd0\u7b97\u903b\u8f91":58,"\u5b8c\u6210\u540evolume\u4e2d\u7684\u6587\u4ef6\u5185\u5bb9\u5927\u81f4\u5982\u4e0b":75,"\u5b8c\u6210\u5f00\u53d1":49,"\u5b8c\u6210\u6570\u636e\u7684\u9884\u5904\u7406":21,"\u5b8c\u6210\u76f8\u5e94\u7684\u8ba1\u7b97":55,"\u5b8c\u6210paddlepaddle\u7684\u5b89\u88c5":51,"\u5b8c\u6574\u4ee3\u7801\u53ef\u4ee5\u53c2\u8003\u793a\u4f8b":46,"\u5b8c\u6574\u6e90\u7801\u53ef\u53c2\u8003":46,"\u5b8c\u6574\u7684\u4ee3\u7801\u89c1":4,"\u5b8c\u6574\u7684\u53c2\u6570\u77e9\u9635\u88ab\u5206\u5e03\u5728\u4e0d\u540c\u7684\u53c2\u6570\u670d\u52a1\u5668\u4e0a":62,"\u5b8c\u6574\u7684\u914d\u7f6e\u6587\u4ef6\u5728":59,"\u5b98\u65b9\u6587\u6863":48,"\u5b9a\u4e49\u4e00\u4e2a\u65f6\u95f4\u6b65\u4e4b\u5185rnn\u5355\u5143\u5b8c\u6210\u7684\u8ba1\u7b97":58,"\u5b9a\u4e49\u4e00\u4e2apython\u7684":2,"\u5b9a\u4e49\u4e86\u4e00\u4e2a\u53ea\u8bfb\u7684memori":58,"\u5b9a\u4e49\u4e86\u7f51\u7edc\u7684\u6570\u636e\u69fd":87,"\u5b9a\u4e49\u4e86\u7f51\u7edc\u7ed3\u6784":81,"\u5b9a\u4e49\u4e86\u7f51\u7edc\u7ed3\u6784\u5e76\u4fdd\u5b58\u4e3a":47,"\u5b9a\u4e49\u4e86lstm\u5355\u5143\u5728\u4e00\u4e2a\u65f6\u95f4\u6b65\u5185\u7684\u8ba1\u7b97\u8fc7\u7a0b":46,"\u5b9a\u4e49\u4f4d\u7f6e":63,"\u5b9a\u4e49\u4fdd\u6301\u4e00\u81f4":63,"\u5b9a\u4e49\u5728\u5916\u5c42":58,"\u5b9a\u4e49\u5f02\u6b65\u8bad\u7ec3\u7684\u957f\u5ea6":70,"\u5b9a\u4e49\u6570\u636e\u6765\u6e90":47,"\u5b9a\u4e49\u6e90\u8bed\u53e5\u7684\u6570\u636e\u5c42":59,"\u5b9a\u4e49\u7c7b\u578b":63,"\u5b9a\u4e49\u89e3\u7801\u5668\u7684memori":59,"\u5b9a\u4e49\u8bad\u7ec3\u6570\u6910\u548c\u6d4b\u8bd5\u6570\u6910\u63d0\u4f9b\u8005":88,"\u5b9a\u4e49\u8f93\u5165":63,"\u5b9a\u4e49\u8f93\u5165\u6570\u636e\u7684\u7c7b\u578b":47,"\u5b9a\u4e49\u8f93\u51fa":63,"\u5b9a\u4e49\u8f93\u51fa\u51fd\u6570":59,"\u5b9a\u4e49\u95e8\u63a7\u5faa\u73af\u5355\u5143\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u5355\u6b65\u51fd\u6570":59,"\u5b9d\u5854\u7684\u5e95\u7aef\u9700\u8981\u575a\u5b9e\u7684\u57fa\u5ea7\u6765\u652f\u6491":51,"\u5b9e\u4f8b\u5982\u4e0b":87,"\u5b9e\u73b0\u4e24\u4e2a\u5b8c\u5168\u7b49\u4ef7\u7684\u5168\u8fde\u63a5rnn":56,"\u5b9e\u73b0\u524d\u5411\u4f20\u64ad\u7684\u90e8\u5206\u6709\u4e0b\u9762\u51e0\u4e2a\u6b65\u9aa4":62,"\u5b9e\u73b0\u5355\u6b65\u51fd\u6570":59,"\u5b9e\u73b0\u540e\u5411\u4f20\u64ad\u7684\u90e8\u5206\u6709\u4e0b\u9762\u51e0\u4e2a\u6b65\u9aa4":62,"\u5b9e\u73b0\u5728":63,"\u5b9e\u73b0\u5bf9":64,"\u5b9e\u73b0\u6570\u636e\u8f93\u5165\u51fd\u6570":2,"\u5b9e\u73b0\u65b0\u7684op\u90fd\u6dfb\u52a0\u81f3\u76ee\u5f55":63,"\u5b9e\u73b0\u6784\u9020\u51fd\u6570":62,"\u5b9e\u73b0\u7684":46,"\u5b9e\u73b0\u7b80\u5355":33,"\u5b9e\u73b0\u7ec6\u8282":62,"\u5b9e\u73b0\u7f51\u7edc\u5c42\u7684\u524d\u5411\u4f20\u64ad":62,"\u5b9e\u73b0\u7f51\u7edc\u5c42\u7684\u540e\u5411\u4f20\u64ad":62,"\u5b9e\u73b0\u8bcd\u8bed\u548c\u53e5\u5b50\u4e24\u4e2a\u7ea7\u522b\u7684\u53cc\u5c42rnn\u7ed3\u6784":58,"\u5b9e\u73b0\u8be5\u5c42\u7684c":62,"\u5b9e\u9645\u4e0a":49,"\u5b9e\u9645\u4e0a\u4f7f\u7528\u4e86":46,"\u5b9e\u9645\u4e0a\u53ea\u6709":82,"\u5b9e\u9645\u4e0a\u662fcsv\u6587\u4ef6":85,"\u5b9e\u9645\u4e0a\u9700\u8981\u7684\u8f93\u51fa\u7ed3\u679c\u662f\u4e24\u4e2a\u77e9\u9635":46,"\u5ba2\u6237":56,"\u5ba2\u6237\u670d\u52a1":85,"\u5bb6":56,"\u5bb9\u5668":[49,73],"\u5bb9\u5668\u4e0d\u4f1a\u4fdd\u7559\u5728\u8fd0\u884c\u65f6\u751f\u6210\u7684\u6570\u636e":73,"\u5bb9\u5668\u4f7f\u7528":49,"\u5bb9\u5668\u8fd0\u884c\u90fd\u8fd0\u884c":75,"\u5bbd\u5ea6\u7b49\u4e8e\u914d\u7f6e\u4e2dlayer\u7684s":46,"\u5bbf\u4e3b\u673a\u662flinux":54,"\u5bbf\u4e3b\u673a\u7684c":[53,54],"\u5bbf\u4e3b\u673a\u76ee\u5f55":73,"\u5bc4\u5b58\u5668\u4f7f\u7528\u60c5\u51b5\u548c\u5171\u4eab\u5185\u5b58\u4f7f\u7528\u60c5\u51b5\u80fd\u8ba9\u6211\u4eec\u5bf9gpu\u7684\u6574\u4f53\u4f7f\u7528\u6709\u66f4\u597d\u7684\u7406\u89e3":67,"\u5bf9":56,"\u5bf9\u4e00\u4e2a5\u7ef4\u975e\u5e8f\u5217\u7684\u7a00\u758f01\u5411\u91cf":[2,51],"\u5bf9\u4e00\u4e2a5\u7ef4\u975e\u5e8f\u5217\u7684\u7a00\u758f\u6d6e\u70b9\u5411\u91cf":[2,51],"\u5bf9\u4e8e":59,"\u5bf9\u4e8e\u4e0d\u540c\u8bed\u8a00":33,"\u5bf9\u4e8e\u4e24\u79cd\u4e0d\u540c\u7684\u8f93\u5165\u6570\u636e\u7c7b\u578b":56,"\u5bf9\u4e8e\u4e60\u60ef\u4f7f\u7528windows\u548cmacos\u7684\u5f00\u53d1\u8005\u6765\u8bf4":60,"\u5bf9\u4e8e\u5185\u5b58\u8f83\u5c0f\u7684\u673a\u5668":2,"\u5bf9\u4e8e\u5355\u5c42rnn":56,"\u5bf9\u4e8e\u5355\u5c42rnn\u7684\u6570\u636e\u4e00\u5171\u6709\u4e24\u4e2a\u6837\u672c":56,"\u5bf9\u4e8e\u53cc\u5c42rnn":56,"\u5bf9\u4e8e\u540c\u4e00\u6bb5c":33,"\u5bf9\u4e8e\u540c\u6837\u7684\u6570\u636e":56,"\u5bf9\u4e8e\u591a\u8bed\u8a00\u63a5\u53e3":33,"\u5bf9\u4e8e\u5927\u591a\u6570\u8bed\u8a00":33,"\u5bf9\u4e8e\u6211\u4eec\u652f\u6301\u7684\u5168\u90e8\u77e9\u9635\u64cd\u4f5c":62,"\u5bf9\u4e8e\u6811\u7684\u6bcf\u4e00\u5c42":89,"\u5bf9\u4e8e\u6bb5\u843d\u7684\u6587\u672c\u5206\u7c7b":56,"\u5bf9\u4e8e\u6bcf\u4e00\u4e2a\u5355\u5c42rnn\u7684\u6570\u636e":56,"\u5bf9\u4e8e\u6bcf\u4f4d\u7528\u6237":86,"\u5bf9\u4e8e\u6bcf\u79cd\u7c7b\u578b":34,"\u5bf9\u4e8e\u6bcf\u79cdc":34,"\u5bf9\u4e8e\u7ed9\u5b9a\u7684\u4e00\u6761\u6587\u672c":84,"\u5bf9\u4e8e\u914d\u5907\u6709\u6ce8\u610f\u529b\u673a\u5236\u7684\u89e3\u7801\u5668":59,"\u5bf9\u4e8eamazon":84,"\u5bf9\u4e8erasspberri":54,"\u5bf9\u4ee3\u7801\u8fdb\u884c\u6027\u80fd\u5206\u6790":67,"\u5bf9\u5168\u8fde\u63a5\u5c42\u6765\u8bf4":62,"\u5bf9\u52a0\u8f7d\u9884\u8bad\u7ec3\u53c2\u6570\u7684\u5c42":46,"\u5bf9\u56fe\u7247\u8fdb\u884c\u9884\u5904\u7406":81,"\u5bf9\u5df2\u7ecfpush\u5230\u8fdc\u7a0b\u4ed3\u5e93\u7684\u591a\u4e2acommit":61,"\u5bf9\u5e94\u4e00\u4e2a\u5b50\u53e5":58,"\u5bf9\u5e94\u4e00\u4e2a\u8bcd":58,"\u5bf9\u5e94\u4e8e\u5b57\u5178":80,"\u5bf9\u5e94\u7684":2,"\u5bf9\u5e94\u7684\u68af\u5ea6op\u8ba1\u7b97\u4e4b\u4e2d":63,"\u5bf9\u6570\u636e\u96c6\u8fdb\u884c\u9884\u5904\u7406\u7684\u57fa\u672c\u547d\u4ee4\u662f":89,"\u5bf9\u6574\u4e2a\u65b0\u5411\u91cf\u96c6\u5408\u7684\u6bcf\u4e00\u4e2a\u7ef4\u5ea6\u53d6\u6700\u5927\u503c\u6765\u8868\u793a\u6700\u540e\u7684\u53e5\u5b50":84,"\u5bf9\u6587\u6863\u5904\u7406\u540e\u5f62\u6210\u7684\u5355\u8bcd\u5411\u91cf":88,"\u5bf9\u673a\u5668\u7ffb\u8bd1\u7684\u4eba\u5de5\u8bc4\u4f30\u5de5\u4f5c\u5f88\u5e7f\u6cdb\u4f46\u4e5f\u5f88\u6602\u8d35":89,"\u5bf9\u6bcf\u4e2a\u8f93\u5165":62,"\u5bf9\u6bcf\u4e2a\u8f93\u5165\u4e58\u4e0a\u53d8\u6362\u77e9\u9635":62,"\u5bf9\u6bd4":33,"\u5bf9\u6bd4\u53cd\u5411op\u4e0d\u540c\u8bbe\u5907":63,"\u5bf9\u6fc0\u6d3b\u6c42\u5bfc":62,"\u5bf9\u7528\u6237\u6765\u8bf4":2,"\u5bf9\u8bad\u7ec3\u6570\u636e\u8fdb\u884cshuffl":2,"\u5bf9\u8bc4\u5ba1\u610f\u89c1\u4e0d\u540c\u610f\u7684":61,"\u5bf9\u8bc4\u5ba1\u610f\u89c1\u540c\u610f\u4e14\u6309\u5176\u4fee\u6539\u5b8c\u7684":61,"\u5bf9\u8be5\u5411\u91cf\u8fdb\u884c\u975e\u7ebf\u6027\u53d8\u6362":84,"\u5bf9\u8c61":46,"\u5bf9\u8c61\u5b58\u50a8\u4e3a\u6587\u4ef6":86,"\u5bf9\u8f93\u5165\u53c2\u6570\u7684\u5b89\u5168\u6027\u8fdb\u884c\u4e86\u5fc5\u8981\u7684\u5224\u65ad":34,"\u5bf9\u8f93\u51fa\u7684\u5408\u5e76":58,"\u5bf9\u8fd9\u4e2a\u7248\u672c\u7684\u63d0\u4ea4":42,"\u5bf9\u9762":56,"\u5bf9check":2,"\u5bf9sparse_binary_vector\u548csparse_float_vector":[2,51],"\u5bfc\u51fa\u8fd9\u4e9b\u63a5\u53e3":34,"\u5bfc\u81f4\u4e86\u6d6e\u70b9\u6570\u6ea2\u51fa":46,"\u5bfc\u81f4\u53c2\u6570\u6536\u655b\u5230\u4e86\u4e00\u4e9b\u5947\u5f02\u7684\u60c5\u51b5":46,"\u5bfc\u81f4\u53c2\u6570\u7d2f\u52a0":46,"\u5bfc\u81f4\u7f16\u8bd1paddlepaddle\u5931\u8d25":46,"\u5bfc\u81f4\u8bad\u7ec3\u65f6\u95f4\u8fc7\u957f":46,"\u5c01\u88c5\u4e86":67,"\u5c01\u88c5\u8be5\u5c42\u7684python\u63a5\u53e3":62,"\u5c06":[2,42,46,67,84],"\u5c06\u4e00\u4e2a\u795e\u7ecf\u7f51\u7edc\u53c2\u6570\u62c6\u5206\u6210\u591a\u4efd":20,"\u5c06\u4e0a\u4e00\u65f6\u95f4\u6b65\u6240\u751f\u6210\u7684\u8bcd\u7684\u5411\u91cf\u6765\u4f5c\u4e3a\u5f53\u524d\u65f6\u95f4\u6b65\u7684\u8f93\u5165":59,"\u5c06\u4e3b\u673a\u4e0a\u7684\u6587\u4ef6\u6216\u76ee\u5f55\u6302\u8f7d\u5230":49,"\u5c06\u4ed6\u4eec\u79fb\u52a8\u5230\u76ee\u5f55":86,"\u5c06\u4f1a\u5c06\u7528\u6237\u4f20\u8fdb\u6765\u7684\u914d\u7f6e\u53c2\u6570\u4f20\u9012cmake\u7cfb\u7edf":53,"\u5c06\u4f1a\u81ea\u52a8\u8ba1\u7b97\u51fa\u4e00\u4e2a\u5408\u9002\u7684\u503c":70,"\u5c06\u5176\u8bbe\u7f6e\u6210":46,"\u5c06\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217\u6570\u636e\u5148\u53d8\u6362\u6210\u5355\u5c42\u65f6\u95f4\u5e8f\u5217\u6570\u636e":56,"\u5c06\u542b\u6709\u5b50\u53e5":58,"\u5c06\u542b\u6709\u8bcd\u8bed\u7684\u53e5\u5b50\u5b9a\u4e49\u4e3a\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":58,"\u5c06\u56fe\u7247\u6309\u7167\u4e0a\u8ff0\u7ed3\u6784\u5b58\u50a8\u597d\u4e4b\u540e":81,"\u5c06\u5728":81,"\u5c06\u5728\u8fd0\u884c\u65f6\u62a5\u9519":68,"\u5c06\u5916\u90e8\u7684\u5b58\u50a8\u670d\u52a1\u5728kubernetes\u4e2d\u63cf\u8ff0\u6210\u4e3a\u7edf\u4e00\u7684\u8d44\u6e90\u5f62\u5f0f":73,"\u5c06\u591a\u53e5\u8bdd\u770b\u6210\u4e00\u4e2a\u6574\u4f53\u540c\u65f6\u4f7f\u7528encoder\u538b\u7f29":56,"\u5c06\u591a\u53f0\u673a\u5668\u7684\u6d4b\u8bd5\u7ed3\u679c\u5408\u5e76":70,"\u5c06\u5927\u91cf\u7684":33,"\u5c06\u5b57\u5178\u7684\u5730\u5740\u4f5c\u4e3aargs\u4f20\u7ed9dataprovid":46,"\u5c06\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u524d\u5411\u548c\u53cd\u5411\u90e8\u5206\u6df7\u5408\u5728\u4e00\u8d77":59,"\u5c06\u6570\u636e\u5904\u7406\u6210\u89c4\u8303\u683c\u5f0f":80,"\u5c06\u6570\u636e\u7ec4\u5408\u6210batch\u8fdb\u884c\u8bad\u7ec3":2,"\u5c06\u6570\u636e\u8f6c\u6362\u4e3apaddle\u7684\u683c\u5f0f":81,"\u5c06\u65b0\u5206\u652f\u7684\u7248\u672c\u6253\u4e0atag":42,"\u5c06\u65b0\u5efa\u7684\u6743\u91cd\u52a0\u5165\u6743\u91cd\u8868":62,"\u5c06\u65e5\u5fd7\u5199\u5165\u6587\u4ef6":86,"\u5c06\u660e\u6587\u53c2\u6570\u8f6c\u5316\u4e3apaddlepaddle\u53ef\u52a0\u8f7d\u7684\u6a21\u578b\u53c2\u6570\u65f6":46,"\u5c06\u672c\u5730\u7684\u4fee\u6539\u63a8\u9001\u5230":61,"\u5c06\u6837\u672c\u7684\u5730\u5740\u653e\u5165\u53e6\u4e00\u4e2a\u6587\u672c\u6587\u4ef6":2,"\u5c06\u6b64\u76ee\u5f55\u6302\u8f7d\u4e3a\u5bb9\u5668\u7684":75,"\u5c06\u6bcf\u4e2a\u6e90\u8bed\u8a00\u5230\u76ee\u6807\u8bed\u8a00\u7684\u5e73\u884c\u8bed\u6599\u5e93\u6587\u4ef6\u5408\u5e76\u4e3a\u4e00\u4e2a\u6587\u4ef6":89,"\u5c06\u73af\u5883\u53d8\u91cf\u8f6c\u6362\u6210paddle\u7684\u547d\u4ee4\u884c\u53c2\u6570":75,"\u5c06\u7528\u4e8epython":63,"\u5c06\u7ed3\u679c\u4fdd\u5b58\u5230\u6b64\u76ee\u5f55\u91cc":75,"\u5c06\u884c\u4e2d\u7684\u6570\u636e\u8f6c\u6362\u6210\u4e0einput_types\u4e00\u81f4\u7684\u683c\u5f0f":2,"\u5c06\u88ab\u5206\u4e3a":80,"\u5c06\u8bad\u7ec3\u6587\u4ef6\u4e0e\u5207\u5206\u597d\u7684\u6570\u636e\u4e0a\u4f20\u5230\u5171\u4eab\u5b58\u50a8":75,"\u5c06\u8be5\u53e5\u8bdd\u5305\u542b\u7684\u6240\u6709\u5355\u8bcd\u5411\u91cf\u6c42\u5e73\u5747":84,"\u5c06\u8df3\u8fc7\u5206\u53d1\u9636\u6bb5\u76f4\u63a5\u542f\u52a8\u6240\u6709\u8282\u70b9\u7684\u96c6\u7fa4\u4f5c\u4e1a":68,"\u5c06\u8fd9\u79cd\u8de8\u8d8a\u65f6\u95f4\u6b65\u7684\u8fde\u63a5\u7528\u4e00\u4e2a\u7279\u6b8a\u7684\u795e\u7ecf\u7f51\u7edc\u5355\u5143\u5b9e\u73b0":56,"\u5c06\u8fdc\u7a0b\u4ed3\u5e93":61,"\u5c06\u900f\u660e":68,"\u5c06\u9700\u8981\u8f93\u51fa\u7684\u5c42\u4f5c\u4e3a":46,"\u5c06ip\u6392\u5e8f\u751f\u6210\u7684\u5e8f\u53f7\u4f5c\u4e3atrain":75,"\u5c06master\u5206\u652f\u7684\u5408\u5165commit\u6253\u4e0atag":42,"\u5c06paddlepaddle\u4fdd\u5b58\u7684\u6a21\u578b\u53c2\u6570\u8fd8\u539f\u56de\u660e\u6587\u65f6":46,"\u5c06recurr":46,"\u5c11\u4e8e5":68,"\u5c1a\u53ef":56,"\u5c31":56,"\u5c31\u4f1a\u751f\u6210\u975e\u5e38\u591a\u7684gener":2,"\u5c31\u53ef\u4ee5\u518d\u8fd0\u884c\u53e6\u4e00\u4e2anginx":49,"\u5c31\u53ef\u4ee5\u5c06\u6570\u636e\u4f20\u9001\u7ed9paddlepaddle\u4e86":2,"\u5c31\u53ef\u4ee5\u5c06\u8fd9\u4e9b\u6587\u4ef6\u6301\u4e45\u5316\u5b58\u50a8":73,"\u5c31\u53ef\u4ee5\u6309":60,"\u5c31\u5f88\u5bb9\u6613\u5bfc\u81f4\u5185\u5b58\u8d85\u9650":46,"\u5c31\u662f":56,"\u5c31\u662f\u6a21\u578b\u7684\u53c2\u6570":47,"\u5c31\u662f\u7528\u4e8e\u5c55\u793a\u4e0a\u8ff0\u5206\u6790\u5de5\u5177\u7684\u7528\u6cd5":67,"\u5c31\u80fd\u591f\u5f88\u65b9\u4fbf\u7684\u5b8c\u6210\u6570\u636e\u4e0b\u8f7d\u548c\u76f8\u5e94\u7684\u9884\u5904\u7406\u5de5\u4f5c":84,"\u5c31\u8fd9\u4e48\u7b80\u5355":49,"\u5c31\u901a\u5e38\u7684gpu\u6027\u80fd\u5206\u6790\u6765\u8bf4":67,"\u5c31\u9700\u8981\u5bf9\u8fd9\u4e2a\u7b2c\u4e09\u65b9\u8bed\u8a00\u589e\u52a0\u4e00\u4e9b\u5b9a\u4e49":33,"\u5c31\u9700\u8981\u9009\u62e9\u4f7f\u7528no":49,"\u5c3a\u5bf8":82,"\u5c3d\u65e9\u62a5\u9519":63,"\u5c40\u90e8\u5173\u8054\u6027\u8d28\u548c\u7a7a\u95f4\u4e0d\u53d8\u6027\u8d28":81,"\u5c42\u540e\u5f97\u5230\u6df1\u5ea6":87,"\u5c42\u548c\u8f93\u5165\u7684\u914d\u7f6e":62,"\u5c42\u6743\u91cd":82,"\u5c42\u6b21\u5316\u7684rnn":58,"\u5c42\u7279\u5f81":82,"\u5c42\u7684\u540d\u79f0\u4e0e":59,"\u5c42\u7684\u5927\u5c0f":62,"\u5c42\u7684\u7279\u5f81":82,"\u5c42\u7684\u7c7b\u578b":62,"\u5c42\u7684\u8f93\u5165":87,"\u5c42\u7684\u8f93\u5165\u548c\u8f93\u51fa\u4f5c\u4e3a\u4e0b\u4e00\u4e2a":87,"\u5c42\u7684\u8f93\u51fa\u88ab\u7528\u4f5c\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684":59,"\u5c42\u7ec4\u6210\u4e00\u5bf9":87,"\u5c45\u7136":56,"\u5c55\u793a\u4e86\u4e00\u79cd\u65b9\u6cd5":89,"\u5c55\u793a\u4e86\u4e0a\u8ff0\u7f51\u7edc\u6a21\u578b\u7684\u8bad\u7ec3\u6548\u679c":84,"\u5c55\u793a\u4e86\u5982\u4f55\u5c06\u6bcf\u4e2a\u7279\u5f81\u6620\u5c04\u5230\u4e00\u4e2a\u5411\u91cf":86,"\u5c5e\u4e8e\u8fd9\u4e00\u7c7b\u7684\u5b9e\u73b0":46,"\u5c5e\u6027":[63,87],"\u5d4c\u5165\u5c42":86,"\u5d4c\u5165\u7279\u5f81\u5b57\u5178":86,"\u5d4c\u5165\u7f16\u53f7\u4f1a\u6839\u636e\u5355\u8bcd\u6392\u5e8f":86,"\u5de5\u4f5c\u6a21\u5f0f":70,"\u5de5\u4f5c\u7a7a\u95f4":68,"\u5de5\u4f5c\u7a7a\u95f4\u4e2d\u7684":68,"\u5de5\u4f5c\u7a7a\u95f4\u6839\u76ee\u5f55":68,"\u5de5\u4f5c\u7a7a\u95f4\u76ee\u5f55\u7684\u5de5\u4f5c\u7a7a\u95f4":68,"\u5de5\u4f5c\u7a7a\u95f4\u914d\u7f6e":68,"\u5de5\u5177":88,"\u5de5\u5177\u4e2d\u7684\u811a\u672c":88,"\u5de5\u5177\u6765\u7ba1\u7406":61,"\u5de5\u5177\u94fe\u7684android":53,"\u5de5\u7a0b\u5e08":85,"\u5de6\u56fe\u6784\u9020\u7f51\u7edc\u6a21\u5757\u7684\u65b9\u5f0f\u88ab\u7528\u4e8e34\u5c42\u7684\u7f51\u7edc\u4e2d":82,"\u5de6\u8fb9\u662f":82,"\u5dee\u8bc4":84,"\u5df2\u6253\u5f00":61,"\u5df2\u7ecf\u5728\u96c6\u7fa4\u63d0\u4ea4\u73af\u5883\u4e2d\u5b8c\u6210\u8bbe\u7f6e":70,"\u5df2\u7ecf\u63d0\u4f9b\u4e86\u811a\u672c\u6765\u5e2e\u52a9\u60a8\u521b\u5efa\u8fd9\u4e24\u4e2a\u6587\u4ef6":68,"\u5e02\u573a":85,"\u5e02\u9762\u4e0a\u5df2\u7ecf\u6709nvidia\u6216\u7b2c\u4e09\u65b9\u63d0\u4f9b\u7684\u4f17\u591a\u5de5\u5177":67,"\u5e0c\u671b\u80fd\u8ba9\u6211\u4eec\u77e5\u6653":86,"\u5e26\u6709\u4e0b\u9762\u4e24\u4e2a\u6a21\u677f\u53c2\u6570":63,"\u5e2e\u52a9\u6211\u4eec\u5b8c\u6210\u5bf9\u8f93\u5165\u5e8f\u5217\u7684\u62c6\u5206":58,"\u5e2e\u52a9\u6211\u4eec\u66f4\u597d\u5730\u63cf\u8ff0\u6bb5\u843d":58,"\u5e2e\u52a9\u6211\u4eec\u6784\u9020\u4e00\u4e9b\u590d\u6742\u7684\u8f93\u5165\u4fe1\u606f":55,"\u5e38\u5e38\u51fa\u73b0":46,"\u5e38\u7528\u4f18\u5316\u7b97\u6cd5\u5305\u62ecmomentum":84,"\u5e38\u89c1\u7684\u53ef\u9009\u5b58\u50a8\u670d\u52a1\u5305\u62ec":73,"\u5e72\u51c0":56,"\u5e73\u53f0\u4e3a\u60f3\u89c2\u6d4b\u8bcd\u5411\u91cf\u7684\u7528\u6237\u63d0\u4f9b\u4e86\u5c06\u4e8c\u8fdb\u5236\u8bcd\u5411\u91cf\u6a21\u578b\u8f6c\u6362\u4e3a\u6587\u672c\u6a21\u578b\u7684\u529f\u80fd":80,"\u5e73\u5747\u6545\u969c\u4fee\u590d\u65f6\u95f4":20,"\u5e73\u5747\u6545\u969c\u7387":20,"\u5e73\u5747\u7279\u5f81\u56fe\u7684\u9ad8\u5ea6\u53ca\u5bbd\u5ea6":81,"\u5e74\u9f84":85,"\u5e74\u9f84\u4ece\u4e0b\u5217\u5217\u8868\u8303\u56f4\u4e2d\u9009\u53d6":85,"\u5e74\u9f84\u548c\u804c\u4e1a":86,"\u5e76\u4e0d\u4fdd\u8bc1":62,"\u5e76\u4e0d\u662f\u4f7f\u7528\u53cc\u5c42rnn\u89e3\u51b3\u5b9e\u9645\u7684\u95ee\u9898":56,"\u5e76\u4e0d\u662fkubernetes\u4e2d\u7684node\u6982\u5ff5":75,"\u5e76\u4e0d\u771f\u6b63\u7684\u548c":56,"\u5e76\u4e0d\u96be":60,"\u5e76\u4e14":[2,59],"\u5e76\u4e14\u4e66\u5199\u4e00\u4efd\u4ee3\u7801":64,"\u5e76\u4e14\u4f7f\u7528":34,"\u5e76\u4e14\u5185\u5c42\u7684\u5e8f\u5217\u64cd\u4f5c\u4e4b\u95f4\u72ec\u7acb\u65e0\u4f9d\u8d56":56,"\u5e76\u4e14\u5206\u522b\u91cd\u547d\u540d\u6587\u4ef6\u540e\u7f00":89,"\u5e76\u4e14\u52a0\u4e0a\u4e0b\u9762\u7684\u547d\u4ee4\u884c\u53c2\u6570":72,"\u5e76\u4e14\u5305\u62ecunit":61,"\u5e76\u4e14\u53ea\u6709\u4e00\u4e2a\u6743\u91cd":82,"\u5e76\u4e14\u53ef\u80fd\u4f1a\u52a0\u901f\u8bad\u7ec3\u8fc7\u7a0b":46,"\u5e76\u4e14\u540e\u7eed\u4ecd\u5728\u4e0d\u65ad\u6539\u8fdb":47,"\u5e76\u4e14\u542f\u52a8\u540e\u6267\u884c":49,"\u5e76\u4e14\u542f\u52a8\u8bad\u7ec3":75,"\u5e76\u4e14\u5728\u5185\u5b58\u8db3\u591f\u7684\u60c5\u51b5\u4e0b\u8d8a\u5927\u8d8a\u597d":2,"\u5e76\u4e14\u5728\u5e38\u89c1\u7684\u5e73\u53f0\u4e0a":33,"\u5e76\u4e14\u5728\u968f\u540e\u7684\u8bfb\u53d6\u6570\u636e\u8fc7\u7a0b\u4e2d\u586b\u5145\u8bcd\u8868":84,"\u5e76\u4e14\u5728dataprovider\u4e2d\u5b9e\u73b0\u5982\u4f55\u8bbf\u95ee\u8bad\u7ec3\u6587\u4ef6\u5217\u8868":1,"\u5e76\u4e14\u5b83\u4eec\u7684\u987a\u5e8f\u4e0e":82,"\u5e76\u4e14\u5bf9\u7528\u6237\u7684\u7279\u5f81\u505a\u540c\u6837\u7684\u64cd\u4f5c":86,"\u5e76\u4e14\u5c06\u9884\u5904\u7406\u597d\u7684\u6570\u636e\u96c6\u5b58\u653e\u5728":89,"\u5e76\u4e14\u5c55\u793a\u4e86\u5982\u4f55\u5229\u7528paddlepaddle\u6765\u89e3\u51b3\u4e00\u4e2a\u7ecf\u5178\u7684\u7ebf\u6027\u56de\u5f52\u95ee\u9898":51,"\u5e76\u4e14\u628a\u5404\u79cd\u5f00\u53d1\u5de5\u5177\u5b89\u88c5\u8fdb\u53bb":60,"\u5e76\u4e14\u628a\u7cfb\u7edf\u751f\u6210\u7684ca":28,"\u5e76\u4e14\u628a\u7ed3\u679c\u8fd4\u56depfsclient\u7aef":28,"\u5e76\u4e14\u67e5\u8be2paddlepaddle\u5355\u5143\u6d4b\u8bd5\u7684\u65e5\u5fd7":46,"\u5e76\u4e14\u7b2c\u4e8c\u4e2a\u662f\u53cd\u5411lstm":88,"\u5e76\u4e14\u7f16\u8bd1\u65f6\u9700\u8981\u6253\u5f00":63,"\u5e76\u4e14\u7f16\u8bd1\u80fd\u901a\u8fc7\u4ee3\u7801\u6837\u5f0f\u68c0\u67e5":61,"\u5e76\u4e14\u8ba9\u63a5\u53e3\u8131\u79bb\u5b9e\u73b0\u7ec6\u8282":33,"\u5e76\u4e14\u8bbe\u7f6e\u9ed8\u8ba4\u503c\u4e3a1":63,"\u5e76\u4e14\u8f93\u51fa\u4e00\u4e2a":61,"\u5e76\u4e14\u8fd0\u884c":60,"\u5e76\u4e14\u901a\u8fc7\u7ed9\u51fa\u5f53\u524d\u76ee\u6807\u5355\u8bcd\u6765\u9884\u6d4b\u4e0b\u4e00\u4e2a\u76ee\u6807\u5355\u8bcd":89,"\u5e76\u4e14\u96c6\u7fa4\u4f5c\u4e1a\u4e2d\u7684\u6240\u6709\u8282\u70b9\u5c06\u5728\u6b63\u5e38\u60c5\u51b5\u4e0b\u5904\u7406\u5177\u6709\u76f8\u540c\u903b\u8f91\u4ee3\u7801\u7684\u6587\u4ef6":68,"\u5e76\u4e14\u9700\u8981\u91cd\u5199\u57fa\u7c7b\u4e2d\u7684\u4ee5\u4e0b\u51e0\u4e2a\u865a\u51fd\u6570":62,"\u5e76\u4e14cpu":63,"\u5e76\u4e14softmax\u5c42\u7684\u4e24\u4e2a\u8f93\u5165\u4e5f\u4f7f\u7528\u4e86\u540c\u6837\u7684\u53c2\u6570":46,"\u5e76\u4f20\u5165\u76f8\u5e94\u7684\u547d\u4ee4\u884c\u53c2\u6570\u521d\u59cb\u5316paddlepaddl":4,"\u5e76\u4f7f\u7528":87,"\u5e76\u4f7f\u7528\u4e86dropout":84,"\u5e76\u4f7f\u7528\u8fd9\u4e2a\u795e\u7ecf\u7f51\u7edc\u6765\u5bf9\u56fe\u7247\u8fdb\u884c\u5206\u7c7b":81,"\u5e76\u521b\u5efa\u4e86\u4e00\u4e2a\u65b0\u6587\u4ef6":61,"\u5e76\u521b\u5efaoptim":51,"\u5e76\u521d\u59cb\u5316":63,"\u5e76\u5220\u9664":42,"\u5e76\u5220\u9664\u66f4\u65e9\u7684\u5feb\u7167":20,"\u5e76\u52a0\u8f7d\u5176\u4e2d\u7684\u53c2\u6570":20,"\u5e76\u53d1\u5e03\u5230pypi":42,"\u5e76\u5728\u4e58\u79ef\u7ed3\u679c\u4e0a\u518d\u52a0\u4e0a\u7ef4\u5ea6\u4e3a":62,"\u5e76\u5728\u6700\u5f00\u59cb\u521d\u59cb\u5316\u4e3a\u8d77\u59cb\u8bcd":59,"\u5e76\u5728\u7b14\u8bb0\u672c\u4e0a\u901a\u8fc7ssh\u4e0e\u5176\u8fde\u63a5":49,"\u5e76\u5728\u7c7b\u6784\u5efa\u51fd\u6570\u4e2d\u628a\u5b83\u653e\u5165\u4e00\u4e2a\u7c7b\u6210\u5458\u53d8\u91cf\u91cc":62,"\u5e76\u5728\u8be5layer\u91cc\u91c7\u7528\u7b2c\u4e00\u79cd\u65b9\u5f0f\u8bbe\u7f6e":46,"\u5e76\u5728\u96c6\u7fa4\u4e2d\u8fd0\u884c\u591a\u4e2a\u5206\u5e03\u5f0f\u6570\u636e\u5904\u7406\u4efb\u52a1":21,"\u5e76\u5728python\u811a\u672c\u4e2d\u5b8c\u6210\u4e0eoperator\u540c\u6837\u7684\u8ba1\u7b97\u903b\u8f91":63,"\u5e76\u5b89\u88c5\u4e86python":46,"\u5e76\u5b9a\u4e49\u5143\u7c7b":63,"\u5e76\u5bf9\u6bd4\u662f\u5426\u548c\u6b63\u5728\u5b89\u88c5\u7684\u540e\u7f00\u4e00\u81f4":46,"\u5e76\u5bf9\u76f8\u5e94\u7684\u53c2\u6570\u8c03\u7528":62,"\u5e76\u5c06\u5176\u6295\u5c04\u5230":59,"\u5e76\u5c06\u5b83\u4eec\u6309\u7167\u542f\u53d1\u4ee3\u4ef7":89,"\u5e76\u5c06\u5b83\u4eec\u653e\u5728":89,"\u5e76\u5c06\u6bcf\u8f6e\u7684\u6a21\u578b\u7ed3\u679c\u5b58\u653e\u5728":47,"\u5e76\u5c06\u8be5layer\u4e0a\u4e00\u65f6\u95f4\u6b65\u7684\u8f93\u51fa\u4f5c\u4e3a\u81ea\u8eab\u5f53\u524d\u65f6\u95f4\u6b65\u7684\u8f93\u51fa":46,"\u5e76\u5c06c":34,"\u5e76\u5c06develop\u548ctest\u6570\u636e\u5206\u522b\u653e\u5165\u4e0d\u540c\u7684\u6587\u4ef6\u5939":89,"\u5e76\u60f3\u4f7f\u7528gpu\u6765\u8bad\u7ec3\u8bbe\u7f6e\u4e3atru":88,"\u5e76\u628a\u5feb\u7167\u4fdd\u5b58\u5230\u8fd9\u4e2a\u76ee\u5f55\u4e0b":20,"\u5e76\u6307\u5b9abatch":89,"\u5e76\u63d0\u4f9b\u4e86\u7b80\u5355\u7684cache\u529f\u80fd":2,"\u5e76\u66f4\u6362job":46,"\u5e76\u6b22\u8fce\u60a8\u6765\u53c2\u4e0e\u8d21\u732e":88,"\u5e76\u6ca1\u6709paddle\u7279\u522b\u9700\u8981\u7684\u7279\u6027":33,"\u5e76\u6dfb\u52a0\u6ce8\u91ca":63,"\u5e76\u7279\u5316\u6a21\u677f\u53c2\u6570\u4e3a":63,"\u5e76\u7531":87,"\u5e76\u7ed9\u51fa\u5206\u7c7b\u7ed3\u679c":81,"\u5e76\u7ed9\u51fa\u7684\u76f8\u5173\u6a21\u578b\u683c\u5f0f\u7684\u5b9a\u4e49":80,"\u5e76\u81ea\u52a8\u7f16\u8bd1\u5bbf\u4e3b\u673a\u7248protoc\u53ef\u6267\u884c\u6587\u4ef6":[53,54],"\u5e76\u88ab\u53cd\u5411\u5904\u7406":87,"\u5e76\u88ab\u5b58\u50a8\u5728\u8bf8\u5982hadoop":21,"\u5e76\u89c2\u5bdf\u7ed3\u679c":67,"\u5e76\u89e3\u91ca\u4e86\u5404\u81ea\u542b\u4e49":63,"\u5e76\u8bb0\u5f55\u5b83\u7684\u7f16\u53f7":61,"\u5e76\u8bbe\u7f6e":68,"\u5e76\u8fdb\u884c\u521d\u59cb\u5316\u64cd\u4f5c":51,"\u5e76\u9002\u5e94github\u7684\u7279\u6027\u505a\u4e86\u4e00\u4e9b\u533a\u522b":42,"\u5e76\u9010\u6e10\u5c55\u793a\u66f4\u52a0\u6df1\u5165\u7684\u529f\u80fd":84,"\u5e76\u91cd\u65b0\u6253\u5305wheel\u5305":42,"\u5e76\u94fe\u63a5\u5230\u751f\u6210\u7684lib\u5e93\u4e2d":63,"\u5e8a\u4e0a\u7528\u54c1":56,"\u5e8a\u57ab":56,"\u5e8f\u5217\u4e2d\u542b\u6709\u5143\u7d20\u7684\u6570\u76ee\u540c":55,"\u5e8f\u5217\u5230\u5e8f\u5217":89,"\u5e8f\u5217\u6570\u636e\u662f\u81ea\u7136\u8bed\u8a00\u5904\u7406\u4efb\u52a1\u9762\u5bf9\u7684\u4e00\u79cd\u4e3b\u8981\u8f93\u5165\u6570\u636e\u7c7b\u578b":58,"\u5e8f\u5217\u662f\u4e00\u79cd\u5e38\u89c1\u7684\u6570\u636e\u7c7b\u578b":55,"\u5e8f\u5217\u751f\u6210\u4efb\u52a1\u5927\u591a\u9075\u5faaencod":58,"\u5e8f\u5217\u751f\u6210\u4efb\u52a1\u7684\u8f93\u5165":58,"\u5e8f\u5217\u7684\u5f00\u59cb":89,"\u5e8f\u5217\u7684\u6bcf\u4e2a\u5143\u7d20\u662f\u539f\u6765\u53cc\u5c42\u5e8f\u5217\u6bcf\u4e2asubseq\u5143\u7d20\u7684\u5e73\u5747\u503c":55,"\u5e8f\u5217\u7684\u7ed3\u5c3e":89,"\u5e8f\u5217\u7684\u7ed3\u675f":89,"\u5e8f\u5217\u8f93\u5165\u65f6\u7b49\u4e8e":46,"\u5e93":68,"\u5e93\u7684\u8def\u5f84":68,"\u5e94\u7528\u524d\u5411\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":59,"\u5e94\u7528\u53cd\u5411\u9012\u5f52\u795e\u7ecf\u7f51\u7edc":59,"\u5e94\u7528\u6a21\u578b":84,"\u5e94\u80fd\u53cd\u6620\u5f53\u524dcommit\u7684\u5185\u5bb9":61,"\u5e94\u8be5":56,"\u5e94\u8be5\u4e0e\u5b83\u7684memory\u540d\u5b57\u76f8\u540c":59,"\u5e94\u8be5\u8bf4\u8c22\u8c22":61,"\u5e94\u8be5\u8bfb\u53d6\u5f53\u524d\u76ee\u5f55\u4e0b\u7684":60,"\u5e94\u8be5\u964d\u4f4e\u5b66\u4e60\u7387":46,"\u5e95\u5c42\u8fdb\u7a0b":68,"\u5efa\u7acb\u4e00\u4e2a":61,"\u5efa\u8bae":[42,61],"\u5efa\u8bae\u5c06\u5176\u8bbe\u7f6e\u4e3a\u8f83\u5927":68,"\u5efa\u8bae\u5c06\u8be5\u53c2\u6570\u8bbe\u4e3atrue":70,"\u5f00\u53d1\u4e86\u6a21\u578b\u9884\u6d4b\u7684\u6837\u4f8b\u4ee3\u7801":34,"\u5f00\u53d1\u4eba\u5458\u4f7f\u7528":61,"\u5f00\u53d1\u4eba\u5458\u53ef\u4ee5\u5728docker\u5f00\u53d1\u955c\u50cf\u4e2d\u5f00\u53d1paddlepaddl":49,"\u5f00\u53d1\u8005\u4f7f\u7528":60,"\u5f00\u53d1\u8005\u4fee\u6539\u81ea\u5df1\u7684\u4ee3\u7801":42,"\u5f00\u53d1\u8005fork\u7684\u7248\u672c\u5e93\u4e2d":42,"\u5f00\u53d1\u8005fork\u7684\u7248\u672c\u5e93\u4f7f\u7528":42,"\u5f00\u53d1\u955c\u50cf":[49,61],"\u5f00\u53d1\u955c\u50cf\u5305\u542b\u4e86\u4ee5\u4e0b\u5de5\u5177":49,"\u5f00\u5934":63,"\u5f00\u59cb":47,"\u5f00\u59cb\u63d0\u4f9b\u670d\u52a1":20,"\u5f00\u59cb\u6807\u8bb0":59,"\u5f00\u59cb\u8bad\u7ec3\u6a21\u578b":84,"\u5f00\u59cb\u9636\u6bb5":67,"\u5f02\u6b65\u968f\u673a\u68af\u5ea6\u4e0b\u964d":69,"\u5f15\u5165\u4e86\u7c7b\u578b\u7684\u5934\u6587\u4ef6":34,"\u5f15\u5165lstm\u6a21\u578b\u4e3b\u8981\u662f\u4e3a\u4e86\u514b\u670d\u6d88\u5931\u68af\u5ea6\u7684\u95ee\u9898":88,"\u5f15\u5165paddlepaddle\u7684pydataprovider2\u5305":2,"\u5f15\u53d1":12,"\u5f15\u5bfc\u5c42":59,"\u5f15\u7528":68,"\u5f15\u7528memory\u5f97\u5230\u8fd9layer\u4e0a\u4e00\u65f6\u523b\u8f93\u51fa":58,"\u5f3a\u70c8\u63a8\u8350":56,"\u5f3a\u70c8\u63a8\u8350\u4f7f\u7528virtualenv\u6765\u521b\u9020\u4e00\u4e2a\u5e72\u51c0\u7684python\u73af\u5883":86,"\u5f52\u4e00\u5316\u6982\u7387\u5411\u91cf":59,"\u5f53":72,"\u5f53\u4f20\u9012\u76f8\u540c\u7684\u6837\u672c\u6570\u65f6":88,"\u5f53\u4f60\u6267\u884c\u547d\u4ee4":62,"\u5f53\u4fdd\u5b58\u7684\u7f51\u7edc\u53c2\u6570\u4e3afloat\u7c7b\u578b\u65f6\u4e3a4":46,"\u5f53\u51fd\u6570\u8fd4\u56de\u7684\u65f6\u5019":2,"\u5f53\u524d\u5355\u8bcd\u5728\u76f8\u6bd4\u4e4b\u4e0b\u603b\u662f\u88ab\u5f53\u4f5c\u771f\u503c":89,"\u5f53\u524d\u5355\u8bcd\u662f\u89e3\u7801\u5668\u6700\u540e\u4e00\u6b65\u7684\u8f93\u51fa":89,"\u5f53\u524d\u65f6\u95f4\u6b65\u5904\u7684memory\u7684\u8f93\u51fa\u4f5c\u4e3a\u4e0b\u4e00\u65f6\u95f4\u6b65memory\u7684\u8f93\u5165":59,"\u5f53\u524d\u7684\u5b66\u4e60\u7387\u4e3a\u6240\u8bbe\u7f6e":46,"\u5f53\u524d\u7684\u5b9e\u73b0\u65b9\u5f0f\u4e0b":62,"\u5f53\u524d\u7684\u8f93\u5165y\u548c\u4e0a\u4e00\u4e2a\u65f6\u95f4\u6b65\u7684\u8f93\u51farnn_state\u505a\u4e86\u4e00\u4e2a\u5168\u94fe\u63a5":56,"\u5f53\u524d\u8bc4\u4f30\u4e2d":89,"\u5f53\u524dbatch\u7684cost":89,"\u5f53\u524dlog_period\u4e2abatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747\u5206\u7c7b\u9519\u8bef\u7387":84,"\u5f53\u524dlog_period\u4e2abatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747cost":84,"\u5f53\u529f\u80fd\u5206\u652f\u5f00\u53d1\u5b8c\u6bd5\u540e":42,"\u5f53\u5728\u7f51\u7edc\u5c42\u914d\u7f6e\u4e2d\u8bbe\u7f6e":70,"\u5f53\u5728\u8bad\u7ec3\u914d\u7f6e\u4e2d\u8bbe\u7f6e":70,"\u5f53\u5bb9\u5668\u56e0\u4e3a\u5404\u79cd\u539f\u56e0\u88ab\u9500\u6bc1\u65f6":73,"\u5f53\u5df2\u8bad\u7ec3\u6837\u672c\u6570\u5927\u4e8e1000\u5c0f\u4e8e\u7b49\u4e8e2000\u65f6":46,"\u5f53\u5df2\u8bad\u7ec3\u6837\u672c\u6570\u5927\u4e8e2000\u65f6":46,"\u5f53\u5df2\u8bad\u7ec3\u6837\u672c\u6570\u5c0f\u4e8e\u7b49\u4e8e1000\u65f6":46,"\u5f53\u5df2\u8bad\u7ec3pass\u6570\u5927\u4e8e1\u5c0f\u4e8e\u7b49\u4e8e2\u65f6":46,"\u5f53\u5df2\u8bad\u7ec3pass\u6570\u5927\u4e8e2\u65f6":46,"\u5f53\u5df2\u8bad\u7ec3pass\u6570\u5c0f\u4e8e\u7b49\u4e8e1\u65f6":46,"\u5f53\u6240\u6709pod\u90fd\u5904\u4e8erunning\u72b6\u6001":75,"\u5f53\u6a21\u578b\u53c2\u6570\u4e0d\u5b58\u5728\u65f6":70,"\u5f53\u6a21\u578b\u8bad\u7ec3\u597d\u4e86\u4e4b\u540e":84,"\u5f53\u6a21\u5f0f\u4e3a":70,"\u5f53\u7136":67,"\u5f53\u7136\u53ef\u4ee5":60,"\u5f53\u7528\u6237\u4f7f\u7528\u5b8c\u8fd9\u4e2a\u53c2\u6570\u540e":34,"\u5f53\u7528\u6237\u6ca1\u6709\u663e\u5f0f\u8bbe\u5b9a\u65f6":46,"\u5f53\u7f51\u7edc\u5c42\u7528\u4e00\u4e2a\u6279\u6b21\u505a\u8bad\u7ec3\u65f6":62,"\u5f53\u89e3\u8bfb\u6bcf\u4e00\u4e2a":59,"\u5f53\u8bad\u7ec3\u6570\u636e\u975e\u5e38\u591a\u65f6":2,"\u5f53\u8d85\u8fc7\u8be5\u9608\u503c\u65f6":70,"\u5f53\u8f93\u5165\u662f\u7ef4\u5ea6\u5f88\u9ad8\u7684\u7a00\u758f\u6570\u636e\u65f6":72,"\u5f53\u9700\u8981\u5b8c\u6210\u8ba1\u7b97\u65f6":64,"\u5f53\u9700\u8981\u5feb\u901f\u6216\u8005\u9891\u7e41\u7684\u8bc4\u4f30\u65f6":89,"\u5f53classif":89,"\u5f53destination\u6587\u4ef6\u4e0d\u5b58\u5728\u6216\u8005\u5927\u5c0f\u548csource\u6587\u4ef6\u4e0d\u4e00\u81f4\u65f6":28,"\u5f53n1":46,"\u5f62\u6210recurr":58,"\u5f62\u6210recurrent\u8fde\u63a5":58,"\u5f62\u72b6":82,"\u5f88":[56,84],"\u5f88\u591a":[56,60],"\u5f88\u591a\u5f00\u53d1\u8005\u4f1a\u4f7f\u7528\u8fdc\u7a0b\u7684\u5b89\u88c5\u6709gpu\u7684\u670d\u52a1\u5668\u5de5\u4f5c":49,"\u5f88\u5b89\u9759":56,"\u5f88\u5e72\u51c0":56,"\u5f88\u65b9\u4fbf":56,"\u5f88\u6709\u53ef\u80fd\u5b9e\u9645\u5e94\u7528\u5c31\u662f\u6ca1\u6709\u6309\u7167\u60a8\u7684\u9884\u671f\u60c5\u51b5\u8fd0\u884c":67,"\u5f88\u6709\u53ef\u80fd\u662f\u975e\u72ec\u5360\u65b9\u5f0f\u6267\u884c\u5bfc\u81f4\u7684\u7aef\u53e3\u51b2\u7a81":46,"\u5f88\u9002\u5408\u6784\u5efa\u7528\u4e8e\u7406\u89e3\u56fe\u7247\u5185\u5bb9\u7684\u6a21\u578b":81,"\u5f88\u96be\u4fdd\u8bc1\u591a\u8bed\u8a00\u4ee3\u7801\u98ce\u683c\u7684\u4e00\u81f4\u6027":33,"\u5f88\u96be\u6574\u4f53\u4fee\u6b63":2,"\u5f8b\u5e08":85,"\u5f97":56,"\u5f97\u4f7f\u7528":33,"\u5f97\u5230\u53e5\u5b50\u7684\u8868\u793a":84,"\u5f97\u5230\u6700\u597d\u8f6e\u6b21\u4e0b\u7684\u6a21\u578b":86,"\u5f97\u5230\u8f93\u51fa\u503c":63,"\u5faa\u73af\u5c55\u5f00\u7684\u6bcf\u4e2a\u65f6\u95f4\u6b65\u603b\u662f\u80fd\u591f\u5f15\u7528\u6240\u6709\u8f93\u5165":58,"\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u4e2d":59,"\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u4f5c\u4e3a\u4f7f\u7528":59,"\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u548c":59,"\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u5728\u6bcf\u4e2a\u65f6\u95f4\u6b65\u9aa4\u987a\u5e8f\u5730\u5904\u7406\u5e8f\u5217":59,"\u5faa\u73af\u7f51\u7edc\u4ece":59,"\u5fc5\u8981":34,"\u5fc5\u987b":62,"\u5fc5\u987b\u4e00\u81f4":2,"\u5fc5\u987b\u4f7f\u7528python\u5173\u952e\u8bcd":2,"\u5fc5\u987b\u5c06\u524d\u4e00\u4e2a\u5b50\u53e5\u7684\u6700\u540e\u4e00\u4e2a\u5143\u7d20":56,"\u5fc5\u987b\u6307\u5411\u4e00\u4e2apaddlepaddle\u5b9a\u4e49\u7684lay":58,"\u5fc5\u987b\u662f\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":58,"\u5fc5\u987b\u662f\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":55,"\u5fc5\u987b\u7531\u53ea\u8bfbmemory\u7684":59,"\u5fc5\u987b\u8bbe\u7f6e\u4e3a":53,"\u5fc5\u987b\u914d\u7f6e\u4e3a":54,"\u5fd8\u8bb0\u95e8\u548c\u8f93\u51fa\u95e8":88,"\u5feb":[56,88],"\u5feb\u901f\u5165\u95e8":83,"\u5feb\u901f\u5728\u672c\u5730\u542f\u52a8\u4e00\u4e2a\u5355\u673a\u7684kubernetes\u670d\u52a1\u5668":73,"\u5feb\u901f\u90e8\u7f72\u96c6\u7fa4":73,"\u6027\u4ef7\u6bd4":56,"\u6027\u522b":[85,86],"\u6027\u80fd\u5206\u6790":67,"\u6027\u80fd\u5206\u6790\u5de5\u5177\u662f\u7528\u4e8e\u7ed9\u5e94\u7528\u7a0b\u5e8f\u7684\u6027\u80fd\u505a\u5b9a\u91cf\u5206\u6790\u7684":67,"\u6027\u80fd\u5206\u6790\u662f\u6027\u80fd\u4f18\u5316\u7684\u5173\u952e\u4e00\u6b65":67,"\u6027\u80fd\u548c\u628a\u7f16\u8bd1\u5de5\u5177\u5b89\u88c5\u5728\u672c\u673a\u8fd0\u884c\u4e00\u6837":60,"\u6027\u80fd\u8c03\u4f18":69,"\u603b\u4f53\u6765\u8bf4":56,"\u603b\u8ba1\u7684\u53c2\u6570\u4e2a\u6570":80,"\u603b\u8bc4\u520610\u5206":88,"\u6050\u6016\u7247":85,"\u60a8\u4f1a\u5728\u63a5\u4e0b\u6765\u7684\u90e8\u5206\u4e2d\u83b7\u5f97\u66f4\u591a\u7684\u7ec6\u8282\u4ecb\u7ecd":67,"\u60a8\u53ef\u4ee5\u4efb\u610f\u4f7f\u7528\u4e00\u4e2a\u6216\u4e24\u4e2a\u6765\u5bf9\u611f\u5174\u8da3\u7684\u4ee3\u7801\u6bb5\u505a\u6027\u80fd\u5206\u6790":67,"\u60a8\u53ef\u4ee5\u5bfc\u5165":67,"\u60a8\u53ef\u4ee5\u91c7\u7528\u4e0b\u9762\u4e94\u4e2a\u6b65\u9aa4":67,"\u60a8\u5c06\u4e86\u89e3\u5982\u4f55":59,"\u60a8\u5c31\u80fd\u83b7\u5f97\u5982\u4e0b\u7684\u5206\u6790\u7ed3\u679c":67,"\u60a8\u6309\u5982\u4e0b\u6b65\u9aa4\u64cd\u4f5c\u5373\u53ef":67,"\u60a8\u6700\u597d\u5148\u786e\u8ba4":67,"\u60a8\u9700\u8981\u66f4\u6539":49,"\u60a8\u9996\u5148\u9700\u8981\u5728\u76f8\u5173\u4ee3\u7801\u6bb5\u4e2d\u52a0\u5165":67,"\u60ac\u7591\u7247":85,"\u60c5\u6001\u52a8\u8bcd":87,"\u60c5\u611f\u5206\u6790":[42,83],"\u60c5\u611f\u5206\u6790\u4e5f\u5e38\u7528\u4e8e\u57fa\u4e8e\u5927\u91cf\u8bc4\u8bba\u548c\u4e2a\u4eba\u535a\u5ba2\u6765\u76d1\u63a7\u793e\u4f1a\u5a92\u4f53":88,"\u60c5\u611f\u5206\u6790\u662f\u81ea\u7136\u8bed\u8a00\u7406\u89e3\u4e2d\u6700\u5178\u578b\u7684\u95ee\u9898\u4e4b\u4e00":88,"\u60c5\u611f\u5206\u6790\u6709\u8bb8\u591a\u5e94\u7528\u573a\u666f":88,"\u60ca\u9669\u7535\u5f71":85,"\u60f3\u4e86\u89e3\u66f4\u591a\u7ec6\u8282\u53ef\u4ee5\u53c2\u8003pydataprovider\u90e8\u5206\u7684\u6587\u6863":88,"\u610f\u5473\u7740\u4e0d\u540c\u65f6\u95f4\u6b65\u7684\u8f93\u5165\u90fd\u662f\u76f8\u540c\u7684\u503c":59,"\u610f\u601d\u662f\u4e0d\u4f7f\u7528\u5e73\u5747\u53c2\u6570\u6267\u884c\u6d4b\u8bd5":70,"\u610f\u601d\u662f\u4e0d\u4fdd\u5b58\u7ed3\u679c":70,"\u610f\u601d\u662f\u4f7f\u7528\u7b2ctest":70,"\u610f\u601d\u662f\u5728gpu\u6a21\u5f0f\u4e0b\u4f7f\u75284\u4e2agpu":70,"\u611f\u89c9":56,"\u620f\u5267":85,"\u6210\u529f\u8bad\u7ec3\u4e14\u9000\u51fa\u7684pod\u6570\u76ee\u4e3a3\u65f6":75,"\u6210\u5458":63,"\u6211\u4eec\u4e0d\u80fd\u901a\u8fc7\u5e38\u89c4\u7684\u68af\u5ea6\u68c0\u67e5\u7684\u65b9\u5f0f\u6765\u8ba1\u7b97\u68af\u5ea6":62,"\u6211\u4eec\u4e3b\u8981\u4f1a\u4ecb\u7ecdnvprof\u548cnvvp":67,"\u6211\u4eec\u4e5f\u53ef\u4ee5\u786e\u5b9a\u6bcf\u4e00\u4e2a\u53c2\u6570\u7684\u7c7b\u578b":34,"\u6211\u4eec\u4ec5\u4ec5\u5bf9\u795e\u7ecf\u7f51\u7edc\u7684\u8f93\u5165\u8fdb\u884c\u4e86\u63cf\u8ff0":51,"\u6211\u4eec\u4ec5\u4ec5\u662f\u5c06\u6bcf\u4e2a\u7279\u5f81\u79cd\u7c7b\u6620\u5c04\u5230\u4e00\u4e2a\u7279\u5f81\u5411\u91cf\u4e2d":86,"\u6211\u4eec\u4ec5\u6709\u4e00\u4e2a\u8f93\u5165":62,"\u6211\u4eec\u4ec5\u7528":86,"\u6211\u4eec\u4ecb\u7ecd\u5982\u4f55\u5728":74,"\u6211\u4eec\u4ecb\u7ecd\u5982\u4f55\u5728kubernetes\u96c6\u7fa4\u4e0a\u8fdb\u884c\u5206\u5e03\u5f0fpaddlepaddle\u8bad\u7ec3\u4f5c\u4e1a":75,"\u6211\u4eec\u4ece\u63d0\u524d\u7ed9\u5b9a\u7684\u7c7b\u522b\u96c6\u5408\u4e2d\u9009\u62e9\u5176\u6240\u5c5e\u7c7b\u522b":84,"\u6211\u4eec\u4ee5mnist\u624b\u5199\u8bc6\u522b\u4e3a\u4f8b":2,"\u6211\u4eec\u4f1a\u53d1\u73b0\u6570\u636e\u96c6":89,"\u6211\u4eec\u4f1a\u5728":49,"\u6211\u4eec\u4f1a\u7ee7\u7eed\u4f7f\u7528\u73b0\u6709\u7684\u5185\u5b58\u5757":62,"\u6211\u4eec\u4f1a\u91cd\u65b0\u5206\u914d\u5185\u5b58":62,"\u6211\u4eec\u4f7f\u7528":[62,68,88],"\u6211\u4eec\u4f7f\u7528\u4e0d\u540c\u7684layer\u8fdb\u884c\u7ec4\u5408":51,"\u6211\u4eec\u4f7f\u7528\u4e86":56,"\u6211\u4eec\u4f7f\u7528\u4e86\u4e00\u4e2a\u7f16\u89e3\u7801\u6a21\u578b\u7684\u6269\u5c55":89,"\u6211\u4eec\u4f7f\u7528\u4e86\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":88,"\u6211\u4eec\u4f7f\u7528\u5176\u4e2d\u7684":89,"\u6211\u4eec\u4f7f\u7528\u52a8\u6001\u5e93\u6765\u5206\u53d1paddl":33,"\u6211\u4eec\u4f7f\u7528\u6700\u5927\u6982\u7387\u7684\u6807\u7b7e\u4f5c\u4e3a\u7ed3\u679c":87,"\u6211\u4eec\u4f7f\u7528\u96c6\u675f\u641c\u7d22":89,"\u6211\u4eec\u4f7f\u7528paddlepaddle\u5728ilsvrc\u7684\u9a8c\u8bc1\u96c6\u517150":82,"\u6211\u4eec\u5047\u8bbe\u4e00\u53f0\u673a\u5668\u4e0a\u67094\u4e2agpu":72,"\u6211\u4eec\u5047\u8bbe\u623f\u4ea7\u7684\u4ef7\u683c":47,"\u6211\u4eec\u5148\u4ece\u4e00\u6761\u968f\u673a\u7684\u76f4\u7ebf":47,"\u6211\u4eec\u5148\u8c03\u7528\u6bcf\u4e2a":64,"\u6211\u4eec\u51c6\u5907\u4e86\u4e00\u4e2a\u811a\u672c":81,"\u6211\u4eec\u5373\u53ef\u5b8c\u6210\u795e\u7ecf\u7f51\u7edc\u7684\u642d\u5efa":51,"\u6211\u4eec\u53ea\u4f7f\u7528\u5df2\u7ecf\u6807\u6ce8\u8fc7\u7684\u8bad\u7ec3\u96c6\u548c\u6d4b\u8bd5\u96c6":88,"\u6211\u4eec\u53ea\u6240\u4ee5\u4f7f\u7528lstm\u6765\u6267\u884c\u8fd9\u4e2a\u4efb\u52a1\u662f\u56e0\u4e3a\u5176\u6539\u8fdb\u7684\u8bbe\u8ba1\u5e76\u4e14\u5177\u6709\u95e8\u673a\u5236":88,"\u6211\u4eec\u53ea\u6f14\u793a\u4e00\u4e2a\u5355\u673a\u4f5c\u4e1a":74,"\u6211\u4eec\u53ea\u9700\u8981":60,"\u6211\u4eec\u53ea\u9700\u8981\u4f7f\u7528lstm":56,"\u6211\u4eec\u53ea\u9700\u8981\u8fd0\u884c":84,"\u6211\u4eec\u53ea\u9700\u8981\u8fd0\u884c\u4e0b\u9762\u547d\u4ee4\u628a\u7f16\u8bd1\u597d\u7684paddlepaddle\u6253\u5305\u6210\u4e00\u4e2a":61,"\u6211\u4eec\u53ea\u9700\u8981\u914d\u7f6e":60,"\u6211\u4eec\u53ef\u4ee5":60,"\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528":81,"\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528\u5176\u4ed6layer\u8fdb\u884c\u7ec4\u5408":51,"\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528\u5b83\u6765\u751f\u6210\u5e8f\u5217":59,"\u6211\u4eec\u53ef\u4ee5\u521b\u5efatrainer\u6765\u5bf9\u7f51\u7edc\u8fdb\u884c\u8bad\u7ec3":51,"\u6211\u4eec\u53ef\u4ee5\u53c2\u8003tensorflow\u7684":64,"\u6211\u4eec\u53ef\u4ee5\u5728":61,"\u6211\u4eec\u53ef\u4ee5\u5b9a\u4e49\u5982\u4e0b\u7684layer\u7ec4\u5408":51,"\u6211\u4eec\u53ef\u4ee5\u5b9a\u4e49\u5982\u4e0blayer\u6765\u63cf\u8ff0\u795e\u7ecf\u7f51\u7edc\u7684\u8f93\u5165":51,"\u6211\u4eec\u53ef\u4ee5\u5c06":68,"\u6211\u4eec\u53ef\u4ee5\u6309\u7167\u5982\u4e0b\u5c42\u6b21\u5b9a\u4e49\u975e\u5e8f\u5217":55,"\u6211\u4eec\u53ef\u4ee5\u751f\u6210":86,"\u6211\u4eec\u53ef\u4ee5\u76f4\u63a5\u89c2\u5bdf\u6a21\u578b\u7684\u53c2\u6570\u662f\u5426\u7b26\u5408\u9884\u671f\u6765\u8fdb\u884c\u68c0\u9a8c":47,"\u6211\u4eec\u53ef\u4ee5\u770b\u5230\u5728\u76ee\u5f55":88,"\u6211\u4eec\u53ef\u4ee5\u8bbe\u8ba1\u642d\u5efa\u4e00\u4e2a\u7075\u6d3b\u7684":58,"\u6211\u4eec\u53ef\u4ee5\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4\u6765\u505ableu\u8bc4\u4f30":89,"\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u4ee5\u4e0b\u547d\u4ee4\u6765\u8bad\u7ec3\u6a21\u578b":89,"\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u4ee5\u4e0b\u547d\u4ee4\u6765\u8fdb\u884c\u4ece\u6cd5\u8bed\u5230\u82f1\u8bed\u7684\u6587\u672c\u7ffb\u8bd1":89,"\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u5982\u4e0b\u547d\u4ee4\u8fdb\u884c\u9884\u5904\u7406\u5de5\u4f5c":81,"\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u65e5\u5fd7\u67e5\u770b\u5bb9\u5668\u8bad\u7ec3\u7684\u60c5\u51b5":75,"\u6211\u4eec\u5728":64,"\u6211\u4eec\u57285\u5929\u91cc\u8bad\u7ec3\u4e8616\u4e2apass":89,"\u6211\u4eec\u5728\u51fd\u6570\u7684\u7ed3\u5c3e\u8fd4\u56de":59,"\u6211\u4eec\u5728\u62e5\u670950\u4e2a\u8282\u70b9\u7684\u96c6\u7fa4\u4e2d\u8bad\u7ec3\u6a21\u578b":89,"\u6211\u4eec\u5728\u8bad\u7ec3\u4e4b\u524d\u9700\u8981\u5e38\u89c1\u4e00\u4e2a\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":89,"\u6211\u4eec\u5728initialzier\u51fd\u6570\u91cc\u521d\u59cb\u5316\u8bcd\u8868":84,"\u6211\u4eec\u5bf9\u6a21\u578b\u8fdb\u884c\u4e86\u4ee5\u4e0b\u66f4\u6539":59,"\u6211\u4eec\u5c06":[75,86],"\u6211\u4eec\u5c06\u4e00\u6bb5\u8bdd\u770b\u6210\u53e5\u5b50\u7684\u6570\u7ec4":56,"\u6211\u4eec\u5c06\u4ecb\u7ecd\u5982\u4f55\u542f\u52a8\u5206\u5e03\u5f0f\u8bad\u7ec3\u4f5c\u4e1a":74,"\u6211\u4eec\u5c06\u4ee5":[68,84],"\u6211\u4eec\u5c06\u4ee5\u6700\u57fa\u672c\u7684\u903b\u8f91\u56de\u5f52\u7f51\u7edc\u4f5c\u4e3a\u8d77\u70b9":84,"\u6211\u4eec\u5c06\u4f7f\u7528":59,"\u6211\u4eec\u5c06\u4f7f\u7528\u7b80\u5355\u7684":59,"\u6211\u4eec\u5c06\u4f7f\u7528cifar":81,"\u6211\u4eec\u5c06\u539f\u59cb\u6570\u636e\u7684\u6bcf\u4e00\u7ec4":56,"\u6211\u4eec\u5c06\u5728\u540e\u9762\u4ecb\u7ecd\u8bad\u7ec3\u548c\u9884\u6d4b\u6d41\u7a0b\u7684\u811a\u672c":84,"\u6211\u4eec\u5c06\u5b83\u4eec\u5212\u5206\u4e3a\u4e0d\u540c\u7684\u7c7b\u522b":69,"\u6211\u4eec\u5c06\u5bf9\u8fd9\u4e24\u4e2a\u6b65\u9aa4\u7ed9\u51fa\u4e86\u8be6\u7ec6\u7684\u89e3\u91ca":84,"\u6211\u4eec\u5c06\u653e\u7f6e\u4f9d\u8d56\u5e93":68,"\u6211\u4eec\u5c06\u8bc4\u5206\u5206\u6210\u4e24\u90e8\u5206":86,"\u6211\u4eec\u5c06\u9610\u91ca\u5982\u4f55\u5728\u96c6\u7fa4\u4e0a\u8fd0\u884c\u5206\u5e03\u5f0f":68,"\u6211\u4eec\u5c31\u53ef\u4ee5\u7740\u624b\u5bf9\u5206\u7c7b\u6a21\u578b\u8fdb\u884c\u8bad\u7ec3\u4e86":81,"\u6211\u4eec\u5c31\u53ef\u4ee5\u8bad\u7ec3\u6a21\u578b\u4e86":84,"\u6211\u4eec\u5c31\u53ef\u4ee5\u8fdb\u884c\u9884\u6d4b\u4e86":84,"\u6211\u4eec\u5c31\u5b8c\u6210\u4e86\u4e00\u6b21\u4ee3\u7801\u8d21\u732e\u7684\u8fc7\u7a0b":61,"\u6211\u4eec\u5c55\u793a\u5982\u4f55\u7528paddlepaddle\u89e3\u51b3":47,"\u6211\u4eec\u5df2\u7ecf\u5b9e\u73b0\u4e86\u5927\u591a\u6570\u5e38\u7528\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u67b6\u6784":59,"\u6211\u4eec\u5e0c\u671b\u80fd\u591f\u68c0\u9a8c\u6a21\u578b\u7684\u597d\u574f":47,"\u6211\u4eec\u5e94\u5f53\u4f1a\u5f97\u5230\u4e00\u4e2a\u540d\u4e3acifar":81,"\u6211\u4eec\u5efa\u8bae\u4f60\u4e3a\u4f60\u7684python\u5c01\u88c5\u5b9e\u73b0\u4e00\u4e2a":62,"\u6211\u4eec\u5efa\u8bae\u4f60\u5728\u5199\u65b0\u7f51\u7edc\u5c42\u65f6\u628a\u6d4b\u8bd5\u4ee3\u7801\u653e\u5165\u65b0\u7684\u6587\u4ef6\u4e2d":62,"\u6211\u4eec\u5efa\u8bae\u4f7f\u7528\u7b2c\u4e8c\u7c7b\u5b9e\u73b0":46,"\u6211\u4eec\u603b\u7ed3\u4e86\u5404\u4e2a\u7f51\u7edc\u7684\u590d\u6742\u5ea6\u548c\u6548\u679c":84,"\u6211\u4eec\u611f\u8c22":89,"\u6211\u4eec\u628apaddlepaddle\u7684\u7f16\u8bd1\u73af\u5883\u6253\u5305\u6210\u4e00\u4e2a\u955c\u50cf":49,"\u6211\u4eec\u63a8\u8350\u4f7f\u7528":49,"\u6211\u4eec\u63a8\u8350\u4f7f\u7528\u6700\u65b0\u7248\u672c\u7684cudnn":48,"\u6211\u4eec\u63d0\u4f9b\u4e24\u4e2a\u7f51\u7edc":88,"\u6211\u4eec\u63d0\u4f9b\u4e24\u4e2a\u8f6c\u6362\u65b9\u5f0f":21,"\u6211\u4eec\u63d0\u4f9b\u4e86\u4e00\u4e2a\u6570\u636e\u9884\u5904\u7406\u811a\u672c":88,"\u6211\u4eec\u63d0\u4f9b\u4e86\u4e00\u4e2a\u793a\u4f8b\u811a\u672c":82,"\u6211\u4eec\u63d0\u4f9b\u4e86\u56fd\u5185\u7684\u955c\u50cf\u670d\u52a1\u5668\u4f9b\u5927\u5bb6\u4f7f\u7528":49,"\u6211\u4eec\u63d0\u4f9b\u4e86\u811a\u672c\u6765\u6784\u5efa\u5b57\u5178\u548c\u9884\u5904\u7406\u6570\u6910":88,"\u6211\u4eec\u63d0\u4f9b\u4e86c":82,"\u6211\u4eec\u63d0\u4f9b\u53ef\u4ee5\u76f4\u63a5\u8fd0\u884cpaddlepaddl":49,"\u6211\u4eec\u63d0\u51fa\u4e86chunk\u7684\u6982\u5ff5":28,"\u6211\u4eec\u65e0\u6cd5\u5728docker\u5bb9\u5668\u4e2d":49,"\u6211\u4eec\u662f\u5bf9\u6bcf\u4e00\u4e2a\u5b50\u5e8f\u5217\u53d6\u6700\u540e\u4e00\u4e2a\u5143\u7d20":56,"\u6211\u4eec\u6700\u7ec8\u7684\u52a8\u6001\u5e93\u4e2d\u4e0d\u5d4c\u5165python\u6216\u8005\u5176\u4ed6\u4efb\u4f55\u8bed\u8a00\u7684\u89e3\u91ca\u5668":33,"\u6211\u4eec\u6709\u4e00\u4e2a\u5e8f\u5217\u4f5c\u4e3a\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u72b6\u6001":59,"\u6211\u4eec\u7528":85,"\u6211\u4eec\u7528\u4ee5\u4e0b\u7684\u4e00\u4e9b":86,"\u6211\u4eec\u7528\u7f16\u53f7\u4f5c\u4e3akei":86,"\u6211\u4eec\u7528paddlepaddle\u89e3\u51b3\u4e86\u5355\u53d8\u91cf\u7ebf\u6027\u56de\u5f52\u95ee\u9898":47,"\u6211\u4eec\u7684":60,"\u6211\u4eec\u7684\u5b57\u5178\u4f7f\u7528\u5185\u90e8\u7684\u5206\u8bcd\u5de5\u5177\u5bf9\u767e\u5ea6\u77e5\u9053\u548c\u767e\u5ea6\u767e\u79d1\u7684\u8bed\u6599\u8fdb\u884c\u5206\u8bcd\u540e\u4ea7\u751f":80,"\u6211\u4eec\u7684\u6807\u51c6\u5f00\u53d1\u6d41\u7a0b\u662f\u628a\u8fd9\u4e9b\u5de5\u5177\u90fd\u88c5\u8fdb\u4e00\u4e2adocker":61,"\u6211\u4eec\u7684\u8bad\u7ec3\u66f2\u7ebf\u5982\u4e0b":87,"\u6211\u4eec\u770b\u4e00\u4e0b\u5355\u5c42rnn\u7684\u914d\u7f6e":56,"\u6211\u4eec\u770b\u4e00\u4e0b\u8bed\u4e49\u76f8\u540c\u7684\u53cc\u5c42rnn\u7684\u7f51\u7edc\u914d\u7f6e":56,"\u6211\u4eec\u771f\u8bda\u5730\u611f\u8c22\u60a8\u7684\u5173\u6ce8":88,"\u6211\u4eec\u771f\u8bda\u5730\u611f\u8c22\u60a8\u7684\u8d21\u732e":61,"\u6211\u4eec\u79f0\u4e4b\u4e3a\u4e00\u4e2a0\u5c42\u7684\u5e8f\u5217":55,"\u6211\u4eec\u8bbe\u8ba1\u8bf4\u660e\u4e86\u540d\u4e3afilemanager\u7cfb\u7edf":28,"\u6211\u4eec\u8c03\u7528\u4e86eigenvector\u7684flatten\u63a5\u53e3":64,"\u6211\u4eec\u8fd8\u53ef\u4ee5\u767b\u5f55\u5230\u5bbf\u4e3b\u673a\u4e0a\u67e5\u770b\u8bad\u7ec3\u7ed3\u679c":74,"\u6211\u4eec\u8fd8\u5c06\u7f16\u7801\u5411\u91cf\u6295\u5c04\u5230":59,"\u6211\u4eec\u9009\u53d6\u5355\u53cc\u5c42\u5e8f\u5217\u914d\u7f6e\u4e2d\u7684\u4e0d\u540c\u90e8\u5206":56,"\u6211\u4eec\u9009\u62e9":21,"\u6211\u4eec\u901a\u5e38\u501f\u52a9":63,"\u6211\u4eec\u901a\u5e38\u5728\u6240\u6709\u8282\u70b9\u4e0a\u521b\u5efa\u4e00\u4e2a":68,"\u6211\u4eec\u901a\u5e38\u5c06\u4e00\u53e5\u8bdd\u7406\u89e3\u6210\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217":56,"\u6211\u4eec\u901a\u8fc7\u8bfb\u53d6":75,"\u6211\u4eec\u9075\u5faa":89,"\u6211\u4eec\u90fd\u63d0\u4f9bpython\u7684\u8f6c\u6362\u5e93":21,"\u6211\u4eec\u91c7\u7528\u4e0a\u9762\u7684\u7279\u5f81\u6a21\u677f":87,"\u6211\u4eec\u91c7\u7528\u5355\u5c42lstm\u6a21\u578b":84,"\u6211\u4eec\u91c7\u7528\u7b2c\u4e8c\u4e2a\u65b9\u6cd5":49,"\u6211\u4eec\u9700\u8981":60,"\u6211\u4eec\u9700\u8981\u5148\u521b\u5efa\u4e00\u4e2a\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":81,"\u6211\u4eec\u9700\u8981\u5148\u628a\u8f93\u5165tensor\u548c\u8f93\u51fatensor\u8f6c\u6362\u4e3aeigen\u652f\u6301\u7684\u683c\u5f0f":64,"\u6211\u4eec\u9700\u8981\u521b\u5efa\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":89,"\u6211\u4eec\u9700\u8981\u5236\u4f5c\u4e00\u4e2a\u5305\u542b\u8bad\u7ec3\u6570\u636e\u7684paddle\u955c\u50cf":74,"\u6211\u4eec\u9700\u8981\u5728\u96c6\u7fa4\u7684\u6240\u6709\u8282\u70b9\u4e0a\u5b89\u88c5":68,"\u6211\u4eec\u9700\u8981\u7b49\u5f0f\u5de6\u8fb9\u7684eigentensor\u8c03\u7528device\u63a5\u53e3":64,"\u6211\u4eec\u9700\u8981\u8ba1\u7b97":62,"\u6211\u4eec\u9700\u8981\u8bbe\u7f6e":86,"\u6211\u4eec\u9700\u8981\u9884\u5904\u7406\u6570\u6910\u5e76\u6784\u5efa\u4e00\u4e2a\u5b57\u5178":88,"\u6211\u4eec\u975e\u5e38\u6b22\u8fce\u60a8\u7528paddlepaddle\u6784\u5efa\u66f4\u597d\u7684\u793a\u4f8b":86,"\u6211\u4eec\u9884\u8bad\u7ec3\u5f97\u52304\u79cd\u4e0d\u540c\u7ef4\u5ea6\u7684\u8bcd\u5411\u91cf":80,"\u6211\u4eec\u9996\u5148\u5904\u7406\u7535\u5f71\u6216\u7528\u6237\u7684\u7279\u5f81\u6587\u4ef6":86,"\u6211\u4eec\u9996\u5148\u9700\u8981\u6839\u636e\u795e\u7ecf\u7f51\u7edc\u7ed3\u6784\u6765\u521b\u5efa\u6240\u9700\u8981\u4f18\u5316\u7684paramet":51,"\u6211\u4eec\u9ed8\u8ba4\u4f7f\u7528imdb\u7684\u6d4b\u8bd5\u6570\u636e\u96c6\u4f5c\u4e3a\u9a8c\u8bc1":88,"\u6211\u5220\u9664\u4e86":61,"\u6211\u53ef\u4ee5\u7528":60,"\u6211\u53ef\u4ee5\u9009\u62e9\u4e0d\u7528docker\u5417":60,"\u6216":[2,67,81,87],"\u6216\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":55,"\u6216\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u53d8\u6210\u4e00\u4e2a0\u5c42\u5e8f\u5217":55,"\u6216\u4e00\u4e2a\u5411\u91cf":58,"\u6216\u4e0d\u786e\u5b9a":85,"\u6216\u5355\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u53d8\u6210\u4e00\u4e2a0\u5c42\u5e8f\u5217":55,"\u6216\u662f\u624b\u52a8\u7f16\u8f91\u751f\u6210":86,"\u6216\u6700\u5927\u503c":55,"\u6216\u6d4b\u8bd5\u6587\u4ef6\u5217\u8868":1,"\u6216\u7b2c\u4e00\u4e2a":55,"\u6216\u7b2c\u4e00\u4e2a\u5143\u7d20":55,"\u6216\u8005":[33,34,46,49,55,56,60,63,67],"\u6216\u8005\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":58,"\u6216\u8005\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":[55,58],"\u6216\u8005\u4ece\u5de5\u5177\u7684\u754c\u9762\u91cc\u8fd0\u884c\u60a8\u7684\u5e94\u7528":67,"\u6216\u8005\u53cd\u5411\u5730\u4ece":59,"\u6216\u8005\u5728cpu\u6a21\u5f0f\u4e0b\u4f7f\u75284\u4e2a\u7ebf\u7a0b":70,"\u6216\u8005\u5c06\u8fd9\u53f0\u8282\u70b9\u8fc1\u79fb\u5230\u53e6\u4e00\u4e2a\u8282\u70b9\u5e76\u542f\u52a8\u5373\u53ef\u6062\u590d\u8bad\u7ec3\u4efb\u52a1":20,"\u6216\u8005\u5df2\u7ecf\u5728\u96c6\u7fa4\u63d0\u4ea4\u73af\u5883\u4e2d\u81ea\u52a8\u8bbe\u7f6e":69,"\u6216\u8005\u6570\u636e\u5e93\u8fde\u63a5\u8def\u5f84\u7b49":1,"\u6216\u8005\u6570\u7ec4\u7684\u6570\u7ec4\u8fd9\u4e2a\u6982\u5ff5":56,"\u6216\u8005\u662f\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":55,"\u6216\u8005\u662f\u4e00\u4e2a\u5c0f\u7684\u6587\u672c\u7247\u6bb5":88,"\u6216\u8005\u662f\u51fd\u6570\u8c03\u7528\u7684\u9891\u7387\u548c\u8017\u65f6\u7b49":67,"\u6216\u8005\u66f4\u65e9":46,"\u6216\u8005\u6bcf\u4e00\u4e2a\u7cfb\u5217\u91cc\u7684\u7279\u5f81\u6570\u636e":56,"\u6216\u8005\u7528tuple\u8868\u793a\u7684\u591a\u4e2a\u503c":21,"\u6216\u8005\u7531\u5b83\u4eec\u7ec4\u6210\u7684list":21,"\u6216\u8005\u76f4\u63a5\u4f7f\u7528\u4e0b\u9762\u7684shell\u547d\u4ee4":82,"\u6216\u8005\u76f4\u63a5\u6254\u6389\u975e\u5e38\u957f\u7684\u5e8f\u5217":46,"\u6216\u8005\u76f8\u5bf9\u4e8e\u6784\u5efa\u76ee\u5f55\u7684\u76f8\u5bf9\u8def\u5f84":[53,54],"\u6216\u8005\u8f93\u5165\u6570\u636e\u5c3a\u5ea6\u8fc7\u5927":46,"\u6216\u8005\u8fd0\u884c":46,"\u6216\u8005\u91c7\u7528\u5e76\u884c\u8ba1\u7b97\u6765\u52a0\u901f\u67d0\u4e9b\u5c42\u7684\u66f4\u65b0":72,"\u6216\u8005\u9700\u8981\u53d1\u5e03\u60a8\u7684\u5e94\u7528\u7684\u955c\u50cf":49,"\u6216\u8005\u9700\u8981\u66f4\u9ad8\u7684\u6548\u7387":1,"\u6216\u8bbe\u7f6e\u4e3anone":1,"\u6216gpu":70,"\u6216gpu\u4e2a\u6570":88,"\u6218\u4e89\u7247":85,"\u622a\u65ad\u5bf9\u8c61\u4e0d\u540c":46,"\u623f":56,"\u623f\u95f4":56,"\u6240\u4ee5":[2,46],"\u6240\u4ee5\u4e00\u822c\u9700\u8981\u5bf9\u8bad\u7ec3\u7528\u7684\u6a21\u578b\u914d\u7f6e\u6587\u4ef6\u7a0d\u4f5c\u76f8\u5e94\u4fee\u6539\u624d\u80fd\u5728\u9884\u6d4b\u65f6\u4f7f\u7528":4,"\u6240\u4ee5\u4e0d\u80fd\u91c7\u7528\u7b2c\u4e00\u79cd\u65b9\u5f0f\u5728\u8fd9\u51e0\u4e2alayer\u91cc\u8bbe\u7f6e":46,"\u6240\u4ee5\u4f60\u53ea\u7528\u6309\u4e0b\u9762\u7684\u7ed3\u6784\u6765\u7ec4\u7ec7\u6570\u6910\u5c31\u884c\u4e86":88,"\u6240\u4ee5\u505a\u6cd5\u53ef\u4ee5\u6709\u4e24\u79cd":46,"\u6240\u4ee5\u53ef\u4ee5\u5229\u7528\u5982\u4e0b\u65b9\u6cd5\u8bfb\u53d6\u6a21\u578b\u7684\u53c2\u6570":47,"\u6240\u4ee5\u53ef\u4ee5\u7b80\u5316\u5bf9\u73af\u5883\u7684\u8981\u6c42":74,"\u6240\u4ee5\u5728\u5199\u5165\u5feb\u7167\u7684\u8fc7\u7a0b\u4e2d":20,"\u6240\u4ee5\u5916\u5c42\u8f93\u51fa\u7684\u5e8f\u5217\u5f62\u72b6":56,"\u6240\u4ee5\u5982\u679c\u9700\u8981\u81ea\u884c\u914d\u7f6e\u5f00\u53d1\u73af\u5883\u9700\u8981\u8003\u8651\u7248\u672c\u7684\u56e0\u7d20":49,"\u6240\u4ee5\u5b83\u4eec\u4f7f\u7528\u540c\u4e00\u4e2aip\u5730\u5740":73,"\u6240\u4ee5\u5bf9":56,"\u6240\u4ee5\u5f88\u591a\u65f6\u5019\u4f60\u9700\u8981\u505a\u7684\u53ea\u662f\u5b9a\u4e49\u6b63\u786e\u7684\u7f51\u7edc\u5c42\u5e76\u628a\u5b83\u4eec\u8fde\u63a5\u8d77\u6765":47,"\u6240\u4ee5\u6027\u80fd\u4e5f\u5c31\u9010\u6b65\u53d8\u6210\u4e86\u6df1\u5ea6\u5b66\u4e60\u9886\u57df\u6700\u91cd\u8981\u7684\u6307\u6807":67,"\u6240\u4ee5\u6211\u4eec\u4f7f\u7528\u8fd9\u4e2a\u955c\u50cf\u6765\u4e0b\u8f7d\u8bad\u7ec3\u6570\u636e\u5230docker":74,"\u6240\u4ee5\u6211\u4eec\u53ef\u4ee5\u5728\u8fd9\u4e2a\u57fa\u7840\u4e0a":75,"\u6240\u4ee5\u6211\u4eec\u9700\u8981\u5c06\u8f93\u5165\u6570\u636e\u6807\u8bb0\u6210":56,"\u6240\u4ee5\u63a8\u8350\u4f7f\u7528\u663e\u5f0f\u6307\u5b9a\u7684\u65b9\u5f0f\u6765\u8bbe\u7f6einput_typ":2,"\u6240\u4ee5\u653e\u4e00\u4e2a\u7a7a\u5217\u8868":47,"\u6240\u4ee5\u65e0\u9700\u5224\u65adavx\u5373\u53ef\u4f7f\u7528":49,"\u6240\u4ee5\u7528\u6237\u9700\u8981\u9996\u5148\u5728":28,"\u6240\u4ee5\u76f8\u6bd4\u4e8erecurr":46,"\u6240\u4ee5\u8bad\u7ec3":68,"\u6240\u4ee5\u8f93\u51fa\u7684value\u5305\u542b\u4e24\u4e2a\u5411\u91cf":4,"\u6240\u4ee5\u8fd9\u4e00\u6b65\u662f\u5fc5\u8981\u7684":62,"\u6240\u4f7f\u7528\u7684\u7f16\u8bd1\u5668\u4e3aarm":54,"\u6240\u5bf9\u5e94\u7684\u8bcd\u8868index\u6570\u7ec4":56,"\u6240\u6709\u4e0e\u7c7b\u578b\u76f8\u5173\u7684\u51fd\u6570":34,"\u6240\u6709\u4ee3\u7801\u5fc5\u987b\u5177\u6709\u5355\u5143\u6d4b\u8bd5":61,"\u6240\u6709\u53c2\u6570\u7f6e\u4e3a\u96f6":70,"\u6240\u6709\u540c\u76ee\u5f55\u4e0b\u7684\u6587\u672c\u5b9e\u4f8b\u6587\u4ef6\u90fd\u662f\u540c\u7ea7\u522b\u7684":88,"\u6240\u6709\u547d\u4ee4\u884c\u9009\u9879\u53ef\u4ee5\u8bbe\u7f6e\u4e3a":68,"\u6240\u6709\u6587\u4ef6\u5217\u8868":2,"\u6240\u6709\u672c\u5730\u8bad\u7ec3":68,"\u6240\u6709\u67b6\u6784":53,"\u6240\u6709\u6807\u8bb0\u7684\u6d4b\u8bd5\u96c6\u548c\u8bad\u7ec3\u96c6":88,"\u6240\u6709\u751f\u6210\u7684\u7684\u6587\u4ef6\u90fd\u5728build\u76ee\u5f55\u4e0b":49,"\u6240\u6709\u7684":[61,62],"\u6240\u6709\u7684\u4eba\u53e3\u7edf\u8ba1\u5b66\u4fe1\u606f\u7531\u7528\u6237\u81ea\u613f\u63d0\u4f9b":85,"\u6240\u6709\u7684\u5355\u6d4b\u90fd\u4f1a\u88ab\u6267\u884c\u4e00\u6b21":62,"\u6240\u6709\u7684\u63a5\u53e3\u5747\u4e3ac\u63a5\u53e3":34,"\u6240\u6709\u7684\u64cd\u4f5c\u90fd\u662f\u9488\u5bf9\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u6765\u8fdb\u884c\u7684":56,"\u6240\u6709\u7684\u7528\u6237\u4fe1\u606f\u90fd\u5305\u542b\u5728":85,"\u6240\u6709\u7684\u7535\u5f71\u4fe1\u606f\u90fd\u5305\u542b\u5728":85,"\u6240\u6709\u7684\u8bc4\u5206\u6570\u636e\u90fd\u5305\u542b\u5728":85,"\u6240\u6709\u7684python\u5c01\u88c5\u90fd\u4f7f\u7528":62,"\u6240\u6709\u7684python\u5c01\u88c5\u90fd\u5728":62,"\u6240\u6709\u7c7b\u578b\u540d\u4e3a":34,"\u6240\u6709\u7f51\u7edc\u5c42\u7684\u68af\u5ea6\u68c0\u67e5\u5355\u6d4b\u90fd\u4f4d\u4e8e":62,"\u6240\u6709\u8282\u70b9\u8fd0\u884c\u96c6\u7fa4\u4f5c\u4e1a\u7684\u4e3b\u673a\u540d\u6216":68,"\u6240\u6709\u8f93\u5165\u5e8f\u5217\u5e94\u8be5\u6709\u76f8\u540c\u7684\u957f\u5ea6":59,"\u6240\u6709\u914d\u7f6e\u90fd\u80fd\u5728":84,"\u6240\u6784\u5efa\u7f51\u7edc\u7ed3\u6784\u7684\u7684\u6df1\u5ea6\u6bd4\u4e4b\u524d\u4f7f\u7528\u7684\u7f51\u7edc\u6709\u5927\u5e45\u5ea6\u7684\u63d0\u9ad8":82,"\u6240\u793a":87,"\u6240\u8c13\u65f6\u95f4\u6b65\u4fe1\u606f":2,"\u6240\u9700\u652f\u6301\u7684\u6700\u4f4eandroid":53,"\u624b\u5199\u591a\u8bed\u8a00\u7ed1\u5b9a":33,"\u624b\u5de5\u827a\u8005":85,"\u624d\u4f1a\u91ca\u653e\u8be5\u6bb5\u5185\u5b58":2,"\u624d\u4f1astop":2,"\u624d\u80fd\u4fdd\u8bc1\u548c\u5355\u5c42\u5e8f\u5217\u7684\u914d\u7f6e\u4e2d":56,"\u624d\u80fd\u5145\u5206\u5229\u7528mac":49,"\u624d\u80fd\u53d1\u6325\u5176\u5168\u90e8\u80fd\u529b":67,"\u6253\u5370\u5728\u5c4f\u5e55\u4e0a":86,"\u6253\u5370\u7684\u65e5\u5fd7\u53d8\u591a":48,"\u6253\u5f00":67,"\u6253\u5f00\u6587\u672c\u6587\u4ef6":2,"\u6253\u5f00\u6d4f\u89c8\u5668\u8bbf\u95ee\u5bf9\u5e94\u76ee\u5f55\u4e0b\u7684index":65,"\u6253\u5f00\u8fd9\u4e2a\u7f16\u8bd1\u9009\u9879":34,"\u6267\u884c":[87,88],"\u6267\u884c\u4e0b\u9762\u7684\u547d\u4ee4\u5c31\u53ef\u4ee5\u9884\u5904\u7406\u6570\u6910":88,"\u6267\u884c\u4e0b\u9762\u7684\u547d\u4ee4\u6765\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5":63,"\u6267\u884c\u4ee5\u4e0b\u547d\u4ee4":[53,54],"\u6267\u884c\u4ee5\u4e0b\u64cd\u4f5c":59,"\u6267\u884c\u5b8c\u5b89\u88c5\u547d\u4ee4\u540e":[53,54],"\u6267\u884c\u60a8\u7684\u4ee3\u7801":67,"\u6267\u884c\u7684\u547d\u4ee4\u5982\u4e0b":82,"\u6269\u5c55\u673a\u5236\u7b49\u529f\u80fd":73,"\u6279\u6b21\u540e\u6253\u5370\u65e5\u5fd7":86,"\u6279\u6b21\u5bf9\u5e73\u5747\u53c2\u6570\u8fdb\u884c\u6d4b\u8bd5":87,"\u6279\u6b21\u7684\u6570\u636e":86,"\u627e\u5230":59,"\u627e\u5230\u6700\u65e9\u62a5\u9519\u7684\u5730\u65b9":46,"\u627e\u5230\u8fd0\u884c\u6162\u7684\u539f\u56e0":67,"\u627e\u5230\u8fd0\u884c\u6162\u7684\u90e8\u5206":67,"\u6280\u672f\u5458":85,"\u628a":[21,62],"\u628a\u4e4b\u524d\u793a\u4f8b\u4e2d\u8f6c\u6362\u5b8c\u6bd5\u7684random":21,"\u628a\u4efb\u610f\u7ef4\u5ea6\u7684tensor\u8f6c\u4e3a\u4e86\u4e00\u7ef4\u7684eigenvector":64,"\u628a\u5de5\u5177\u548c\u914d\u7f6e\u90fd\u5b89\u88c5\u5728\u4e00\u4e2a":60,"\u628a\u7528\u6237\u5728\u8d2d\u7269\u7f51\u7ad9":88,"\u628a\u7f16\u8bd1\u51fa\u6765\u7684paddlepaddle\u4e5f\u6253\u5305\u6210\u4e00\u4e2a\u955c":49,"\u628a\u8bad\u7ec3\u6570\u636e\u76f4\u63a5\u653e\u5728":74,"\u628a\u8fd9\u4e9b\u5de5\u5177\u5b89\u88c5\u5230\u672c\u673a":60,"\u6293\u53d6\u4ea7\u54c1\u7684\u7528\u6237\u8bc4\u8bba\u5e76\u5206\u6790\u4ed6\u4eec\u7684\u60c5\u611f":88,"\u6295\u5c04\u53cd\u5411rnn\u7684\u7b2c\u4e00\u4e2a\u5b9e\u4f8b\u5230":59,"\u6295\u5c04\u7f16\u7801\u5411\u91cf\u5230":59,"\u62bd\u53d6\u51fa\u7684\u65b0\u8bcd\u8868\u7684\u4fdd\u5b58\u8def\u5f84":80,"\u62bd\u53d6\u5bf9\u5e94\u7684\u8bcd\u5411\u91cf\u6784\u6210\u65b0\u7684\u8bcd\u8868":80,"\u62c6\u5206\u5230\u4e0d\u540c\u6587\u4ef6\u5939":89,"\u62c6\u89e3":58,"\u62c6\u89e3\u6210\u7684\u6bcf\u4e00\u53e5\u8bdd\u518d\u901a\u8fc7\u4e00\u4e2alstm\u7f51\u7edc":56,"\u62f7\u8d1d\u8bad\u7ec3\u6587\u4ef6\u5230\u5bb9\u5668\u5185":75,"\u62fc\u63a5":46,"\u62fc\u63a5\u6210\u4e00\u4e2a\u65b0\u7684\u5411\u91cf":84,"\u6301\u7eed\u96c6\u6210\u7cfb\u7edf\u4e5f\u53ef\u4ee5\u7b80\u5355\u5730\u590d\u73b0\u8fd9\u4e2a\u8fc7\u7a0b":49,"\u6302\u8f7dvolum":49,"\u6307\u5411\u4e00\u4e2alayer":58,"\u6307\u5b9a":[46,58,59],"\u6307\u5b9a\u4e00\u53f0\u673a\u5668\u4e0a\u4f7f\u7528\u7684\u7ebf\u7a0b\u6570":70,"\u6307\u5b9a\u4f7f\u75282":46,"\u6307\u5b9a\u521d\u59cb\u5316\u6a21\u578b\u8def\u5f84":84,"\u6307\u5b9a\u524d\u5411\u7f51\u7edc\u6700\u7ec8\u7684\u8f93\u51fa\u76ee\u6807\u53d8\u91cf":63,"\u6307\u5b9a\u52a0\u8f7d\u7684\u65b9\u5f0f":70,"\u6307\u5b9a\u5bf9\u8f93\u5165\u53d8\u91cf":63,"\u6307\u5b9a\u5de5\u4f5c\u6a21\u578b\u8fdb\u884c\u9884\u6d4b":82,"\u6307\u5b9a\u5de5\u4f5c\u6a21\u5f0f\u6765\u63d0\u53d6\u7279\u5f81":82,"\u6307\u5b9a\u63d0\u53d6\u7279\u5f81\u7684\u5c42":82,"\u6307\u5b9a\u662f\u5426\u4f7f\u7528gpu":82,"\u6307\u5b9a\u751f\u6210\u6570\u636e\u7684\u51fd\u6570":84,"\u6307\u5b9a\u7684\u5b57\u5178\u5355\u8bcd\u6570":89,"\u6307\u5b9a\u7684\u6570\u636e\u5c06\u4f1a\u88ab\u6d4b\u8bd5":84,"\u6307\u5b9a\u7684\u8f93\u5165\u4e0d\u4f1a\u88ab":58,"\u6307\u5b9a\u8981\u8f93\u51fa\u7684\u5b57\u6bb5\u8fdb\u884c\u8f93\u51fa":46,"\u6307\u5b9a\u8bad\u7ec3\u6570\u636e\u548c\u6d4b\u8bd5\u6570\u636e":84,"\u6307\u5b9abatch":89,"\u6307\u5b9acudnn\u7684\u6700\u5927\u5de5\u4f5c\u7a7a\u95f4\u5bb9\u9650":70,"\u6307\u5bf9\u4e8e\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217\u8f93\u5165\u6570\u636e":56,"\u6307\u6df1\u5ea6\u5b66\u4e60\u8bad\u7ec3\u4e4b\u540e\u5f97\u5230\u7684\u6240\u6709\u53c2\u6570":20,"\u6307\u793a\u4f7f\u7528\u54ea\u4e2agpu\u6838":70,"\u6307\u793a\u5728\u7b80\u5355\u7684recurrentlayer\u5c42\u7684\u8ba1\u7b97\u4e2d\u662f\u5426\u4f7f\u7528\u6279\u5904\u7406\u65b9\u6cd5":70,"\u6307\u793a\u5f53\u6307\u5b9a\u8f6e\u7684\u6d4b\u8bd5\u6a21\u578b\u4e0d\u5b58\u5728\u65f6":70,"\u6307\u793a\u662f\u5426\u4f7f\u7528\u591a\u7ebf\u7a0b\u6765\u8ba1\u7b97\u4e00\u4e2a\u795e\u7ecf\u7f51\u7edc":70,"\u6307\u793a\u662f\u5426\u5f00\u542f\u53c2\u6570\u670d\u52a1\u5668":70,"\u6307\u793a\u662f\u5426\u663e\u793a\u53c2\u6570\u670d\u52a1\u5668\u4e0a\u7684\u7a00\u758f\u53c2\u6570\u5206\u5e03\u7684\u65e5\u5fd7\u7ec6\u8282":70,"\u6307\u793a\u662f\u5426\u68c0\u67e5\u6240\u6709\u53c2\u6570\u670d\u52a1\u5668\u4e0a\u7684\u7a00\u758f\u53c2\u6570\u7684\u5206\u5e03\u662f\u5747\u5300\u7684":70,"\u6307\u793a\u6d4b\u8bd5\u4efb\u52a1":87,"\u6307\u793a\u6d4b\u8bd5\u4efb\u52a1\u7684\u6807\u8bb0":87,"\u6309\u542f\u53d1\u5f0f\u635f\u5931\u7684\u5927\u5c0f\u9012\u589e\u6392\u5e8f":70,"\u6309\u7167":54,"\u6309\u7167\u4e0b\u9762\u6b65\u9aa4\u5373\u53ef":75,"\u6309\u7167\u5176\u5185\u5bb9\u521b\u5efa\u4e00\u4e2a\u540d\u4e3a":60,"\u6309\u7167\u5177\u4f53\u5b9e\u73b0\u65b9\u5f0f\u53ef\u4ee5\u5f52\u7eb3\u4e3a2\u7c7b":46,"\u6309\u94ae":61,"\u633a":56,"\u633a\u597d":56,"\u6355\u83b7":84,"\u6362":56,"\u6392\u6210\u4e00\u5217\u7684\u591a\u4e2a\u5143\u7d20":55,"\u63a5\u4e0a\u4e00\u4e2a\u5168\u8fde\u63a5\u5c42":51,"\u63a5\u4e0a\u5e73\u65b9\u8bef\u5dee\u5c42":51,"\u63a5\u4e0b\u6765":[63,84,88],"\u63a5\u4e0b\u6765\u53ef\u4ee5\u8003\u8651\u4e0b\u65f6\u95f4\u7ebf\u7684\u5206\u6790":67,"\u63a5\u4e0b\u6765\u5c31\u53ef\u4ee5\u4f7f\u7528":67,"\u63a5\u4e0b\u6765\u6211\u4eec\u521b\u5efa\u4e00\u4e2a\u539f\u59cb":61,"\u63a5\u4e0b\u6765\u6211\u4eec\u53d6\u6d88\u5bf9":61,"\u63a5\u4e0b\u6765\u6211\u4eec\u5c06\u5c55\u793a\u5982\u4f55\u7528paddlepaddle\u8bad\u7ec3\u4e00\u4e2a\u6587\u672c\u5206\u7c7b\u6a21\u578b":84,"\u63a5\u4e0b\u6765\u7b49\u5f85":61,"\u63a5\u53d7":87,"\u63a5\u53d7\u4e00\u4e2a\u8f93\u5165\u53c2\u6570":63,"\u63a5\u53d7\u7684\u4e1c\u897f":87,"\u63a5\u53d7\u8005":87,"\u63a5\u53e3":[33,34,63,64],"\u63a5\u53e3\u4f1a\u88ab\u8c03\u7528":64,"\u63a5\u53e3\u540d\u79f0":68,"\u63a5\u53e3\u5c42\u505a\u8fc7\u591a\u5c01\u88c5":34,"\u63a5\u53e3\u63d0\u53d6\u7684\u7ed3\u679c\u662f\u4e00\u81f4\u7684":82,"\u63a5\u53e3\u662f":21,"\u63a5\u53e3\u6700\u7ec8\u4f1a\u8c03\u7528\u5bf9\u5e94":64,"\u63a5\u53e3\u6709\u4e00\u4e2a":46,"\u63a5\u53e3\u6765\u52a0\u8f7d\u6570\u636e":84,"\u63a5\u53e3\u6765\u52a0\u8f7d\u8be5\u6587\u4ef6":82,"\u63a5\u53e3\u6765\u6253\u5f00\u8be5\u6587\u4ef6":82,"\u63a5\u53e3\u7684":46,"\u63a5\u6536\u5904\u7406pfsclient\u7aef\u7684\u6587\u4ef6\u7ba1\u7406\u8bf7\u6c42":28,"\u63a5\u7740\u6211\u4eec\u5c31\u80fd\u591f\u6253\u5f00\u6d4f\u89c8\u5668\u5728":49,"\u63a7\u5236":70,"\u63a7\u5236\u5982\u4f55\u6539\u53d8\u6a21\u578b\u53c2\u6570":47,"\u63a7\u5236\u662f\u5426\u751f\u6210\u5355\u5143\u6d4b\u8bd5":49,"\u63a7\u5236\u751f\u6210\u7684\u751f\u4ea7\u955c\u50cf\u662f\u5426\u652f\u6301avx":49,"\u63a7\u5236\u751f\u6210\u7684\u751f\u4ea7\u955c\u50cf\u662f\u5426\u652f\u6301gpu":49,"\u63a7\u5236\u7528\u6237\u6743\u9650":21,"\u63a8\u5bfc\u8be5\u5c42\u524d\u5411\u548c\u540e\u5411\u4f20\u9012\u7684\u65b9\u7a0b":62,"\u63a8\u8350":56,"\u63a8\u8350\u4f7f\u7528":2,"\u63a8\u8350\u56fd\u5185\u7528\u6237\u4f7f\u7528docker":49,"\u63a8\u8350\u6e05\u7406\u6574\u4e2a\u7f16\u8bd1\u76ee\u5f55":48,"\u63a8\u8350\u76f4\u63a5\u5b58\u653e\u5230\u8bad\u7ec3\u76ee\u5f55":1,"\u63a8\u8350\u7cfb\u7edf":68,"\u63a8\u9001\u5230\u8fdc\u7a0b\u4ed3\u5e93":61,"\u63a8\u9500\u5458":85,"\u63cf\u8ff0":48,"\u63cf\u8ff0\u7684\u9ed8\u8ba4\u5165\u53e3\u7a0b\u5e8f":60,"\u63cf\u8ff0\u7f51\u7edc\u7ed3\u6784\u548c\u4f18\u5316\u7b97\u6cd5":84,"\u63cf\u8ff0\u8be5op\u7684\u8f93\u5165":63,"\u63cf\u8ff0\u95ee\u9898":61,"\u63cf\u8ff0kubernetes\u4e0a\u8fd0\u884c\u7684\u4f5c\u4e1a":73,"\u63d0\u4ea4\u65b9\u5f0f\u53c2\u89c1":65,"\u63d0\u4ea4pull":61,"\u63d0\u4f9b":68,"\u63d0\u4f9b\u4e03\u5c42\u534f\u8bae\u7684\u53cd\u5411\u4ee3\u7406":28,"\u63d0\u4f9b\u4e86\u4e00\u4e2a\u542f\u52a8\u811a\u672c":75,"\u63d0\u4f9b\u4e86\u547d\u4ee4\u6837\u4f8b\u6765\u8fd0\u884c":68,"\u63d0\u4f9b\u4e86\u81ea\u52a8\u5316\u811a\u672c\u6765\u542f\u52a8\u4e0d\u540c\u8282\u70b9\u4e2d\u7684\u6240\u6709":68,"\u63d0\u4f9b\u51e0\u4e4e\u6240\u6709\u8bad\u7ec3\u7684\u5185\u90e8\u8f93\u51fa\u65e5\u5fd7":68,"\u63d0\u4f9b\u5e38\u7528\u7684\u547d\u4ee4\u884c\u7ba1\u7406\u547d\u4ee4\u7ba1\u7406\u6587\u4ef6\u548c\u76ee\u5f55":28,"\u63d0\u4f9b\u6269\u5c55\u7684\u957f\u5ea6\u4fe1\u606f":55,"\u63d0\u4f9b\u6700\u65b0":49,"\u63d0\u4f9b\u7528\u6237\u7ba1\u7406\u6587\u4ef6\u7684\u547d\u4ee4":28,"\u63d0\u4f9b\u7ed9paddle\u4f5c\u4e3a\u8bad\u7ec3\u6570\u636e":21,"\u63d0\u4f9b\u8bad\u7ec3\u8fc7\u7a0b\u7684":68,"\u63d0\u51fa\u7684\u4ee3\u7801\u9700\u6c42":80,"\u63d0\u793a":46,"\u642d\u5efa\u795e\u7ecf\u7f51\u7edc\u5c31\u50cf\u4f7f\u7528\u79ef\u6728\u642d\u5efa\u5b9d\u5854\u4e00\u6837":51,"\u64cd\u4f5c":56,"\u64cd\u4f5c\u7cfb\u7edf":60,"\u64cd\u6301\u5bb6\u52a1\u8005":85,"\u652f\u6301\u53cc\u5c42\u5e8f\u5217\u4f5c\u4e3a\u8f93\u5165\u7684layer":[57,58],"\u652f\u6301\u5927\u6587\u4ef6\u7684\u65ad\u70b9\u4e0a\u4f20":28,"\u652f\u6301\u7684\u6700\u5c0f\u7684android":53,"\u652f\u6301rbd":73,"\u653e\u5728\u8fd9\u4e2a\u76ee\u5f55\u91cc\u7684\u6587\u4ef6\u5176\u5b9e\u662f\u4fdd\u5b58\u5230\u4e86mfs\u4e0a":75,"\u653e\u5fc3":56,"\u6545\u800c\u662f\u4e00\u4e2a\u5355\u5c42\u65f6\u95f4\u5e8f\u5217":56,"\u6548\u679c\u603b\u7ed3":84,"\u6559\u7a0b\u6587\u6863\u5230":68,"\u6559\u80b2\u5de5\u4f5c\u8005":85,"\u6570":[58,87],"\u6570\u5fc5\u987b\u4e25\u683c\u76f8\u7b49":58,"\u6570\u636e":[28,89],"\u6570\u636e\u4e0b\u8f7d\u4e4b\u540e":81,"\u6570\u636e\u4e2d0":46,"\u6570\u636e\u5217\u8868":82,"\u6570\u636e\u5c06\u4fdd\u5b58\u5728":80,"\u6570\u636e\u5c42":[47,86],"\u6570\u636e\u5e94\u8be5\u5728\u542f\u52a8\u96c6\u7fa4\u4f5c\u4e1a\u4e4b\u524d\u51c6\u5907\u597d":68,"\u6570\u636e\u63d0\u4f9b\u5668":69,"\u6570\u636e\u63d0\u4f9b\u811a\u672c\u4ec5\u4ec5\u662f\u8bfb\u53d6meta":86,"\u6570\u636e\u63d0\u4f9b\u811a\u672c\u7684\u7ec6\u8282\u6587\u6863\u53ef\u4ee5\u53c2\u8003":86,"\u6570\u636e\u76ee\u5f55\u4e2d\u7684\u6240\u6709\u6587\u4ef6\u88ab":68,"\u6570\u636e\u7c7b\u578b":4,"\u6570\u636e\u7f13\u5b58\u7684\u7b56\u7565":2,"\u6570\u636e\u8bbf\u95ee":0,"\u6570\u636e\u8bfb\u53d6\u5747\u4ea4\u7531\u5176\u4ed6\u8bed\u8a00\u5b8c\u6210":33,"\u6570\u636e\u8f93\u5165":58,"\u6570\u636e\u8f93\u5165\u683c\u5f0f":2,"\u6570\u636e\u957f\u5ea6\u53ca\u6821\u9a8c\u503c\u7ec4\u6210":28,"\u6570\u636e\u96c6":85,"\u6570\u636e\u96c6\u63cf\u8ff0":86,"\u6570\u636e\u96c6\u6587\u4ef6\u5939\u540d\u79f0":89,"\u6570\u636e\u96c6\u9700\u8981\u9884\u5148\u88ab\u8f6c\u6362\u6210paddlepaddle\u5206\u5e03\u5f0f\u8bad\u7ec3\u4f7f\u7528\u7684\u5b58\u50a8\u683c":21,"\u6570\u636e\u9884\u5904\u7406\u4efb\u52a1":21,"\u6570\u636e\u9884\u5904\u7406\u5b8c\u6210\u4e4b\u540e":84,"\u6570\u636e\u9884\u6d4b":87,"\u6570\u6910\u5b9a\u4e49":88,"\u6570\u6910\u8bf4\u660e\u6587\u6863":88,"\u6570\u6910\u96c6\u548c":88,"\u6570\u76ee":72,"\u6574\u4f53":56,"\u6574\u4f53\u6570\u636e\u548c\u539f\u59cb\u6570\u636e\u5b8c\u5168\u4e00\u6837":56,"\u6574\u4f53\u7684\u7ed3\u6784\u56fe\u5982\u4e0b":75,"\u6574\u6570":62,"\u6574\u6570\u6807\u7b7e":[2,51],"\u6574\u6d01":56,"\u6587\u4e66\u5de5\u4f5c":85,"\u6587\u4ef6":[33,60,61,63,74,87],"\u6587\u4ef6\u4e2d":[63,75,82,85,87],"\u6587\u4ef6\u4e2d\u6307\u5b9a\u6a21\u578b\u8def\u5f84\u548c\u8f93\u51fa\u7684\u76ee\u5f55":82,"\u6587\u4ef6\u4e2d\u6307\u5b9a\u8981\u63d0\u53d6\u7279\u5f81\u7684\u7f51\u7edc\u5c42\u7684\u540d\u5b57":82,"\u6587\u4ef6\u4e2d\u6ce8\u518c\u524d\u5411":63,"\u6587\u4ef6\u4e2d\u6ce8\u518c\u8be5op\u548ckernel":63,"\u6587\u4ef6\u4e2d\u6ce8\u518cgpu":63,"\u6587\u4ef6\u4e2d\u7684":82,"\u6587\u4ef6\u4e2d\u7684\u6bcf\u884c\u90fd\u5fc5\u987b\u662f\u4e00\u4e2a\u53e5\u5b50":89,"\u6587\u4ef6\u4e3a":[46,89],"\u6587\u4ef6\u4e4b\u5916":61,"\u6587\u4ef6\u4e5f\u53ef\u4ee5\u7528\u4e8e\u5bf9\u6837\u672c\u8fdb\u884c\u9884\u6d4b":82,"\u6587\u4ef6\u4f20\u8f93\u7684\u7684\u5173\u952e\u5728\u4e8e\u9700\u8981pfsclient\u7aef\u5bf9\u6bd4source\u548cdestination\u7684\u6587\u4ef6chunks\u7684checksum\u662f\u5426\u4fdd\u6301\u4e00\u81f4":28,"\u6587\u4ef6\u5185\u5bb9\u4e3a":33,"\u6587\u4ef6\u5206\u5272\u4e3a\u8bad\u7ec3\u548c\u6d4b\u8bd5\u6587\u4ef6":86,"\u6587\u4ef6\u540d\u4e3a\u6b64uuid":20,"\u6587\u4ef6\u540d\u79f0\u4e3a":86,"\u6587\u4ef6\u547d\u540d\u4ee5":63,"\u6587\u4ef6\u59390":75,"\u6587\u4ef6\u5939\u4e0b":65,"\u6587\u4ef6\u5939\u4e2d\u7684\u6bcf\u4e2a\u6587\u4ef6\u7684\u6bcf\u4e00\u884c\u5305\u542b\u4e24\u90e8\u5206":89,"\u6587\u4ef6\u5bf9\u5e94\u7684data":21,"\u6587\u4ef6\u7684\u4e0a\u4f20\u548c\u4e0b\u8f7d\u90fd\u662f\u901a\u8fc7\u5bf9chunk\u7684\u64cd\u4f5c\u6765\u5b9e\u73b0\u7684":28,"\u6587\u4ef6\u7684\u5206\u9694\u7b26\u4e3a":86,"\u6587\u4ef6\u7684\u6539\u53d8":61,"\u6587\u4ef6\u7684\u683c\u5f0f\u53ef\u4ee5":86,"\u6587\u4ef6\u7a0d\u6709\u5dee\u522b":81,"\u6587\u4ef6\u7cfb\u7edf\u4ee5\u53ca\u7f51\u7edc\u4e4b\u4e0a":49,"\u6587\u4ef6\u7d22\u5f15":68,"\u6587\u4ef6\u7ed9\u51fa\u4e86\u5b8c\u6574\u4f8b\u5b50":84,"\u6587\u4ef6model":72,"\u6587\u5b57\u7684\u4ea4\u4e92\u5f0f\u6587\u6863":49,"\u6587\u672c\u4e2d\u7684\u5355\u8bcd\u7528\u7a7a\u683c\u5206\u9694":84,"\u6587\u672c\u4fe1\u606f\u5c31\u662f\u4e00\u4e2a\u5e8f\u5217\u6570\u636e":2,"\u6587\u672c\u5206\u7c7b\u95ee\u9898":84,"\u6587\u672c\u5377\u79ef\u5206\u53ef\u4e3a\u4e09\u4e2a\u6b65\u9aa4":84,"\u6587\u672c\u5377\u79ef\u91c7\u6837\u5c42":86,"\u6587\u672c\u6295\u5f71\u5c42":86,"\u6587\u672c\u683c\u5f0f\u7684\u5b9e\u4f8b\u6587\u4ef6":88,"\u6587\u6863":46,"\u6587\u68631":64,"\u6587\u68632":64,"\u6587\u6863\u7f16\u5199\u7b49\u5de5\u4f5c":49,"\u6587\u6863\u81ea\u52a8\u5206\u7c7b\u548c\u95ee\u7b54":87,"\u6587\u6863\u8f83\u5c11":64,"\u6587\u6863\u90fd\u662f\u901a\u8fc7":65,"\u6587\u7ae0":75,"\u65b0":56,"\u65b0\u5efa\u4e00\u4e2a\u6743\u91cd":62,"\u65b0\u624b\u5165\u95e8":78,"\u65b0\u624b\u5165\u95e8\u7ae0\u8282":42,"\u65b9\u4fbf":56,"\u65b9\u4fbf\u4eca\u540e\u7684\u5d4c\u5165\u5f0f\u79fb\u690d\u5de5\u4f5c":48,"\u65b9\u4fbf\u5f00\u53d1\u8005\u76f4\u63a5\u767b\u5f55\u5230\u955c\u50cf\u4e2d\u8fdb\u884c\u5f00\u53d1":49,"\u65b9\u4fbf\u6d4b\u8bd5\u4eba\u5458\u6d4b\u8bd5paddlepaddle\u7684\u884c\u4e3a":42,"\u65b9\u4fbf\u7528\u6237\u4e0a\u4f20\u81ea\u5df1\u7684\u8bad\u7ec3\u6570\u636e\u4ee5\u8fdb\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3":28,"\u65b9\u5f0f1":46,"\u65b9\u5f0f2":46,"\u65b9\u6848\u6765\u6807\u8bb0\u6bcf\u4e2a\u53c2\u6570":87,"\u65b9\u6cd5\u4e00":72,"\u65b9\u6cd5\u4e09":72,"\u65b9\u6cd5\u4e8c":72,"\u65c1\u8fb9":56,"\u65c5\u6e38\u7f51\u7ad9":88,"\u65e0":56,"\u65e0\u4e1a\u4eba\u58eb":85,"\u65e0\u5ef6\u8fdf":70,"\u65e0\u6cd5\u505a\u5230\u5bf9\u4e8e\u5404\u79cd\u8bed\u8a00\u9519\u8bef\u5904\u7406\u65b9\u5f0f\u7684\u9002\u914d":33,"\u65e0\u8bba\u5728\u672c\u5730\u8fd8\u662f\u5728\u4e91\u7aef":21,"\u65e0\u8bba\u662f\u4ece":21,"\u65e0\u8bba\u662f\u5728\u672c\u5730\u6216\u662f\u4e91\u7aef\u8f6c\u6362":21,"\u65e0\u9ed8\u8ba4\u503c":[53,54],"\u65e5\u5fd7\u5c06\u4fdd\u5b58\u5728":88,"\u65e5\u5fd7\u62a5\u9519\u4e3a\u7f51\u7edc\u901a\u4fe1\u7c7b\u9519\u8bef":46,"\u65e8\u5728\u5efa\u7acb\u4e00\u4e2a\u53ef\u4ee5\u88ab\u534f\u540c\u8c03\u81f3\u6700\u4f18\u7ffb\u8bd1\u6548\u679c\u7684\u5355\u795e\u7ecf\u5143\u7f51\u7edc":89,"\u65e9\u9910":56,"\u65f6":[20,46,55,59,62,70,75],"\u65f6\u5019":56,"\u65f6\u52a0\u4e0a":88,"\u65f6\u5e8f\u6a21\u578b\u5747\u4f7f\u7528\u8be5\u811a\u672c":84,"\u65f6\u5e8f\u6a21\u578b\u662f\u6307\u6570\u636e\u7684\u67d0\u4e00\u7ef4\u5ea6\u662f\u4e00\u4e2a\u5e8f\u5217\u5f62\u5f0f":2,"\u65f6\u76ee\u6807\u8bed\u8a00\u7684\u6587\u4ef6":89,"\u65f6\u88ab\u8bad\u7ec3\u7684":62,"\u65f6\u8bbe\u5907id\u53f7\u7684\u5206\u914d":72,"\u65f6\u95f4":56,"\u65f6\u95f4\u6233":85,"\u65f6\u95f4\u6233\u8868\u793a\u4e3a\u4ece1970":85,"\u65f6\u95f4\u6b65\u7684\u6982\u5ff5":56,"\u65f6\u987b\u4ece\u7b2c17\u5b57\u8282\u5f00\u59cb":46,"\u6620\u5c04\u4e3a":60,"\u6620\u5c04\u5230\u4e00\u4e2a\u7ef4\u5ea6\u4e3a":62,"\u662f":[28,46,48,56],"\u662f\u4e00\u4e2a\u51681\u7684\u5411\u91cf":62,"\u662f\u4e00\u4e2a\u5185\u7f6e\u7684\u5b9a\u65f6\u5668\u5c01\u88c5":67,"\u662f\u4e00\u4e2a\u52a8\u6001\u7a0b\u5e8f\u5206\u6790\u7684\u672f\u8bed":67,"\u662f\u4e00\u4e2a\u5305\u88c5\u6570\u636e\u7684":87,"\u662f\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":55,"\u662f\u4e00\u4e2a\u53cc\u5c42\u7684\u5e8f\u5217":55,"\u662f\u4e00\u4e2a\u591a\u8bed\u8a00\u63a5\u53e3\u7684\u4ee3\u7801\u751f\u6210\u5668":33,"\u662f\u4e00\u4e2a\u5c01\u88c5\u5bf9\u8c61":67,"\u662f\u4e00\u4e2a\u5f88\u6709\u7528\u7684\u53c2\u6570":72,"\u662f\u4e00\u4e2a\u7b26\u5408\u9ad8\u65af\u5206\u5e03\u7684\u968f\u673a\u53d8\u91cf":47,"\u662f\u4e00\u4e2a\u7c7b\u578b\u7684\u6807\u5fd7":34,"\u662f\u4e00\u4e2a\u7edf\u8ba1\u5b66\u7684\u673a\u5668\u7ffb\u8bd1\u7cfb\u7edf":89,"\u662f\u4e00\u4e2a\u914d\u7f6e\u6587\u4ef6\u7684\u4f8b\u5b50":88,"\u662f\u4e00\u4e2a\u975e\u7ebf\u6027\u7684":62,"\u662f\u4e00\u4e2apython\u7684":2,"\u662f\u4e00\u4e2aunbound":58,"\u662f\u4e00\u6761\u65f6\u95f4\u5e8f\u5217":[2,51],"\u662f\u4e00\u79cd\u4efb\u610f\u590d\u6742\u7684rnn\u5355\u5143":58,"\u662f\u4e00\u7ec4":73,"\u662f\u4e0d\u5305\u62ec\u6e90\u7801\u7684":74,"\u662f\u4e0d\u5e38\u89c1\u7684\u505a\u6cd5":33,"\u662f\u4e0d\u662f\u5f88\u7b80\u5355\u5462":2,"\u662f\u4e0d\u662f\u8981\u5bf9\u6570\u636e\u505ashuffl":2,"\u662f\u4e3b\u8981\u7684\u53ef\u6267\u884cpython\u811a\u672c":87,"\u662f\u4ec0\u4e48\u4e5f\u6ca1\u5173\u7cfb":2,"\u662f\u4f17\u591a\u8bef\u5dee\u4ee3\u4ef7\u51fd\u6570\u5c42\u7684\u4e00\u79cd":47,"\u662f\u4f7f\u5f97\u8981\u5171\u4eab\u7684\u53c2\u6570\u4f7f\u7528\u540c\u6837\u7684":46,"\u662f\u504f\u5dee":59,"\u662f\u5176\u5927\u5c0f":47,"\u662f\u51e0\u4e4e\u4e0d\u5360\u5185\u5b58\u7684":2,"\u662f\u539f\u59cb\u6cd5\u8bed\u6587\u4ef6":89,"\u662f\u5404\u4e2a\u5b9e\u73b0\u4e2d\u5171\u4eab\u7684\u5934\u6587\u4ef6":34,"\u662f\u5411\u91cf":62,"\u662f\u5426\u4ee5\u9006\u5e8f\u5904\u7406\u8f93\u5165\u5e8f\u5217":58,"\u662f\u5426\u4f7f\u7528\u53cc\u7cbe\u5ea6\u6d6e\u70b9\u6570":48,"\u662f\u5426\u4f7f\u7528\u65e7\u7684remoteparameterupdat":70,"\u662f\u5426\u4f7f\u7528\u6743\u91cd":62,"\u662f\u5426\u4f7f\u7528arm\u6a21\u5f0f":53,"\u662f\u5426\u4f7f\u7528gpu":86,"\u662f\u5426\u4f7f\u7528gpu\u8bad\u7ec3":89,"\u662f\u5426\u4f7f\u7528neon\u6307\u4ee4":[53,54],"\u662f\u5426\u5141\u8bb8\u6682\u5b58\u7565\u5fae\u591a\u4f59pool_size\u7684\u6570\u636e":2,"\u662f\u5426\u5185\u5d4cpython\u89e3\u91ca\u5668":48,"\u662f\u5426\u5219\u5171\u4eab\u540c\u4e00\u4e2a":63,"\u662f\u5426\u5c06\u5168\u5c40\u79cd\u5b50\u5e94\u7528\u4e8e\u672c\u5730\u7ebf\u7a0b\u7684\u968f\u673a\u6570":70,"\u662f\u5426\u5f00\u542f\u5355\u5143\u6d4b\u8bd5":48,"\u662f\u5426\u5f00\u542f\u8ba1\u65f6\u529f\u80fd":48,"\u662f\u5426\u5f00\u542frdma":48,"\u662f\u5426\u6253\u5370\u7248\u672c\u4fe1\u606f":70,"\u662f\u5426\u652f\u6301gpu":48,"\u662f\u5426\u663e\u793a":70,"\u662f\u5426\u7a00\u758f":62,"\u662f\u5426\u7f16\u8bd1\u4e2d\u82f1\u6587\u6587\u6863":48,"\u662f\u5426\u7f16\u8bd1\u542b\u6709avx\u6307\u4ee4\u96c6\u7684paddlepaddle\u4e8c\u8fdb\u5236\u6587\u4ef6":48,"\u662f\u5426\u7f16\u8bd1\u65f6\u8fdb\u884c\u4ee3\u7801\u98ce\u683c\u68c0\u67e5":48,"\u662f\u5426\u7f16\u8bd1python\u7684swig\u63a5\u53e3":48,"\u662f\u5426\u8fd0\u884c\u65f6\u52a8\u6001\u52a0\u8f7dcuda\u52a8\u6001\u5e93":48,"\u662f\u5426\u9700\u8981\u7b49\u5f85\u8be5\u8f6e\u6a21\u578b\u53c2\u6570":70,"\u662f\u56e0\u4e3a\u8fd9\u4e2a\u6d41\u7a0b\u6bd4\u5176\u4ed6\u65b9\u6cd5\u90fd\u66f4\u7b80\u4fbf":60,"\u662f\u56e0\u4e3ac99\u652f\u6301":33,"\u662f\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u52a0\u8f7d\u5b57\u5178\u5e76\u5b9a\u4e49\u6570\u636e\u63d0\u4f9b\u7a0b\u5e8f\u6a21\u5757\u548c\u7f51\u7edc\u67b6\u6784\u7684\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6":87,"\u662f\u5728paddlepaddle\u4e2d\u6784\u9020\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u65f6\u6700\u91cd\u8981\u7684\u6982\u5ff5":59,"\u662f\u57fa\u7840\u7684\u8ba1\u7b97\u5355\u5143":47,"\u662f\u5b58\u6709\u4e00\u7cfb\u5217\u53d8\u6362\u77e9\u9635\u7684\u6743\u91cd":62,"\u662f\u5b58\u6709\u504f\u7f6e\u5411\u91cf\u7684\u6743\u91cd":62,"\u662f\u5bf9\u7528\u6237\u6587\u4ef6\u5b58\u50a8\u7a7a\u95f4\u7684\u62bd\u8c61":28,"\u662f\u5e8f\u5217":86,"\u662f\u5f85\u6269\u5c55\u7684\u6570\u636e":55,"\u662f\u6211\u4eec":61,"\u662f\u6307":34,"\u662f\u6307\u4e00\u7cfb\u5217\u7684\u7279\u5f81\u6570\u636e":56,"\u662f\u6307recurrent_group\u7684\u591a\u4e2a\u8f93\u5165\u5e8f\u5217":56,"\u662f\u6570\u636e\u8f93\u5165":59,"\u662f\u6574\u4e2a\u7684\u8bcd\u5d4c\u5165":86,"\u662f\u6709\u610f\u4e49\u7684":56,"\u662f\u6784\u5efa\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u6700\u91cd\u8981\u7684\u5de5\u5177":59,"\u662f\u6a21\u578b\u53c2\u6570\u4f18\u5316\u7684\u76ee\u6807\u51fd\u6570":47,"\u662f\u6ca1\u6709\u540d\u5b57\u7684":60,"\u662f\u6d45\u5c42\u8bed\u4e49\u89e3\u6790\u7684\u4e00\u79cd\u5f62\u5f0f":87,"\u662f\u6e90\u8bed\u8a00\u7684\u6587\u4ef6":89,"\u662f\u7528\u6237\u4f7f\u7528c":34,"\u662f\u7684":60,"\u662f\u76ee\u6807\u82f1\u8bed\u6587\u4ef6":89,"\u662f\u77e9\u9635":62,"\u662f\u795e\u7ecf\u7f51\u7edc\u7684\u5165\u53e3":47,"\u662f\u7f51\u7edc\u548c\u6570\u636e\u914d\u7f6e\u6587\u4ef6":81,"\u662f\u7f51\u7edc\u5c42\u5b9e\u4f8b\u7684\u540d\u5b57\u6807\u8bc6\u7b26":62,"\u662f\u7f51\u7edc\u5c42\u7684\u6807\u8bc6\u7b26":62,"\u662f\u7f51\u7edc\u5c42\u7684\u7c7b\u578b":62,"\u662f\u7f51\u7edc\u5c42\u8f93\u51fa\u7684\u5927\u5c0f":62,"\u662f\u8be5\u5c42\u7684\u6807\u8bc6\u7b26":62,"\u662f\u8be5\u5c42\u7684\u7c7b\u540d":62,"\u662f\u8be5\u7f51\u7edc\u5c42\u7684":62,"\u662f\u8f93\u5165":59,"\u662f\u8fd9\u4e00\u7c7b\u7684":46,"\u662f\u901a\u7528\u7269\u4f53\u5206\u7c7b\u9886\u57df\u4e00\u4e2a\u4f17\u6240\u5468\u77e5\u7684\u6570\u636e\u5e93":82,"\u662f\u9700\u8981\u4e86\u89e3\u54ea\u4e9b\u6b65\u9aa4\u62d6\u6162\u4e86\u6574\u4f53":67,"\u662fc":34,"\u662fdecoder\u7684\u6570\u636e\u8f93\u5165":58,"\u662fgoogle\u5f00\u6e90\u7684\u5bb9\u5668\u96c6\u7fa4\u7ba1\u7406\u7cfb\u7edf":73,"\u662fnvidia\u6027\u80fd\u5206\u6790\u5de5\u5177":67,"\u662fpaddlepaddle\u652f\u6301\u7684\u4e00\u79cd\u4efb\u610f\u590d\u6742\u7684rnn\u5355\u5143":58,"\u662fpaddlepaddle\u8d1f\u8d23\u63d0\u4f9b\u6570\u636e\u7684\u6a21\u5757":84,"\u662fpod\u5185\u7684\u5bb9\u5668\u90fd\u53ef\u4ee5\u8bbf\u95ee\u7684\u5171\u4eab\u76ee\u5f55":73,"\u662fpython\u5c01\u88c5\u7684\u7c7b\u540d":62,"\u662frnn\u72b6\u6001":59,"\u663e":84,"\u663e\u5f0f\u6307\u5b9a\u8fd4\u56de\u7684\u662f\u4e00\u4e2a28":2,"\u665a":56,"\u666e\u901a\u7528\u6237\u8bf7\u8d70\u5b89\u88c5\u6d41\u7a0b":50,"\u6682\u4e0d\u8003\u8651\u5728\u5185":46,"\u6682\u65f6\u4e0d\u8003\u8651\u591a\u4e2aparamet":20,"\u66b4\u9732\u8fd9\u4e2a\u6982\u5ff5\u5fc5\u8981\u51fd\u6570":34,"\u66f4\u522b\u63d0\u7b80\u5316\u95ee\u9898\u590d\u73b0\u5e26\u6765\u7684\u597d\u5904\u4e86":60,"\u66f4\u591a\u5173\u4e8edocker\u7684\u5b89\u88c5\u4e0e\u4f7f\u7528":46,"\u66f4\u591a\u5185\u5bb9\u53ef\u67e5\u770b\u53c2\u8003\u6587\u732e":88,"\u66f4\u591a\u7684\u7ec6\u8282\u53ef\u4ee5\u5728\u6587\u732e\u4e2d\u627e\u5230":88,"\u66f4\u591a\u7684\u7f16\u8bd1\u914d\u7f6e\u89c1":54,"\u66f4\u591a\u7684\u8f6c\u6362\u65b9\u6cd5\u8bf7\u53c2\u8003eigen":64,"\u66f4\u597d\u5730\u5b8c\u6210\u4e00\u4e9b\u590d\u6742\u7684\u8bed\u8a00\u7406\u89e3\u4efb\u52a1":58,"\u66f4\u5feb":59,"\u66f4\u65b0":46,"\u66f4\u65b0\u53ef\u80fd\u5bfc\u81f4\u9700\u8981\u65b0\u7684\u5f00\u53d1\u5de5\u5177":60,"\u66f4\u65b0\u6a21\u5f0f":46,"\u66f4\u65b0\u7684\u6587\u6863\u4ee5pr\u7684\u5f62\u5f0f\u63d0\u4ea4\u5230github\u4e2d":65,"\u66f4\u65b0\u7f51\u7edc\u53c2\u6570\u65f6\u5e94\u7528":46,"\u66f4\u65b9\u4fbf\u7684\u8bbe\u7f6e\u65b9\u5f0f":46,"\u66f4\u8be6\u7ec6\u6570\u636e\u683c\u5f0f\u548c\u7528\u4f8b\u8bf7\u53c2\u8003":84,"\u66f4\u8be6\u7ec6\u7684\u7f51\u7edc\u914d\u7f6e\u8fde\u63a5\u8bf7\u53c2\u8003":84,"\u66f4\u8be6\u7ec6\u7684\u8bf4\u660e":84,"\u66f4\u8fdb\u4e00\u6b65":58,"\u66f4\u9ad8":59,"\u66ff\u6211\u4eec\u5b8c\u6210\u4e86\u539f\u59cb\u8f93\u5165\u6570\u636e\u7684\u62c6\u5206":58,"\u6700":56,"\u6700\u4e0d\u540c\u7684\u7279\u8272\u662f\u5b83\u5e76\u6ca1\u6709\u5c06\u8f93\u5165\u8bed\u53e5\u7f16\u7801\u4e3a\u4e00\u4e2a\u5355\u72ec\u7684\u5b9a\u957f\u5411\u91cf":89,"\u6700\u4e3b\u8981\u7684\u5de5\u4f5c\u5c31\u662f\u89e3\u6790\u51fa":75,"\u6700\u4f73\u63a8\u8350":2,"\u6700\u540e":[2,61,62,68,84,88],"\u6700\u540e\u4e00\u4e2a":55,"\u6700\u540e\u4e00\u5c42cost\u4e2d\u8bb0\u5f55\u4e86\u795e\u7ecf\u7f51\u7edc\u7684\u6240\u6709\u62d3\u6251\u7ed3\u6784":51,"\u6700\u540e\u4e00\u90e8\u5206\u662f\u795e\u7ecf\u7f51\u7edc\u7684\u914d\u7f6e":47,"\u6700\u540e\u518d\u8c03\u7528mutabl":64,"\u6700\u540e\u5220\u9664":42,"\u6700\u540e\u6211\u4eec\u4f7f\u7528\u94fe\u5f0f\u6cd5\u5219\u8ba1\u7b97":62,"\u6700\u597d\u7684\u6a21\u578b\u662f":88,"\u6700\u5c11\u663e\u793a\u591a\u5c11\u4e2a\u8282\u70b9":70,"\u6700\u5e38\u89c1\u7684\u9519\u8bef\u5904\u7406\u65b9\u5f0f\u662fexcept":33,"\u6700\u65b0\u7684\u4ee3\u7801":61,"\u6700\u65b0\u7684paddlepaddl":46,"\u6700\u65b0log":88,"\u6700\u7ec8":62,"\u6700\u7ec8\u5b9e\u73b0\u4e00\u4e2a\u5c42\u6b21\u5316\u7684\u590d\u6742rnn":58,"\u6700\u7ec8\u6211\u4eec\u53ef\u4ee5\u8c03\u7528trainer\u7684train\u65b9\u6cd5\u542f\u52a8\u8bad\u7ec3":51,"\u6700\u7ec8\u751f\u6210\u540d\u4e3apaddl":49,"\u6700\u7ec8\u7684\u8f93\u51fa\u7ed3\u679c":58,"\u6700\u7ec8\u8d8b\u4e8e\u63a5\u8fd1":47,"\u6708\u6e56":56,"\u6709":56,"\u6709\u4e00\u4e2a\u57fa\u672c\u7684\u8ba4\u8bc6":73,"\u6709\u4e00\u4e9b\u5fc5\u987b\u914d\u7f6e\u7684\u53c2\u6570":[53,54],"\u6709\u4e24\u4e2a\u7279\u6b8a\u6807\u8bc6":89,"\u6709\u4e86meta\u914d\u7f6e\u6587\u4ef6\u4e4b\u540e":86,"\u6709\u4e9b\u5c42\u53ef\u80fd\u9700\u8981\u9ad8\u7cbe\u5ea6\u6765\u4fdd\u8bc1\u68af\u5ea6\u68c0\u67e5\u5355\u6d4b\u6b63\u786e\u6267\u884c":62,"\u6709\u4e9b\u5c42\u6216\u8005\u6fc0\u6d3b\u9700\u8981\u505a\u5f52\u4e00\u5316\u4ee5\u4fdd\u8bc1\u5b83\u4eec\u7684\u8f93\u51fa\u7684\u548c\u662f\u4e00\u4e2a\u5e38\u6570":62,"\u6709\u4e9b\u7279\u5f81\u7684\u53d6\u503c\u8fbe\u5230\u6570\u767e\u4e07":46,"\u6709\u4e9b\u7535\u5f71id\u53ef\u80fd\u4e0e\u5b9e\u9645\u7535\u5f71\u4e0d\u76f8\u7b26\u5408":85,"\u6709\u4eba\u7528\u865a\u62df\u673a\u6765\u7c7b\u6bd4":60,"\u6709\u5173":56,"\u6709\u5173\u53c2\u6570\u914d\u7f6e\u7684\u8be6\u7ec6\u8bf4\u660e\u89c1":53,"\u6709\u5173\u7ebf\u6027\u56de\u5f52\u7684\u5b9e\u9645\u5e94\u7528":51,"\u6709\u5173kubernetes\u76f8\u5173\u6982\u5ff5\u4ee5\u53ca\u5982\u4f55\u642d\u5efa\u548c\u914d\u7f6ekubernetes\u96c6\u7fa4":75,"\u6709\u52a9\u4e8e\u7406\u89e3\u7528\u6237\u5bf9\u4e0d\u540c\u516c\u53f8":88,"\u6709\u52a9\u4e8e\u8bca\u65ad\u5206\u5e03\u5f0f\u9519\u8bef":68,"\u6709\u591a\u96be":60,"\u6709\u65f6\u5019\u6211\u4eec\u4f1a\u5e0c\u671b\u6e05\u7406\u6389\u5df2\u7ecf\u4e0b\u8f7d\u7684\u7b2c\u4e09\u65b9\u4f9d\u8d56\u4ee5\u53ca\u5df2\u7ecf\u7f16\u8bd1\u7684\u4e8c\u8fdb\u5236\u6587\u4ef6":60,"\u6709\u65f6\u5019\u6211\u4eec\u53ea\u60f3\u8fd0\u884c\u4e00\u4e2a\u7279\u5b9a\u7684\u5355\u5143\u6d4b\u8bd5":60,"\u6709\u65f6\u79f0\u4e3a":88,"\u6709\u6807\u51c6\u7684":33,"\u6709\u7684\u65f6\u5019":33,"\u6709\u7684\u65f6\u5019\u7b80\u7b80\u5355\u5355\u7684\u6539\u53d8\u5c31\u80fd\u5728\u6027\u80fd\u4e0a\u4ea7\u751f\u660e\u663e\u7684\u4f18\u5316\u6548\u679c":67,"\u6709\u7684\u8bdd\u9700\u8981\u5148\u5378\u8f7d":46,"\u670d\u52a1":56,"\u670d\u52a1\u5458":56,"\u672a\u5305\u542b\u5728\u5b57\u5178\u4e2d\u7684\u5355\u8bcd":89,"\u672a\u6307\u5b9a\u6309\u7167double\u7cbe\u5ea6\u7f16\u8bd1":46,"\u672a\u6807\u8bb0\u7684\u8bc4\u4ef7\u6837\u672c":88,"\u672a\u77e5\u8bcd":80,"\u672c\u4f8b\u4e2d\u4e3a0":80,"\u672c\u4f8b\u4e2d\u4e3a32":80,"\u672c\u4f8b\u4e2d\u4e3a4":80,"\u672c\u4f8b\u4e2d\u4f7f\u7528for\u5faa\u73af\u8fdb\u884c\u591a\u6b21\u8c03\u7528":2,"\u672c\u4f8b\u4e2d\u7684\u539f\u59cb\u6570\u636e\u4e00\u5171\u670910\u4e2a\u6837\u672c":56,"\u672c\u4f8b\u4e2d\u7684\u8f93\u5165\u7279\u5f81\u662f\u8bcdid\u7684\u5e8f\u5217":2,"\u672c\u4f8b\u6839\u636e\u7f51\u7edc\u914d\u7f6e\u4e2d":2,"\u672c\u4f8b\u6bcf\u884c\u4fdd\u5b58\u4e00\u6761\u6837\u672c":84,"\u672c\u4f8b\u7531\u6613\u5230\u96be\u5c55\u793a4\u79cd\u4e0d\u540c\u7684\u6587\u672c\u5206\u7c7b\u7f51\u7edc\u914d\u7f6e":84,"\u672c\u4f8b\u7684":2,"\u672c\u4f8b\u7684\u6240\u6709\u5b57\u7b26\u90fd\u5c06\u8f6c\u6362\u4e3a\u8fde\u7eed\u6574\u6570\u8868\u793a\u7684id\u4f20\u7ed9\u6a21\u578b":84,"\u672c\u4f8b\u91c7\u7528\u82f1\u6587\u60c5\u611f\u5206\u7c7b\u7684\u6570\u636e":2,"\u672c\u4f8b\u91c7\u7528adam\u4f18\u5316\u65b9\u6cd5":84,"\u672c\u5217\u8868\u8bf4\u660epaddlepaddle\u53d1\u7248\u4e4b\u524d\u9700\u8981\u6d4b\u8bd5\u7684\u529f\u80fd\u70b9":42,"\u672c\u5730":46,"\u672c\u5730\u6d4b\u8bd5":69,"\u672c\u5730\u8bad\u7ec3":69,"\u672c\u5730\u8bad\u7ec3\u7684\u5b9e\u9a8c":72,"\u672c\u5b9e\u4f8b\u4e2d":80,"\u672c\u5c0f\u8282\u6211\u4eec\u5c06\u4ecb\u7ecd\u6a21\u578b\u7f51\u7edc\u7ed3\u6784":84,"\u672c\u5c42\u5c3a\u5bf8":82,"\u672c\u5c42\u6709\u56db\u4e2a\u53c2\u6570":82,"\u672c\u6559\u7a0b\u4e2d\u6211\u4eec\u7ed9\u51fa\u4e86\u4e09\u4e2aresnet\u6a21\u578b":82,"\u672c\u6559\u7a0b\u4e3b\u8981\u4ecb\u7ecd\u5e26kernel\u7684op\u5982\u4f55\u5199":63,"\u672c\u6559\u7a0b\u5c06\u4ecb\u7ecd\u4f7f\u7528\u6df1\u5ea6\u53cc\u5411\u957f\u77ed\u671f\u8bb0\u5fc6":87,"\u672c\u6559\u7a0b\u5c06\u6307\u5bfc\u4f60\u5982\u4f55\u5728":59,"\u672c\u6559\u7a0b\u5c06\u6307\u5bfc\u60a8\u5b8c\u6210\u957f\u671f\u77ed\u671f\u8bb0\u5fc6":88,"\u672c\u6559\u7a0b\u63d0\u4f9b\u4e86\u4e00\u4e2a\u7528\u4e8eimagenet\u4e0a\u7684\u5377\u79ef\u5206\u7c7b\u7f51\u7edc\u6a21\u578b":82,"\u672c\u6587\u4e2d\u6240\u6709\u7684\u4f8b\u5b50":56,"\u672c\u6587\u4e2d\u7531\u4e8e\u8f93\u5165\u6570\u636e\u662f\u968f\u673a\u751f\u6210\u7684\u4e0d\u9700\u8981\u8bfb\u8f93\u5165\u6587\u4ef6":47,"\u672c\u6587\u4e2d\u7684\u4f8b\u5b50\u91cc":60,"\u672c\u6587\u4e2d\u793a\u4f8b\u6240\u4f7f\u7528\u7684\u5355\u5143\u6d4b\u8bd5\u6587\u4ef6\u662f":56,"\u672c\u6587\u4ee5paddlepaddle\u7684\u53cc\u5c42rnn\u5355\u5143\u6d4b\u8bd5\u4e3a\u793a\u4f8b":56,"\u672c\u6587\u53ea\u4f7f\u7528\u4e86\u9ed8\u8ba4\u547d\u540d\u7a7a\u95f4":73,"\u672c\u6587\u5c06\u4ecb\u7ecd\u5728kubernetes\u5bb9\u5668\u7ba1\u7406\u5e73\u53f0\u4e0a\u5feb\u901f\u6784\u5efapaddlepaddle\u5bb9\u5668\u96c6\u7fa4":75,"\u672c\u6587\u6863\u4ecb\u7ecd\u5982\u4f55\u5728paddlepaddle\u5e73\u53f0\u4e0a":80,"\u672c\u6587\u6863\u5185\u4e0d\u91cd\u590d\u4ecb\u7ecd":73,"\u672c\u6587\u6863\u5c06\u4ee5linux":53,"\u672c\u6587\u6863\u63cf\u8ff0paddl":34,"\u672c\u6765":56,"\u672c\u6b21\u8bad\u7ec3\u6587\u4ef6\u6240\u5728\u76ee\u5f55":75,"\u672c\u6b21\u8bad\u7ec3\u7684yaml\u6587\u4ef6\u53ef\u4ee5\u5199\u6210":75,"\u672c\u6b21\u8bad\u7ec3\u8981\u6c42\u67093\u4e2apaddlepaddle\u8282\u70b9":75,"\u672c\u6b21\u8bd5\u9a8c":84,"\u672c\u793a\u4f8b\u4e2d\u4f7f\u7528\u7684\u539f\u59cb\u6570\u636e\u5982\u4e0b":56,"\u672c\u793a\u4f8b\u610f\u56fe\u4f7f\u7528\u5355\u5c42rnn\u548c\u53cc\u5c42rnn\u5b9e\u73b0\u4e24\u4e2a\u5b8c\u5168\u7b49\u4ef7\u7684\u5168\u8fde\u63a5rnn":56,"\u672c\u793a\u4f8b\u7684\u9884\u6d4b\u7ed3\u679c":88,"\u672c\u7bc7\u6559\u7a0b\u5728paddlepaddle\u4e2d\u91cd\u73b0\u4e86\u8fd9\u4e00\u826f\u597d\u7684\u8bad\u7ec3\u7ed3\u679c":89,"\u672c\u7bc7\u6559\u7a0b\u5c06\u4f1a\u6307\u5bfc\u4f60\u901a\u8fc7\u8bad\u7ec3\u4e00\u4e2a":89,"\u672c\u8d28\u4e0a\u4e0e\u8bad\u7ec3\u6a21\u578b\u4e00\u6837":89,"\u673a\u5668\u7684\u8bbe\u5907":72,"\u673a\u5668\u7ffb\u8bd1":[42,83],"\u6743\u91cd\u66f4\u65b0\u7684\u68af\u5ea6":70,"\u6761\u4ef6\u4e0b":73,"\u6765":56,"\u6765\u4e0b\u8f7ddocker\u955c\u50cf":49,"\u6765\u4ee3\u66ff":61,"\u6765\u4f7f\u7528\u4e00\u4e2a\u955c\u50cf\u542f\u52a8\u4e00\u4e2a\u5bb9\u5668":49,"\u6765\u4f7f\u7528dropout":46,"\u6765\u4f7f\u7528dropout\u7684":46,"\u6765\u4fdd\u8bc1\u8bad\u7ec3\u8fc7\u7a0b\u53ef\u4ee5\u4ece\u4e2d\u95f4\u72b6\u6001\u91cd\u65b0\u542f\u52a8":20,"\u6765\u505a\u68af\u5ea6\u68c0\u67e5":62,"\u6765\u505ableu\u8bc4\u4f30":89,"\u6765\u505c\u6b62\u8bad\u7ec3":86,"\u6765\u5206\u6790\u6267\u884c\u6587\u4ef6":67,"\u6765\u5206\u79bb\u8bad\u7ec3\u548c\u6d4b\u8bd5\u6587\u4ef6":86,"\u6765\u5206\u9694\u6bcf\u4e00\u884c":86,"\u6765\u5217\u51fa\u5f53\u524d\u7cfb\u7edf\u4e2d\u7684\u6240\u6709\u955c\u50cf":49,"\u6765\u521d\u59cb\u5316\u53c2\u6570":46,"\u6765\u542f\u52a8\u548c":60,"\u6765\u5b89\u88c5":68,"\u6765\u5b8c\u6210\u7f51\u7edc\u7684\u8bad\u7ec3":51,"\u6765\u5b9a\u4e49\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":59,"\u6765\u5bf9\u6bd4\u5206\u6790\u4e24\u8005\u8bed\u4e49\u76f8\u540c\u7684\u539f\u56e0":56,"\u6765\u5e2e\u52a9\u4f60\u7406\u89e3paddlepaddle\u7684\u5185\u90e8\u8fd0\u884c\u673a\u5236":84,"\u6765\u5f00\u542f\u672c\u5730\u7684\u8bad\u7ec3":88,"\u6765\u5f71\u54cdpaddlepaddle\u7684\u7f16\u8bd1\u8fc7\u7a0b":[53,54],"\u6765\u5f97\u5230\u67d0\u4e2a\u7279\u5b9a\u53c2\u6570\u7684\u68af\u5ea6\u77e9\u9635":62,"\u6765\u6267\u884c":60,"\u6765\u6307\u5b9a\u7f51\u7edc\u5c42\u7684\u6570\u76ee":82,"\u6765\u63a5\u53d7\u4e0d\u4f7f\u7528\u7684\u51fd\u6570\u4ee5\u4fdd\u8bc1\u517c\u5bb9\u6027":2,"\u6765\u63cf\u8ff0\u7684":64,"\u6765\u63cf\u8ff0\u8be5op\u7684\u8f93\u5165":63,"\u6765\u642d\u5efa\u795e\u7ecf\u7f51\u7edc":51,"\u6765\u6ce8\u518c\u8be5\u5c42":62,"\u6765\u6df7\u5408\u4f7f\u7528gpu\u548ccpu\u8ba1\u7b97\u7f51\u7edc\u5c42\u7684\u53c2\u6570":72,"\u6765\u6e05\u7406\u8fd9\u4e9b\u5185\u5bb9":60,"\u6765\u751f\u6210\u5e8f\u5217":89,"\u6765\u7684\u79d2\u6570":85,"\u6765\u786e\u4fdd\u628a":33,"\u6765\u786e\u5b9a\u5bf9\u5e94\u5173\u7cfb":2,"\u6765\u7f16\u8bd1":60,"\u6765\u81ea\u5b9a\u4e49\u4f20\u6570\u636e\u7684\u8fc7\u7a0b":1,"\u6765\u83b7\u5f97\u8f93\u51fa\u7684\u68af\u5ea6":62,"\u6765\u8868\u793a":59,"\u6765\u8868\u793a\u53c2\u6570\u4f4d\u7f6e":87,"\u6765\u8868\u793a\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":59,"\u6765\u8868\u793apaddle\u5185\u90e8\u7c7b":33,"\u6765\u89e3\u51b3\u4e0a\u9762\u7684\u95ee\u9898":46,"\u6765\u8ba1\u7b97\u68af\u5ea6":62,"\u6765\u8bb2\u89e3\u5982\u4f55\u4f7f\u7528\u53cc\u5c42rnn":56,"\u6765\u8bbe\u7f6e":46,"\u6765\u8bbf\u95ee\u7528\u6237\u81ea\u5df1\u7684\u6570\u636e":21,"\u6765\u8bf4\u660epydataprovider2\u7684\u7b80\u5355\u4f7f\u7528\u573a\u666f":2,"\u6765\u8c03\u6574":61,"\u6765\u8c03\u7528":60,"\u6765\u8fd0\u884c":68,"\u6765\u8fd0\u884c\u5305\u62ec":60,"\u6765\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5\u4e86":60,"\u6765\u8fd0\u884c\u6027\u80fd\u5206\u6790\u548c\u8c03\u4f18":67,"\u6765\u8fd0\u884c\u955c\u50cf":49,"\u6765\u8fdb\u884c\u8ba8\u8bba":34,"\u6765\u9884\u6d4b\u8fd9\u4e2a\u4e2d\u95f4\u7684\u8bcd":46,"\u676f\u5b50":56,"\u6784\u5efa":53,"\u6784\u6210\u4e86\u8f93\u51fa\u53cc\u5c42\u5e8f\u5217\u7684\u7b2ci\u4e2a":55,"\u6784\u9020":75,"\u6784\u9020\u51fd\u6570\u542b\u67092\u4e2a\u53c2\u6570":63,"\u6784\u9020\u51fd\u6570\u91cc\u901a\u8fc7":63,"\u6784\u9020paddl":4,"\u67b6\u6784\u6765\u8bad\u7ec3\u60c5\u611f\u5206\u6790\u6a21\u578b":88,"\u67d0\u4e00\u4e2a\u795e\u7ecf\u5143\u7684\u4e00\u4e2a\u8f93\u5165\u4e3a\u4e0a\u4e00\u4e2a\u65f6\u95f4\u6b65\u7f51\u7edc\u4e2d\u67d0\u4e00\u4e2a\u795e\u7ecf\u5143\u7684\u8f93\u51fa":56,"\u67d0\u4e9b\u53c2\u6570\u53ea\u53ef\u7528\u4e8e\u7279\u5b9a\u7684\u5c42\u4e2d":69,"\u67e5\u770b":[61,84],"\u67e5\u770b\u5f53\u524d\u72b6\u6001":61,"\u67e5\u770b\u5f53\u524d\u8fdc\u7a0b\u4ed3\u5e93\u7684\u540d\u5b57":61,"\u67e5\u770b\u6587\u4ef6\u5177\u4f53\u88ab\u4fee\u6539\u7684\u5185\u5bb9":61,"\u67e5\u770b\u662f\u5426\u662f\u5176\u4ed6\u9519\u8bef\u5f15\u53d1\u7684\u62a5\u9519":46,"\u67e5\u770bjob\u7684\u8be6\u7ec6\u60c5\u51b5":74,"\u6807\u51c6\u5dee\u4e3a":46,"\u6807\u51c6\u8868\u793apaddlepaddle\u7248\u672c\u53f7":42,"\u6807\u51c6lstm\u4ee5\u6b63\u5411\u5904\u7406\u8be5\u5e8f\u5217":87,"\u6807\u793a\u56fe\u7247\u662f\u5f69\u8272\u56fe\u6216\u7070\u5ea6\u56fe":81,"\u6807\u793a\u662f\u5426\u4e3a\u5f69\u8272\u56fe\u7247":81,"\u6807\u7b7e0\u8868\u793a\u8d1f\u9762\u7684\u8bc4\u8bba":88,"\u6807\u7b7e1\u8868\u793a\u6b63\u9762\u7684\u8bc4\u8bba":88,"\u6807\u7b7e\u4e0b\u627e\u5230\u6700\u65b0\u7684paddle\u955c\u50cf\u7248\u672c":49,"\u6807\u7b7e\u6587\u4ef6":87,"\u6807\u7b7e\u65b9\u6848\u6765\u81ea":87,"\u6807\u8bc6\u662f\u5426\u4e3a\u8fde\u7eed\u7684batch\u8ba1\u7b97":70,"\u6838\u4e00\u6837\u591a\u7684\u8fdb\u7a0b\u6765\u5e76\u884c\u7f16\u8bd1":60,"\u6839\u636e\u4f60\u7684\u4efb\u52a1":72,"\u6839\u636e\u524d\u6587\u7684\u63cf\u8ff0":75,"\u6839\u636e\u5728\u6a21\u578b\u914d\u7f6e\u6587\u4ef6\u4e2d\u4f7f\u7528\u7684\u540d\u4e3a":68,"\u6839\u636e\u6570\u636e\u91cf\u89c4\u6a21":85,"\u6839\u636e\u7528\u6237\u6307\u5b9a\u7684\u5b57\u5178":80,"\u6839\u636e\u7f51\u7edc\u914d\u7f6e\u4e2d\u7684":70,"\u6839\u636e\u8f93\u5165tensor\u7684\u5927\u5c0f\u6765\u8bbe\u7f6e\u8f93\u51fatensor\u7684\u5927\u5c0f":64,"\u6839\u636e\u8fd9\u4e9b\u53c2\u6570\u7684\u4f7f\u7528\u573a\u5408":69,"\u6839\u636e\u9ed8\u8ba4\u503c\u9012\u589e":70,"\u6839\u636e\u9ed8\u8ba4\u7aef\u53e3\u53f7\u9012\u589e":70,"\u6839\u636ecpu":49,"\u6839\u636ejob\u5bf9\u5e94\u7684pod\u4fe1\u606f":74,"\u683c\u5f0f":70,"\u683c\u5f0f\u5982\u4e0b":84,"\u683c\u5f0f\u7684\u6587\u4ef6\u6765\u5b58\u653e":63,"\u683c\u5f0f\u8bf4\u660e":80,"\u68af\u5ea6\u4f1a\u5c31\u5730":62,"\u68af\u5ea6\u53c2\u6570\u7684\u5206\u5757\u6570\u76ee":70,"\u68af\u5ea6\u5c31\u53ef\u4ee5\u901a\u8fc7\u8fd9\u4e2a\u65b9\u7a0b\u8ba1\u7b97\u5f97\u5230":62,"\u68af\u5ea6\u670d\u52a1\u5668\u7684\u6570\u91cf":70,"\u68af\u5ea6\u68c0\u67e5\u5355\u5143\u6d4b\u8bd5\u901a\u8fc7\u6709\u9650\u5dee\u5206\u6cd5\u6765\u9a8c\u8bc1\u4e00\u4e2a\u5c42\u7684\u68af\u5ea6":62,"\u68af\u5ea6\u68c0\u67e5\u7684\u8f93\u5165\u6570\u636e\u7684\u6279\u6b21\u5927\u5c0f":62,"\u68c0\u67e5\u70b9\u4fdd\u5b58\u7a0b\u5e8f\u6d41\u7a0b":20,"\u68c0\u67e5\u8f93\u5165\u6570\u636e\u7ef4\u5ea6":63,"\u68d2":84,"\u697c\u5c42":56,"\u6a21\u5757":81,"\u6a21\u5757\u4e0b\u7684\u76f8\u5173":64,"\u6a21\u5757\u4e2d\u7684":2,"\u6a21\u578b":87,"\u6a21\u578b\u4e00\u76f4\u4e0d\u6536\u655b":46,"\u6a21\u578b\u4fdd\u5b58\u5728\u76ee\u5f55":88,"\u6a21\u578b\u5171\u5305\u542b1":80,"\u6a21\u578b\u5217\u8868\u6587\u4ef6":87,"\u6a21\u578b\u53c2\u6570\u68c0\u67e5\u70b9\u901a\u8fc7\u5b9a\u671f\u5411\u78c1\u76d8\u4e0a\u4fdd\u5b58\u4e00\u4efd\u5b58\u50a8\u5728paramet":20,"\u6a21\u578b\u53ca\u53c2\u6570\u4f1a\u88ab\u4fdd\u5b58\u5728\u8def\u5f84":81,"\u6a21\u578b\u5b58\u50a8\u8def\u5f84":84,"\u6a21\u578b\u5c06\u4fdd\u5b58\u5728\u76ee\u5f55":87,"\u6a21\u578b\u5c31\u8bad\u7ec3\u6210\u529f\u4e86":89,"\u6a21\u578b\u6570\u636e\u68c0\u67e5\u70b9\u7684\u5b9e\u73b0":20,"\u6a21\u578b\u6587\u4ef6\u5c06\u88ab\u5199\u5165\u8282\u70b9":68,"\u6a21\u578b\u6765\u5c06\u6cd5\u8bed\u7ffb\u8bd1\u6210\u82f1\u8bed":89,"\u6a21\u578b\u6765\u6307\u5bfc\u4f60\u5b8c\u6210\u8fd9\u4e9b\u6b65\u9aa4":59,"\u6a21\u578b\u6f14\u793a\u5982\u4f55\u914d\u7f6e\u590d\u6742\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u6a21\u578b":59,"\u6a21\u578b\u7684\u4ee3\u7801\u53ef\u4ee5\u5728":59,"\u6a21\u578b\u7684\u7ed3\u6784\u548c\u8bad\u7ec3\u8fc7\u7a0b":80,"\u6a21\u578b\u7684\u7f16\u7801\u5668\u90e8\u5206\u5982\u4e0b\u6240\u793a":59,"\u6a21\u578b\u88ab\u4fdd\u5b58\u5728":86,"\u6a21\u578b\u8bad\u7ec3\u4f1a\u770b\u5230\u7c7b\u4f3c\u4e0a\u9762\u8fd9\u6837\u7684\u65e5\u5fd7\u4fe1\u606f":84,"\u6a21\u578b\u8bad\u7ec3\u548c\u6700\u540e\u7684\u7ed3\u679c\u9a8c\u8bc1":47,"\u6a21\u578b\u8bad\u7ec3\u7b49\u4efb\u52a1":51,"\u6a21\u578b\u8def\u5f84":[82,87],"\u6a21\u578b\u8f93\u51fa\u8def\u5f84":87,"\u6a21\u578b\u914d\u7f6e":0,"\u6a21\u578b\u914d\u7f6e\u89e3\u6790":33,"\u6a21\u578b\u91c7\u7528":80,"\u6a21\u578b\u9884\u6d4b":4,"\u6a2a\u5411\u62fc\u63a5":46,"\u6b21":56,"\u6b22\u8fce\u901a\u8fc7":61,"\u6b63\u6837\u672c":84,"\u6b63\u786e\u7684\u89e3\u51b3\u65b9\u6cd5\u662f":46,"\u6b63\u8d1f\u5bf9\u9a8c\u8bc1":69,"\u6b63\u9762\u7684\u8bc4\u8bba\u7684\u5f97\u5927\u4e8e\u7b49\u4e8e7":88,"\u6b63\u9762\u8bc4\u4ef7\u6837\u672c":88,"\u6b64\u547d\u4ee4\u5c06\u5728your":53,"\u6b64\u5904":80,"\u6b64\u5904\u90fd\u4e3a2":56,"\u6b64\u5916":[46,53,60,61],"\u6b64\u6559\u7a0b\u5c06\u5411\u60a8\u5206\u6b65\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528\u5185\u7f6e\u7684\u5b9a\u65f6\u5de5\u5177":67,"\u6b64\u6570\u636e\u96c6\u5305\u542b\u7535\u5f71\u8bc4\u8bba\u53ca\u5176\u76f8\u5173\u8054\u7684\u7c7b\u522b\u6807\u7b7e":88,"\u6b64\u65f6\u53ea\u9700\u8981":60,"\u6b64\u65f6\u53ef\u4ee5\u5728\u8c03\u7528infer\u63a5\u53e3\u65f6\u901a\u8fc7\u8bbe\u7f6e":46,"\u6b64\u65f6\u53ef\u4ee5\u8df3\u8fc7paddlepaddle\u6a21\u578b\u53c2\u6570\u6587\u4ef6\u7684\u5934\u4fe1\u606f":46,"\u6b64\u65f6\u60a8\u53ef\u4ee5\u8fd0\u884c\u8fd9\u4e2a\u547d\u4ee4\u5728\u5f00\u53d1\u673a\u4e0a\u8fdb\u884c\u6d4b\u8bd5\u8fd0\u884c":49,"\u6b64\u65f6master\u5c06\u8d1f\u8d23\u542f\u52a8\u4e00\u4e2a\u65b0\u7684train":20,"\u6b64\u7c7b\u62a5\u9519\u901a\u5e38\u662f\u7531\u4e8e\u67d0\u4e00\u4e2a\u8282\u70b9\u7684\u9519\u8bef\u5bfc\u81f4\u8fd9\u4e2a\u8282\u70b9\u7684\u8bad\u7ec3\u8fdb\u7a0b\u9000\u51fa":46,"\u6b65\u9aa4":46,"\u6bb5\u843d\u53ef\u4ee5\u770b\u4f5c\u662f\u4e00\u4e2a\u5d4c\u5957\u7684\u53cc\u5c42\u7684\u5e8f\u5217":58,"\u6bcf100\u4e2abatch\u6253\u5370\u4e00\u6b21\u7edf\u8ba1\u4fe1\u606f":88,"\u6bcf100\u4e2abatch\u663e\u793a\u53c2\u6570\u7edf\u8ba1":87,"\u6bcf20\u4e2abatch\u6253\u5370\u4e00\u6b21\u65e5\u5fd7":88,"\u6bcf20\u4e2abatch\u8f93\u51fa\u65e5\u5fd7":87,"\u6bcf\u4e00\u4e2a":42,"\u6bcf\u4e00\u4e2a\u4efb\u52a1\u6d41\u7a0b\u90fd\u53ef\u4ee5\u88ab\u5212\u5206\u4e3a\u5982\u4e0b\u4e94\u4e2a\u6b65\u9aa4":84,"\u6bcf\u4e00\u4e2a\u6587\u4ef6\u662f\u6570\u636e\u96c6\u7684\u4e00\u4e2ashard":21,"\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65":56,"\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u4e4b\u95f4\u7684\u795e\u7ecf\u7f51\u7edc\u5177\u6709\u4e00\u5b9a\u7684\u76f8\u5173\u6027":56,"\u6bcf\u4e00\u4e2a\u6d4b\u8bd5\u5468\u671f\u6d4b\u8bd5\u4e00\u6b21\u6240\u6709\u6570\u636e":86,"\u6bcf\u4e00\u4e2a\u8282\u70b9\u90fd\u6709\u76f8\u540c\u7684\u65e5\u5fd7\u7ed3\u6784":68,"\u6bcf\u4e00\u4e2akey\u7531":86,"\u6bcf\u4e00\u4e2alayer\u8f93\u51fa\u77e9\u9635\u7684\u9ad8\u5ea6":46,"\u6bcf\u4e00\u7ec4\u5185\u7684\u6240\u6709\u53e5\u5b50\u548clabel":56,"\u6bcf\u4e00\u884c\u8868\u793a\u4e00\u4e2a\u5b9e\u4f8b":88,"\u6bcf\u4e2a":[68,87],"\u6bcf\u4e2a\u503c\u7684\u7c7b\u578b\u53ef\u4ee5\u662f\u6574\u5f62":21,"\u6bcf\u4e2a\u5143\u7d20\u662f\u4e00\u4e2a0\u5c42\u5e8f\u5217":55,"\u6bcf\u4e2a\u5143\u7d20\u662f\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":55,"\u6bcf\u4e2a\u5355\u5c42rnn":58,"\u6bcf\u4e2a\u5355\u8bcd\u7684\u9884\u6d4b\u9519\u8bef\u7387":89,"\u6bcf\u4e2a\u53e5\u5b50\u53c8\u662f\u5355\u8bcd\u7684\u6570\u7ec4":56,"\u6bcf\u4e2a\u53e5\u5b50\u90fd\u4ee5\u5f00\u59cb\u6807\u8bb0\u5f00\u5934":59,"\u6bcf\u4e2a\u53e5\u5b50\u90fd\u4ee5\u7ed3\u675f\u6807\u8bb0\u7ed3\u5c3e":59,"\u6bcf\u4e2a\u5b50\u5e8f\u5217\u957f\u5ea6\u53ef\u4ee5\u4e0d\u4e00\u81f4":56,"\u6bcf\u4e2a\u5b50\u6587\u4ef6\u5939\u4e0b\u5b58\u50a8\u76f8\u5e94\u5206\u7c7b\u7684\u56fe\u7247":81,"\u6bcf\u4e2a\u5b57\u5178\u5305\u542b\u603b\u517130000\u4e2a\u5355\u8bcd":89,"\u6bcf\u4e2a\u5b57\u5178\u90fd\u6709dictsize\u4e2a\u5355\u8bcd":89,"\u6bcf\u4e2a\u5c42\u5728\u5176":62,"\u6bcf\u4e2a\u5c42\u90fd\u6709\u4e00\u4e2a\u6216\u591a\u4e2ainput":84,"\u6bcf\u4e2a\u6279\u6b21\u6570\u636e":70,"\u6bcf\u4e2a\u6587\u4ef6\u5939\u90fd\u5305\u542b\u6cd5\u8bed\u5230\u82f1\u8bed\u7684\u5e73\u884c\u8bed\u6599\u5e93":89,"\u6bcf\u4e2a\u6587\u4ef6\u662f\u4e00\u4e2a\u7535\u5f71\u8bc4\u8bba":88,"\u6bcf\u4e2a\u6587\u672c\u6587\u4ef6\u5305\u542b\u4e00\u4e2a\u6216\u8005\u591a\u4e2a\u5b9e\u4f8b":88,"\u6bcf\u4e2a\u65f6\u95f4\u6b65\u4e4b\u5185\u7684\u8fd0\u7b97\u662f\u72ec\u7acb\u7684":58,"\u6bcf\u4e2a\u65f6\u95f4\u6b65\u90fd\u7528\u4e86\u4e0a\u4e00\u4e2a\u65f6\u95f4\u6b65\u7684\u8f93\u51fa\u7ed3\u679c":56,"\u6bcf\u4e2a\u6743\u91cd\u5bf9\u5e94\u4e00\u4e2a\u8f93\u5165":62,"\u6bcf\u4e2a\u6837\u672c\u7531\u4e24\u90e8\u5206\u7ec4\u6210":56,"\u6bcf\u4e2a\u6837\u672c\u95f4\u7528\u7a7a\u884c\u5206\u5f00":56,"\u6bcf\u4e2a\u6d4b\u8bd5\u5468\u671f\u6d4b\u8bd5":86,"\u6bcf\u4e2a\u7279\u5f81\u7684meta\u914d\u7f6e":86,"\u6bcf\u4e2a\u72b6\u6001":58,"\u6bcf\u4e2a\u7c7b\u522b\u4e2d\u968f\u673a\u62bd\u53d6\u4e8610\u5f20\u56fe\u7247":81,"\u6bcf\u4e2a\u7c7b\u5305\u542b6000\u5f20":81,"\u6bcf\u4e2a\u7ebf\u7a0b":70,"\u6bcf\u4e2a\u7ebf\u7a0b\u5206\u914d\u5230128\u4e2a\u6837\u672c\u7528\u4e8e\u8bad\u7ec3":70,"\u6bcf\u4e2a\u8282\u70b9\u6709\u4e24\u4e2a6\u6838cpu":89,"\u6bcf\u4e2a\u8bad\u7ec3\u8282\u70b9\u5fc5\u987b\u6307\u5b9a\u4e00\u4e2a\u552f\u4e00\u7684id\u53f7":70,"\u6bcf\u4e2a\u8bb0\u5fc6\u5355\u5143\u5305\u542b\u56db\u4e2a\u4e3b\u8981\u7684\u5143\u7d20":88,"\u6bcf\u4e2a\u8bc4\u8bba\u7684\u7f51\u5740":88,"\u6bcf\u4e2a\u8f93\u5165\u90fd\u662f\u4e00\u4e2a":62,"\u6bcf\u4e2a\u8f93\u51fa\u8282\u70b9\u90fd\u8fde\u63a5\u5230\u6240\u6709\u7684\u8f93\u5165\u8282\u70b9\u4e0a":62,"\u6bcf\u4e2a\u91cc\u9762\u90fd\u5305\u542b202mb\u7684\u5168\u90e8\u7684\u6a21\u578b\u53c2\u6570":89,"\u6bcf\u4e2acommit\u53ea\u505a\u4e86\u5c11\u91cf\u7684\u4fee\u6539":61,"\u6bcf\u4e2adata":21,"\u6bcf\u4e2aparamet":20,"\u6bcf\u4e2apass\u7684\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747\u5206\u7c7b\u9519\u8bef\u7387":84,"\u6bcf\u4e2apass\u7684\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747cost":84,"\u6bcf\u4e2apod\u5305\u542b\u4e00\u4e2apaddlepaddle\u5bb9\u5668":75,"\u6bcf\u4e2ashard\u5206\u522b\u5b58\u50a8\u5728\u5176\u4e2d\u4e00\u53f0paramet":20,"\u6bcf\u4f4d\u7528\u6237\u81f3\u5c11\u670920\u6761\u8bc4\u5206":85,"\u6bcf\u5c42\u4e0a\u53ea\u80fd\u4fdd\u5b58\u56fa\u5b9a\u6570\u76ee\u4e2a\u6700\u597d\u7684\u72b6\u6001":70,"\u6bcf\u5c42\u4f7f\u7528\u7684gpu\u53f7\u4f9d\u8d56\u4e8e\u53c2\u6570train":72,"\u6bcf\u5f53\u6a21\u578b\u5728\u7ffb\u8bd1\u8fc7\u7a0b\u4e2d\u751f\u6210\u4e86\u4e00\u4e2a\u5355\u8bcd":89,"\u6bcf\u6279\u6b21":70,"\u6bcf\u6b21":49,"\u6bcf\u6b21\u63d0\u4ea4\u4ee3\u7801":61,"\u6bcf\u6b21\u63d0\u4ea4\u65f6":61,"\u6bcf\u6b21\u6d4b\u8bd5\u90fd\u6d4b\u8bd5\u6240\u6709\u6570\u636e":88,"\u6bcf\u6b21\u751f\u62101\u4e2a\u5e8f\u5217":89,"\u6bcf\u6b21\u8bfb\u53d6\u4e00\u6761\u6570\u636e\u540e":84,"\u6bcf\u6b21\u8f93\u51fa\u4e00\u4e2adata":21,"\u6bcf\u6b21\u90fd\u4f1a\u4ecepython\u7aef\u8bfb\u53d6\u6570\u636e":2,"\u6bcf\u884c\u5b58\u50a8\u4e00\u4e2a\u8bcd":80,"\u6bcf\u884c\u5b58\u50a8\u7684\u662f\u4e00\u4e2a\u6837\u672c\u7684\u7279\u5f81":82,"\u6bcf\u884c\u6253\u537032\u4e2a\u53c2\u6570\u4ee5":80,"\u6bcf\u884c\u8868\u793a\u4e00\u4e2a\u6279\u6b21\u4e2d\u7684\u5355\u4e2a\u8f93\u5165":62,"\u6bcf\u884c\u90fd\u662f\u4e00\u4e2a\u6cd5\u8bed\u6216\u8005\u82f1\u8bed\u7684\u53e5\u5b50":89,"\u6bcf\u8f6e\u4f1a\u5c06\u6570\u636e\u96c6\u4e2d\u7684\u6240\u6709\u8bad\u7ec3\u6837\u672c\u4f7f\u7528\u4e00\u6b21":70,"\u6bcf\u8f6e\u7ed3\u675f\u65f6\u5bf9\u6240\u6709\u6d4b\u8bd5\u6570\u636e\u8fdb\u884c\u6d4b\u8bd5":70,"\u6bcf\u8f6e\u90fd\u4f1a\u4fdd\u5b58\u9884\u6d4b\u7ed3\u679c":70,"\u6bcf\u8fd0\u884c\u591a\u5c11\u4e2a\u6279\u6b21\u6267\u884c\u4e00\u6b21\u7a00\u758f\u53c2\u6570\u5206\u5e03\u7684\u68c0\u67e5":70,"\u6bcf\u969410\u5206\u949f":20,"\u6bcf\u9694\u591a\u5c11batch\u6253\u5370\u4e00\u6b21\u65e5\u5fd7":84,"\u6bcfdot":70,"\u6bcflog":70,"\u6bcfsave":70,"\u6bcftest":70,"\u6bd4\u5982":[21,46,49,53,60,61,84],"\u6bd4\u5982\u4e00\u53e5\u8bdd\u4e2d\u7684\u6bcf\u4e00\u4e2a\u5355\u8bcd":56,"\u6bd4\u5982\u5982\u679c\u8981build\u4e00\u4e2a\u4e0d\u4f9d\u8d56gpu":61,"\u6bd4\u5982\u5c06":42,"\u6bd4\u5982\u5e0c\u671b\u6700\u5c0f\u5316\u751f\u6210\u7684\u5e93\u7684\u5927\u5c0f":[53,54],"\u6bd4\u5982\u6bcf\u969410\u5206\u949f\u6700\u65b0\u7684\u5feb\u7167":20,"\u6bd4\u5982\u6d41\u5f0f\u6570\u636e\u5904\u7406":21,"\u6bd4\u5982\u8bbe\u7f6e\u4e00\u4e2a\u5168\u8fde\u63a5\u5c42\u7684\u53c2\u6570\u521d\u59cb\u5316\u65b9\u5f0f\u548cbias\u521d\u59cb\u5316\u65b9\u5f0f":46,"\u6bd4\u5982\u901a\u8fc78080\u7aef\u53e3":73,"\u6bd4\u5982fpe":46,"\u6bd4\u5982ide\u914d\u7f6e\u91cc":61,"\u6bd4\u5982imagenet\u8fd9\u4e2a\u6570\u636e\u96c6\u53ef\u80fd\u88ab\u5206\u62101000\u4e2ashard":21,"\u6c34\u6e29":56,"\u6c49\u5ead":56,"\u6c60\u5316\u5c42":81,"\u6ca1":56,"\u6ca1\u6709\u4f5c\u7528":2,"\u6ca1\u6709\u5b9e\u9645\u610f\u4e49":80,"\u6ca1\u6709\u627e\u5230\u548c\u5f53\u524d\u7cfb\u7edf\u5339\u914d\u7684paddlepaddle\u5b89\u88c5\u5305":46,"\u6ca1\u6709\u6d4b\u8bd5\u6570\u636e":2,"\u6ca1\u6709\u8bbe\u7f6e":[53,54],"\u6ca1\u6709\u8fdb\u884c\u6b63\u786e\u6027\u7684\u68c0\u67e5":85,"\u6ca1\u6709\u8fdb\u884c\u7ed3\u6784\u7684\u5fae\u8c03":86,"\u6cd5\u8bed":89,"\u6ce8":[20,49],"\u6ce8\u518c":63,"\u6ce8\u518ccpu":63,"\u6ce8\u518cop":63,"\u6ce8\u518cop\u65f6\u7684\u7c7b\u578b\u540d":63,"\u6ce8\u610f":[2,48,49,51,53,54,59,62,75,81],"\u6ce8\u610f\u4e0a\u8ff0\u547d\u4ee4\u4e2d":75,"\u6ce8\u610f\u5230\u6211\u4eec\u5df2\u7ecf\u5047\u8bbe\u673a\u5668\u4e0a\u67094\u4e2agpu":72,"\u6ce8\u610f\u5e94\u8be5\u786e\u4fdd\u9ed8\u8ba4\u6a21\u578b\u8def\u5f84":88,"\u6ce8\u610f\u9884\u6d4b\u6570\u636e\u901a\u5e38\u4e0d\u5305\u542blabel":4,"\u6ce8\u610fnode":75,"\u6ce8\u91ca":63,"\u6ce8\u91ca\u6389":88,"\u6cf3\u6c60":56,"\u6d41":56,"\u6d41\u7a0b\u6765\u63d0\u4ea4\u4ee3\u7801":61,"\u6d44":56,"\u6d4b\u8bd5":61,"\u6d4b\u8bd5\u524d\u5411oper":63,"\u6d4b\u8bd5\u6570\u636e":68,"\u6d4b\u8bd5\u6570\u636e\u4e5f\u5305\u542b":68,"\u6d4b\u8bd5\u6570\u636e\u548c\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":68,"\u6d4b\u8bd5\u6570\u636e\u548c\u751f\u6210\u6570\u636e":89,"\u6d4b\u8bd5\u6570\u636e\u653e\u7f6e\u5728\u5de5\u4f5c\u7a7a\u95f4\u4e2d\u4e0d\u540c\u76ee\u5f55\u7684\u8981\u6c42":68,"\u6d4b\u8bd5\u6570\u636e\u7684\u6240\u6709\u76f8\u5bf9\u6216\u7edd\u5bf9\u6587\u4ef6\u8def\u5f84":68,"\u6d4b\u8bd5\u6570\u6910\u96c6":88,"\u6d4b\u8bd5\u65f6\u6307\u5b9a\u7684\u5b58\u50a8\u6a21\u578b\u5217\u8868\u7684\u6587\u4ef6":70,"\u6d4b\u8bd5\u65f6\u9ed8\u8ba4\u4e0dshuffl":2,"\u6d4b\u8bd5\u662f":61,"\u6d4b\u8bd5\u6837\u672c":68,"\u6d4b\u8bd5\u6a21\u578b\u662f\u6307\u4f7f\u7528\u8bad\u7ec3\u51fa\u7684\u6a21\u578b\u8bc4\u4f30\u5df2\u6807\u8bb0\u7684\u9a8c\u8bc1\u96c6":88,"\u6d4b\u8bd5\u7684\u6a21\u578b\u5305\u62ec\u4ece\u7b2cm\u8f6e\u5230\u7b2cn":72,"\u6d4b\u8bd5\u811a\u672c\u662f":87,"\u6d4b\u8bd5\u96c6\u548c\u8bad\u7ec3\u96c6\u76ee\u5f55\u5305\u542b\u4e0b\u9762\u7684\u6587\u4ef6":88,"\u6d4b\u8bd5docker\u955c\u50cf":42,"\u6d4b\u8bd5model_list":69,"\u6d4b\u8bd5save_dir":69,"\u6d4f\u89c8\u4ee3\u7801":49,"\u6d6a\u6f2b\u7247":85,"\u6d6e\u70b9\u578b\u6570\u636e":21,"\u6d6e\u70b9\u5f02\u5e38\u901a\u5e38\u7684\u539f\u56e0\u662f\u6d6e\u70b9\u6570\u6ea2\u51fa":46,"\u6d6e\u70b9\u6570\u5360\u7528\u7684\u5b57\u8282\u6570":80,"\u6d6e\u70b9\u7a00\u758f\u6570\u636e":62,"\u6dd8\u5b9d\u7b49":88,"\u6df1\u5ea6\u53cc\u5411lstm\u5c42\u63d0\u53d6softmax\u5c42\u7684\u7279\u5f81":87,"\u6df1\u5ea6\u5b66\u4e60\u5165\u95e8\u8bfe\u7a0b":52,"\u6df7\u5408":87,"\u6df7\u5408\u5f53\u524d\u8bcd\u5411\u91cf\u548cattention\u52a0\u6743\u7f16\u7801\u5411\u91cf":59,"\u6dfb\u52a0\u4e86\u4e00\u4e2a\u8f93\u51fa":63,"\u6dfb\u52a0\u542f\u52a8\u811a\u672c":75,"\u6dfb\u52a0\u8f93\u5165\u53c2\u6570":63,"\u6dfb\u52a0\u8f93\u51fa\u53c2\u6570":63,"\u6dfb\u52a0op\u7684\u6ce8\u91ca":63,"\u6e05\u7406":60,"\u6e05\u7406\u6389\u8001\u65e7\u7684paddlepaddle\u5b89\u88c5\u5305":46,"\u6e29\u99a8":56,"\u6e90":89,"\u6e90\u4ee3\u7801":[49,84],"\u6e90\u4ee3\u7801\u683c\u5f0f":61,"\u6e90\u5b57\u5178":89,"\u6e90\u5e8f\u5217":59,"\u6e90\u7801":49,"\u6e90\u7801\u4e0edemo":74,"\u6e90\u7801\u6811\u6839\u76ee\u5f55":60,"\u6e90\u7801\u7f16\u8bd1paddlepaddl":54,"\u6e90\u8bed\u8a00\u5230\u76ee\u6807\u8bed\u8a00\u7684\u5e73\u884c\u8bed\u6599\u5e93\u6587\u4ef6":89,"\u6e90\u8bed\u8a00\u548c\u76ee\u6807\u8bed\u8a00\u5171\u4eab\u76f8\u540c\u7684\u7f16\u7801\u5b57\u5178":80,"\u6e90\u8bed\u8a00\u548c\u76ee\u6807\u8bed\u8a00\u90fd\u662f\u76f8\u540c\u7684\u8bed\u8a00":80,"\u6e90\u8bed\u8a00\u77ed\u8bed\u548c\u76ee\u6807\u8bed\u8a00\u77ed\u8bed\u7684\u5b57\u5178\u5c06\u88ab\u5408\u5e76":80,"\u6ee4\u6ce2\u5668\u6838\u5728\u5782\u76f4\u65b9\u5411\u4e0a\u7684\u5c3a\u5bf8":82,"\u6ee4\u6ce2\u5668\u6838\u5728\u6c34\u5e73\u65b9\u5411\u4e0a\u7684\u5c3a\u5bf8":82,"\u6f14\u793a\u4e2d\u4f7f\u7528\u7684":87,"\u6f14\u793a\u91c7\u7528":87,"\u6fc0\u6d3b":62,"\u6fc0\u6d3b\u51fd\u6570\u7c7b\u578b":84,"\u6fc0\u6d3b\u65b9\u7a0b":62,"\u6fc0\u6d3b\u7684\u7c7b\u578b":62,"\u70ed\u60c5":56,"\u7136\u540e":[67,68,80,86],"\u7136\u540e\u4ea4\u7ed9\u7528\u6237\u81ea\u5b9a\u4e49\u7684\u51fd\u6570":47,"\u7136\u540e\u4ea4\u7ed9step\u51fd\u6570":58,"\u7136\u540e\u4f60\u53ea\u9700\u8981\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4":89,"\u7136\u540e\u4f7f\u7528\u4e0b\u9762\u7684\u811a\u672c":88,"\u7136\u540e\u4f7f\u7528resize\u63a5\u53e3\u8bbe\u7f6etensor\u7684\u5927\u5c0f":64,"\u7136\u540e\u518d\u505a\u4e00\u6b21\u6587\u672c\u5377\u79ef\u7f51\u7edc\u64cd\u4f5c":86,"\u7136\u540e\u5229\u7528\u89c2\u6d4b\u6570\u636e\u8c03\u6574":47,"\u7136\u540e\u5355\u51fb":61,"\u7136\u540e\u53ef\u4ee5\u4f7f\u7528\u547d\u4ee4\u884c\u5de5\u5177\u521b\u5efajob":75,"\u7136\u540e\u53ef\u4ee5\u8f6c\u6362\u4e3a\u56fe\u7247":82,"\u7136\u540e\u5728":89,"\u7136\u540e\u5728\u4e0b\u4e00\u4e2a\u65f6\u95f4\u6b65\u8f93\u5165\u7ed9\u53e6\u4e00\u4e2a\u795e\u7ecf\u5143":56,"\u7136\u540e\u5728\u6d4f\u89c8\u5668\u4e2d\u8f93\u5165\u4ee5\u4e0b\u7f51\u5740":49,"\u7136\u540e\u5728\u89e3\u7801\u88ab\u7ffb\u8bd1\u7684\u8bed\u53e5\u65f6":89,"\u7136\u540e\u5728dataprovider\u91cc\u9762\u6839\u636e\u8be5\u5730\u5740\u52a0\u8f7d\u5b57\u5178":46,"\u7136\u540e\u5728etcd\u7684":20,"\u7136\u540e\u5b89\u88c5paddle\u7684python\u73af\u5883":46,"\u7136\u540e\u5b9a\u4e49":59,"\u7136\u540e\u5c06\u6784\u5efa\u6210\u529f\u7684\u955c\u50cf\u4e0a\u4f20\u5230\u955c\u50cf\u4ed3\u5e93":75,"\u7136\u540e\u5c06\u8fd9\u4e9blayer\u7684\u53c2\u6570":46,"\u7136\u540e\u5c31\u53ef\u4ee5\u5e76\u53d1\u5199\u5165\u591a\u4e2achunk":28,"\u7136\u540e\u5f97\u5230\u5e73\u5747\u91c7\u6837\u7684\u7ed3\u679c":86,"\u7136\u540e\u6211\u4eec\u5229\u7528\u591a\u8f93\u5165\u7684":86,"\u7136\u540e\u6211\u4eec\u53d1\u73b0pass":89,"\u7136\u540e\u6211\u4eec\u5806\u53e0\u4e00\u5bf9\u5bf9\u7684":87,"\u7136\u540e\u6211\u4eec\u6c42\u8fd9\u4e24\u4e2a\u7279\u5f81\u7684\u4f59\u5f26\u76f8\u4f3c\u5ea6":86,"\u7136\u540e\u6240\u6709\u7528":61,"\u7136\u540e\u624d\u80fd\u4f7f\u7528pfsclient":28,"\u7136\u540e\u6267\u884c\u4e0b\u9762\u7684\u547d\u4ee4":82,"\u7136\u540e\u628a\u8fd9\u4e2a\u5305\u542b\u4e86\u8bad\u7ec3\u6570\u636e\u7684container\u4fdd\u5b58\u4e3a\u4e00\u4e2a\u65b0\u7684\u955c\u50cf":74,"\u7136\u540e\u63d0\u4ea4\u65b0\u6dfb\u52a0\u7684":61,"\u7136\u540e\u63d0\u53d6\u9690\u85cflstm\u5c42\u7684\u6240\u6709\u65f6\u95f4\u6b65\u957f\u7684\u6700\u5927\u8bcd\u5411\u91cf\u4f5c\u4e3a\u6574\u4e2a\u5e8f\u5217\u7684\u8868\u793a":88,"\u7136\u540e\u662f\u5bf9\u5e94\u7684\u82f1\u8bed\u5e8f\u5217":89,"\u7136\u540e\u6dfb\u52a0\u4e00\u4e2a\u5168\u8fde\u63a5\u5c42":86,"\u7136\u540e\u70b9\u51fb":61,"\u7136\u540e\u7528\u5bc6\u7801":49,"\u7136\u540e\u7528pickle\u547d\u4ee4\u5c06\u7279\u5f81":86,"\u7136\u540e\u7533\u660e\u4e00\u4e2a\u5b58\u50a8\u5377":75,"\u7136\u540e\u89c2\u5bdf\u5230\u8f93\u51fa\u7684\u53d8\u5316\u4e3a":62,"\u7136\u540e\u89e3\u538b":89,"\u7136\u540e\u89e3\u7801\u5668\u901a\u8fc7\u8fd9\u4e2a\u5411\u91cf\u751f\u6210\u4e00\u4e2a\u76ee\u6807\u8bed\u53e5":89,"\u7136\u540e\u8f93\u51fa\u9884\u6d4b\u5206\u6570":86,"\u7136\u540e\u8fd4\u56de\u7ed9paddlepaddle\u8fdb\u7a0b":2,"\u7136\u540e\u8fdb\u884c\u968f\u673a\u6253\u4e71":86,"\u7136\u540e\u901a\u8fc7\u51fd\u6570":75,"\u7136\u540e\u901a\u8fc7\u81ea\u8eab\u7684ip\u5730\u5740\u5728":75,"\u7136\u540e\u91cd\u65b0cmake\u5373\u53ef":46,"\u7136\u800c":[59,70],"\u7136\u800c\u6709\u4e9b\u8bc4\u8bba\u4e0a\u4e0b\u6587\u975e\u5e38\u957f":88,"\u7248\u672c":60,"\u7248\u672c\u5206\u652f":42,"\u7248\u672c\u53f7":42,"\u7248\u672c\u53f7rc":42,"\u7248\u672c\u5728":61,"\u7248\u672cfork\u51fa\u81ea\u5df1\u7684\u529f\u80fd\u5206\u652f":42,"\u7279\u522b\u662f\u5728lstm\u7b49rnn\u4e2d":46,"\u7279\u522b\u662f\u5f53\u76f8\u540c\u7684\u8bcd\u5728\u53e5\u5b50\u4e2d\u51fa\u73b0\u591a\u4e8e\u4e00\u6b21\u65f6":87,"\u7279\u5f81":86,"\u7279\u5f81\u56fe\u5747\u503c":82,"\u7279\u5f81\u56fe\u65b9\u5dee":82,"\u7279\u5f81\u5c06\u4f1a\u5b58\u5230":82,"\u7279\u5f81\u6587\u4ef6":87,"\u7279\u5f81\u7684\u7c7b\u578b\u548c\u7ef4\u5ea6":86,"\u72af\u7f6a\u7247":85,"\u72ec\u7acb\u5de5\u5177\u94fe":53,"\u72ec\u7acb\u5de5\u5177\u94fe\u6240\u5728\u7684\u7edd\u5bf9\u8def\u5f84":53,"\u73af\u5883\u53d8\u91cf":75,"\u73af\u5883\u53d8\u91cf\u6765\u6307\u5b9a\u7279\u5b9a\u7684gpu":46,"\u73b0\u5728\u6211\u4eec\u53ef\u4ee5\u5f00\u59cbpaddle\u8bad\u7ec3\u4e86":86,"\u73b0\u9636\u6bb5paddle\u6709\u4e00\u4e2a\u95ee\u9898\u662f":33,"\u7406\u89e3":60,"\u751a\u81f3\u4e0d\u540c\u7ade\u4e89\u5bf9\u624b\u4ea7\u54c1\u7684\u504f\u597d":88,"\u751a\u81f3\u80fd\u89e3\u91ca\u4e3a\u4ec0\u4e48\u67d0\u4e2a\u64cd\u4f5c\u82b1\u4e86\u5f88\u957f\u65f6\u95f4":67,"\u751f\u4ea7\u73af\u5883\u4e2d\u7684\u8bad\u7ec3\u6570\u636e\u96c6\u901a\u5e38\u4f53\u79ef\u5f88\u5927":21,"\u751f\u4ea7\u73af\u5883\u7684\u65e5\u5fd7\u6570\u636e\u4f1a\u901a\u8fc7\u5b9e\u65f6\u6d41\u7684\u65b9\u5f0f":21,"\u751f\u4ea7\u955c\u50cf":[49,61],"\u751f\u4ea7\u955c\u50cf\u7684\u751f\u6210\u5206\u4e3a\u4e24\u6b65":49,"\u751f\u6210":75,"\u751f\u6210\u5404\u79cd\u8bed\u8a00\u7684\u7ed1\u5b9a\u4ee3\u7801":33,"\u751f\u6210\u540e\u7684\u6587\u6863\u5206\u522b\u5b58\u50a8\u5728\u7f16\u8bd1\u76ee\u5f55\u7684":65,"\u751f\u6210\u5e8f\u5217\u7684\u6700\u5927\u957f\u5ea6":59,"\u751f\u6210\u5f53\u524d\u5c42\u7684\u6240\u6709\u540e\u7ee7\u72b6\u6001":89,"\u751f\u6210\u6570\u636e\u7684\u76ee\u5f55":89,"\u751f\u6210\u6587\u6863":33,"\u751f\u6210\u751f\u4ea7\u955c\u50cf\u53ea\u9700\u8981\u8fd0\u884c":49,"\u751f\u6210\u7684":21,"\u751f\u6210\u7684\u6570\u636e\u5c06\u4f1a\u5b58\u50a8\u5728\u8fd9\u4e2avolume\u4e0b":75,"\u751f\u6210\u7684\u6570\u636e\u7f13\u5b58\u5728\u5185\u5b58\u91cc":46,"\u751f\u6210\u7684\u7ed3\u679c\u6587\u4ef6":89,"\u751f\u6210\u7684html\u7248\u672c\u7684c":49,"\u751f\u6210\u7684meta\u914d\u7f6e\u6587\u4ef6\u5982\u4e0b\u6240\u793a":86,"\u751f\u6210\u7ed9\u5b9a":21,"\u751f\u6210\u7f51\u7edc\u5c42\u914d\u7f6e":62,"\u751f\u6210\u81ea\u5df1\u76ee\u5f55\u4e0b\u7684\u4ed3\u5e93":61,"\u751f\u6210\u8bad\u7ec3\u9700\u8981\u7684\u6837\u672c":86,"\u751f\u6210\u8fd0\u884c\u7a0b\u5e8f":49,"\u751f\u6210\u968f\u673a\u7684\u8f93\u5165\u6570\u636e":63,"\u751f\u6210api\u6587\u6863":33,"\u751f\u6210docker\u955c\u50cf\u7684\u65b9\u5f0f\u6709\u4e24\u4e2a":49,"\u751f\u6210pfsclient\u548cpfsserver\u7684\u6846\u67b6\u90e8\u5206":28,"\u7528":[28,85,86,87],"\u75280\u548c1\u8868\u793a":2,"\u7528\u4e86\u4e24\u4e2a\u6708\u4e4b\u540e\u8fd9\u4e2a\u663e\u793a\u5668\u5c4f\u5e55\u788e\u4e86":84,"\u7528\u4e8e":68,"\u7528\u4e8e\u5207\u5206\u5355\u5355\u8bcd\u548c\u6807\u70b9\u7b26\u53f7":88,"\u7528\u4e8e\u521d\u59cb\u5316\u53c2\u6570\u548c\u8bbe\u7f6e":62,"\u7528\u4e8e\u5c06\u4e0b\u4e00\u884c\u7684\u6570\u636e\u8f93\u5165\u51fd\u6570\u6807\u8bb0\u6210\u4e00\u4e2apydataprovider2":2,"\u7528\u4e8e\u5c06\u53c2\u6570\u4f20\u9012\u7ed9\u7f51\u7edc\u914d\u7f6e":72,"\u7528\u4e8e\u6307\u5b9a\u5176\u8981\u5173\u8054\u7684layer":46,"\u7528\u4e8e\u6307\u5b9a\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6":70,"\u7528\u4e8e\u653e\u7f6e":68,"\u7528\u4e8e\u6784\u6210\u65b0\u7684\u8bcd\u8868":80,"\u7528\u4e8e\u6ce8\u518c\u6ca1\u6709\u53cd\u5411\u7684op":63,"\u7528\u4e8e\u7a00\u758f\u8bad\u7ec3\u4e2d":70,"\u7528\u4e8e\u7edf\u8ba1\u8bcd\u9891\u7684bow\u6a21\u578b\u7279\u5f81":88,"\u7528\u4e8e\u81ea\u5b9a\u4e49\u6bcf\u6761\u6570\u636e\u7684batch":2,"\u7528\u4e8e\u83b7\u53d6\u7279\u5b9alayer\u4e0a\u4e00\u65f6\u95f4\u6b65\u7684\u8f93\u51fa":46,"\u7528\u4e8e\u8ba1\u7b97\u7f16\u7801\u5411\u91cf\u7684\u52a0\u6743\u548c":59,"\u7528\u4e8e\u8bbe\u7f6e\u8bad\u7ec3\u7b97\u6cd5":81,"\u7528\u4e8e\u8bfb\u53d6\u8bad\u7ec3":68,"\u7528\u4e8e\u96c6\u7fa4\u901a\u4fe1\u901a\u9053\u7684\u7aef\u53e3\u6570":68,"\u7528\u53cc\u5411\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7f16\u7801":59,"\u7528\u591a\u5bf9\u6548\u679c\u5b8c\u5168\u76f8\u540c\u7684":56,"\u7528\u6237":68,"\u7528\u62371\u7684\u7279\u5f81":86,"\u7528\u6237\u4e0a\u4f20\u6570\u636e\u540e":21,"\u7528\u6237\u4e5f\u53ef\u4ee5\u4e0a\u4f20label":21,"\u7528\u6237\u4e5f\u53ef\u4ee5\u5728\u81ea\u5df1\u719f\u6089\u7684\u5f00\u53d1\u5e73\u53f0\u4e0a":54,"\u7528\u6237\u4e5f\u53ef\u4ee5\u5728c":1,"\u7528\u6237\u53ea\u9700\u5b9a\u4e49rnn\u5728\u4e00\u4e2a\u65f6\u95f4\u6b65\u5185\u5b8c\u6210\u7684\u8ba1\u7b97":58,"\u7528\u6237\u53ea\u9700\u6267\u884c":87,"\u7528\u6237\u53ea\u9700\u6267\u884c\u4ee5\u4e0b\u547d\u4ee4\u5c31\u53ef\u4ee5\u4e0b\u8f7d\u5e76\u5904\u7406\u539f\u59cb\u6570\u636e":87,"\u7528\u6237\u53ef\u4ee5\u4f7f\u7528\u5f00\u53d1\u955c\u50cf\u4ee3\u66ff\u914d\u7f6e\u672c\u5730\u73af\u5883":49,"\u7528\u6237\u53ef\u4ee5\u4f7f\u7528ssh\u767b\u5f55\u5230\u8fd9\u53f0\u670d\u52a1\u5668\u4e0a\u5e76\u6267\u884c":49,"\u7528\u6237\u53ef\u4ee5\u5206\u522b\u67e5\u770b\u6700\u65b0\u7684":65,"\u7528\u6237\u53ef\u4ee5\u53c2\u8003sphinx\u6559\u7a0b\u8fdb\u884c\u4e66\u5199":65,"\u7528\u6237\u53ef\u4ee5\u5728\u8f93\u51fa\u7684\u6587\u672c\u6a21\u578b\u4e2d\u770b\u5230":80,"\u7528\u6237\u53ef\u4ee5\u5b89\u5168\u7684\u91ca\u653e\u67d0\u4e2ac":34,"\u7528\u6237\u53ef\u4ee5\u628a\u81ea\u5df1\u7684\u6570\u636e\u5206\u4eab\u7ed9\u522b\u4eba":21,"\u7528\u6237\u53ef\u4ee5\u6839\u636e\u8bad\u7ec3\u65e5\u5fd7":84,"\u7528\u6237\u53ef\u4ee5\u76f4\u63a5\u4f7f\u7528\u8fd9\u4e2a\u52a8\u6001\u5e93\u6765\u5f15\u5165paddl":34,"\u7528\u6237\u53ef\u4ee5\u81ea\u5b9a\u4e49beam":70,"\u7528\u6237\u53ef\u4ee5\u8bbe\u7f6e":72,"\u7528\u6237\u53ef\u4ee5\u901a\u8fc7\u7b80\u5355\u4f7f\u7528python\u63a5\u53e3":1,"\u7528\u6237\u53ef\u4ee5\u901a\u8fc7\u7f51\u9875\u6d4f\u89c8\u6587\u6863":49,"\u7528\u6237\u53ef\u5728\u8c03\u7528cmake\u7684\u65f6\u5019\u8bbe\u7f6e\u5b83\u4eec":48,"\u7528\u6237\u53ef\u5728cmake\u7684\u547d\u4ee4\u884c\u4e2d":48,"\u7528\u6237\u53ef\u5c06":53,"\u7528\u6237\u53ef\u6839\u636e\u81ea\u5df1\u7684\u7f16\u8bd1\u76ee\u6807\u67b6\u6784":53,"\u7528\u6237\u53ef\u81ea\u884c\u524d\u5f80":54,"\u7528\u6237\u53ef\u81ea\u884c\u524d\u5f80\u4e0b\u8f7d\u9884\u7f16\u8bd1\u597d\u7684\u7248\u672c":53,"\u7528\u6237\u53ef\u901a\u8fc7\u4ea4\u53c9\u7f16\u8bd1\u7684\u65b9\u5f0f":53,"\u7528\u6237\u53ef\u901a\u8fc7ssh\u7b49\u65b9\u5f0f\u767b\u5f55\u5230raspberri":54,"\u7528\u6237\u5728\u4f7f\u7528\u8fd9\u4e00\u7c7brecurr":46,"\u7528\u6237\u5728\u4f7f\u7528paddlepaddl":46,"\u7528\u6237\u5728\u672c\u5730\u8f6c\u6362\u597d\u518d\u4e0a\u4f20":21,"\u7528\u6237\u5b9a\u4e49\u7684\u53c2\u6570":2,"\u7528\u6237\u5c06\u53c2\u6570\u8f7d\u5165":46,"\u7528\u6237\u5c06\u914d\u7f6e\u4e0e\u8bad\u7ec3\u6570\u636e\u5207\u5206\u597d\u653e\u5728\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u9884\u5148\u5206\u914d\u597d\u7684\u76ee\u5f55\u4e2d":75,"\u7528\u6237\u5e94\u8be5\u63d0\u4f9b\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6":87,"\u7528\u6237\u5f3a\u5236\u6307\u5b9a\u7279\u5b9a\u7684python\u7248\u672c":46,"\u7528\u6237\u6307\u5b9a\u65b0\u7684\u5b57\u5178\u7684\u8def\u5f84":80,"\u7528\u6237\u6587\u4ef6\u4e2d\u6709\u56db\u79cd\u7c7b\u578b\u7684\u5b57\u6bb5":86,"\u7528\u6237\u6587\u4ef6\u53ef\u80fd\u662f\u6bd4\u8f83\u5927\u7684":28,"\u7528\u6237\u7279\u5f81":86,"\u7528\u6237\u8fd8\u53ef\u4ee5\u6839\u636e\u6982\u7387\u5206\u5e03\u77e9\u9635\u5b9e\u73b0\u67f1\u641c\u7d22\u6216\u7ef4\u7279\u6bd4\u89e3\u7801":87,"\u7528\u6237\u8fd8\u53ef\u6839\u636e\u81ea\u5df1\u7684\u9700\u6c42\u8bbe\u7f6e\u5176\u4ed6\u7f16\u8bd1\u53c2\u6570":[53,54],"\u7528\u6237\u901a\u8fc7\u53c2\u6570":46,"\u7528\u6237\u901a\u8fc7c":34,"\u7528\u6237\u9700\u8981\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u6307\u5b9a":72,"\u7528\u6237\u9700\u8981\u5728cmake\u65f6\u624b\u52a8\u8bbe\u7f6e\u8fd9\u4e9b\u503c":[53,54],"\u7528\u6237\u9700\u8981\u6307\u5b9a\u672c\u673a\u4e0apython\u7684\u8def\u5f84":46,"\u7528\u6237\u9700\u8981\u63d0\u524d\u51c6\u5907\u597d\u4ea4\u53c9\u7f16\u8bd1\u73af\u5883":[53,54],"\u7528\u6237\u9884\u6d4b\u7684\u547d\u4ee4\u884c\u754c\u9762\u5982\u4e0b":86,"\u7528\u6237id":85,"\u7528\u6237id\u8303\u56f4\u4ece1\u52306040":85,"\u7528\u672c\u673a\u7684\u7b2c\u4e00\u4e2a":60,"\u7528\u6765\u4ece\u53c2\u6570\u670d\u52a1\u5668\u9884\u53d6\u53c2\u6570\u77e9\u9635\u76f8\u5e94\u7684\u884c":62,"\u7528\u6765\u4f30\u8ba1\u7ebf\u6027\u51fd\u6570\u7684\u53c2\u6570w":47,"\u7528\u6765\u5177\u4f53\u63cf\u8ff0":86,"\u7528\u6765\u5177\u4f53\u8bf4\u660e\u6570\u636e\u96c6\u7684\u5b57\u6bb5\u548c\u6587\u4ef6\u683c\u5f0f":86,"\u7528\u6765\u8ba1\u7b97\u6a21\u578b\u7684\u8bef\u5dee":47,"\u7528\u8fd9\u4e2a\u955c\u50cf\u521b\u5efa\u7684\u5bb9\u5668\u9700\u8981\u6709\u4ee5\u4e0b\u4e24\u4e2a\u529f\u80fd":75,"\u7531":[46,58],"\u7531\u4e8e\u4e0a\u4e00\u6b65cmake\u914d\u7f6e\u4e2d":54,"\u7531\u4e8e\u4e0d\u540c\u7684paddle\u7684\u7248\u672c\u53ef\u80fd\u9700\u8981\u4e0d\u540c\u7684\u4f9d\u8d56\u548c\u5de5\u5177":49,"\u7531\u4e8e\u5b83\u5185\u90e8\u5305\u542b\u4e86\u6bcf\u7ec4\u6570\u636e\u4e2d\u7684\u6240\u6709\u53e5\u5b50":56,"\u7531\u4e8e\u5bb9\u5668\u4e4b\u95f4\u5171\u4eabnet":73,"\u7531\u4e8e\u5bf9parameters\u7684\u66f4\u65b0\u9700\u8981\u83b7\u53d6parameters\u5185\u5b58\u7684":20,"\u7531\u4e8e\u5df2\u7ecf\u77e5\u9053\u4e86\u771f\u5b9e\u7b54\u6848":47,"\u7531\u4e8e\u610f\u5916\u7684\u526f\u672c\u8bb0\u5f55\u548c\u6d4b\u8bd5\u8bb0\u5f55":85,"\u7531\u4e8e\u6211\u4eec\u60f3\u8981\u7684\u53d8\u6362\u662f\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217":56,"\u7531\u4e8e\u6211\u4eec\u652f\u6301\u8bad\u7ec3\u6570\u636e\u6709\u4e0d\u540c\u7684\u6279\u6b21\u5927\u5c0f":62,"\u7531\u4e8e\u6570\u636e\u8bb8\u53ef\u7684\u539f\u56e0":87,"\u7531\u4e8e\u6807\u51c6\u7684\u7ffb\u8bd1\u7ed3\u679c\u5df2\u7ecf\u4e0b\u8f7d\u5230\u8fd9\u91cc":89,"\u7531\u4e8e\u6bcf\u4e2a\u5377\u79ef\u5c42\u540e\u9762\u8fde\u63a5\u7684\u662fbatch":82,"\u7531\u4e8e\u8fd9\u4e2a\u5730\u5740\u4f1a\u88abdataprovider\u4f7f\u7528":1,"\u7531\u4e8e\u8fd9\u6837\u505a\u53ef\u4ee5\u907f\u514d\u5f88\u591a\u6b7b\u9501\u95ee\u9898":2,"\u7531\u4e8e\u96c6\u7fa4\u4e2d\u540c\u65f6\u5b58\u5728\u4e24\u53f0\u673a\u5668\u6545\u969c\u7684\u6982\u7387\u6781\u4f4e":20,"\u7531\u4e8e\u987a\u5e8f\u8c03\u7528\u8fd9\u4e9bgenerator\u4e0d\u4f1a\u51fa\u73b0\u4e0a\u8ff0\u95ee\u9898":2,"\u7531\u4e8ec":33,"\u7531\u4e8echunk\u6bd4\u8f83\u5c0f":28,"\u7531\u4e8eeigen":64,"\u7531\u4e8epaddlepaddle\u5df2\u7ecf\u5b9e\u73b0\u4e86\u4e30\u5bcc\u7684\u7f51\u7edc\u5c42":47,"\u7531\u4e8epypi":42,"\u7531\u4e8estep":58,"\u7531\u4e8etensor\u7684rank\u662f\u6a21\u677f\u53c2\u6570":64,"\u7531\u4e8etest_data\u5305\u542b\u4e24\u6761\u9884\u6d4b\u6570\u636e":4,"\u7531\u8bcd\u8bed\u6784\u6210\u7684\u53e5\u5b50":55,"\u7531grouplen":85,"\u7533\u8bf7\u7528\u6237\u7a7a\u95f4":28,"\u7535\u5f711\u7684\u7279\u5f81":86,"\u7535\u5f71\u4fe1\u606f\u4ee5\u53ca\u7535\u5f71\u8bc4\u5206":85,"\u7535\u5f71\u540d\u5b57\u6bb5":86,"\u7535\u5f71\u540d\u79f0":85,"\u7535\u5f71\u548c\u7528\u6237":86,"\u7535\u5f71\u548c\u7528\u6237\u6709\u8bb8\u591a\u7684\u7279\u5f81":86,"\u7535\u5f71\u5927\u90e8\u5206\u662f\u624b\u5de5\u8f93\u5165\u6570\u636e":85,"\u7535\u5f71\u7279\u5f81":86,"\u7535\u5f71\u7c7b\u578b":85,"\u7535\u5f71\u7c7b\u578b\u5982\u7b26\u5408\u591a\u79cd\u7528\u7ba1\u9053\u7b26\u53f7":85,"\u7535\u5f71id":85,"\u7535\u5f71id\u8303\u56f4\u4ece1\u52303952":85,"\u7535\u8111":56,"\u767e\u4e07\u6570\u636e\u96c6":85,"\u7684":[56,60,61,64,68,74,75,84,88],"\u768410\u7ef4\u6574\u6570\u6807\u7b7e":2,"\u7684\u4e00\u4e2a\u7b80\u5355\u8c03\u7528\u5982\u4e0b":58,"\u7684\u4e00\u4e2a\u7ebf\u6027\u51fd\u6570":47,"\u7684\u4e00\u79cd":89,"\u7684\u4e3a0":70,"\u7684\u4e3b\u8981\u90e8\u5206":87,"\u7684\u4efb\u4e00\u4e00\u79cd":46,"\u7684\u4efb\u52a1":89,"\u7684\u4f5c\u7528\u662f\u5ef6\u8fdf\u5206\u914d\u5185\u5b58":64,"\u7684\u4f7f\u7528\u793a\u4f8b\u5982\u4e0b":55,"\u7684\u503c":[53,54],"\u7684\u503c\u81ea\u52a8\u63a8\u5bfc\u5f97\u5230":53,"\u7684\u504f\u7f6e\u5411\u91cf":62,"\u7684\u5177\u4f53\u8ba1\u7b97\u903b\u8f91":63,"\u7684\u5185\u5b58":46,"\u7684\u5185\u5bb9\u5982\u4e0b\u6240\u793a":89,"\u7684\u5185\u5bb9\u6765\u5b9a\u5236imag":75,"\u7684\u5185\u6838block\u4f7f\u7528\u60c5\u51b5":67,"\u7684\u5206\u7c7b\u4efb\u52a1\u4e2d\u8d62\u5f97\u4e86\u7b2c\u4e00\u540d":82,"\u7684\u522b\u540d":[5,6,8],"\u7684\u5355\u5143\u6d4b\u8bd5":63,"\u7684\u5355\u8bcd\u7ea7\u522b\u7684\u5377\u79ef\u795e\u7ecf\u7f51\u7edc":86,"\u7684\u53cd\u5411\u4f20\u64ad\u5c06\u4f1a\u6253\u5370\u65e5\u5fd7\u4fe1\u606f":70,"\u7684\u53d8\u6362\u77e9\u9635":62,"\u7684\u53e5\u5b50\u7684\u60c5\u611f":88,"\u7684\u540d\u5b57":2,"\u7684\u540d\u79f0\u76f8\u540c":59,"\u7684\u540e\u7f00":85,"\u7684\u5411\u91cf":62,"\u7684\u542f\u52a8\u53c2\u6570":75,"\u7684\u542f\u52a8\u53c2\u6570\u5e76\u6267\u884c\u8fdb\u7a0b":75,"\u7684\u547d\u540d\u98ce\u683c\u5e76\u4e0d\u80fd\u9002\u5e94\u5176\u4ed6\u7b2c\u4e09\u65b9\u8bed\u8a00":33,"\u7684\u5730\u5740":73,"\u7684\u5730\u65b9":61,"\u7684\u5747\u5300\u5206\u5e03":46,"\u7684\u5934\u6587\u4ef6":33,"\u7684\u5b9e\u73b0":63,"\u7684\u5dee\u8ddd\u4e0d\u65ad\u51cf\u5c0f":47,"\u7684\u5e73\u5747\u503c":55,"\u7684\u5e8f\u5217\u5f62\u72b6\u4e00\u81f4":56,"\u7684\u5f00\u53d1\u5de5\u4f5c\u90fd\u5e94\u8be5\u5728\u4e00\u4e2a\u65b0\u7684\u5206\u652f\u4e0a\u5b8c\u6210":61,"\u7684\u5f00\u53d1\u6d41\u7a0b":60,"\u7684\u5f00\u59cb\u8bf7\u52a0\u4e0a\u5b8f\u5b9a\u4e49":63,"\u7684\u603b":68,"\u7684\u63a5\u53e3\u6837\u5f0f":33,"\u7684\u63cf\u8ff0\u8bf4\u660e\u4e2d":61,"\u7684\u64cd\u4f5c":64,"\u7684\u6570\u636e\u6d41\u56fe":21,"\u7684\u6570\u636e\u8bfb\u53d6\u811a\u672c\u548c\u7c7b\u4f3c\u4e8e":84,"\u7684\u6570\u76ee\u4e00\u81f4":55,"\u7684\u6587\u4ef6\u4e5f\u5e26\u5230\u65b0\u5206\u652f\u4e0a":61,"\u7684\u65b9\u5f0f":49,"\u7684\u65b9\u6cd5\u5df2\u88ab\u8bc1\u660e\u662f\u4e00\u79cd\u5f3a\u5927\u7684\u6a21\u578b":89,"\u7684\u65b9\u7a0b":62,"\u7684\u65f6\u95f4\u6b65\u4fe1\u606f\u6210\u6b63\u6bd4":46,"\u7684\u66f4\u8be6\u7ec6\u51c6\u786e\u7684\u5b9a\u4e49":56,"\u7684\u6700\u5c0f\u503c":70,"\u7684\u6700\u65b0\u4ee3\u7801\u5e76\u66f4\u65b0\u5f53\u524d\u5206\u652f":61,"\u7684\u6784\u9020\u51fd\u6570":63,"\u7684\u67b6\u6784\u7684\u793a\u4f8b":59,"\u7684\u6837\u5f0f":61,"\u7684\u6838\u5fc3\u662f\u8bbe\u8ba1step\u51fd\u6570\u7684\u8ba1\u7b97\u903b\u8f91":58,"\u7684\u6bb5\u843d\u5b9a\u4e49\u4e3a\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":58,"\u7684\u6d4b\u8bd5\u6570\u636e\u96c6":87,"\u7684\u6e90\u7801\u4ee5\u53ca\u751f\u6210\u6587\u6863\u9700\u8981\u591a\u79cd\u5f00\u53d1\u5de5\u5177":61,"\u7684\u6e90\u7801\u91cc\u4f7f\u7528\u4e86":33,"\u7684\u7248\u672c":60,"\u7684\u7248\u672c\u53f7":80,"\u7684\u7279\u5f81":82,"\u7684\u72b6\u6001":58,"\u7684\u7528\u6237\u53c2\u8003":68,"\u7684\u76ee\u5f55\u7ed3\u6784\u4e3a":86,"\u7684\u771f\u5b9e\u5173\u7cfb\u4e3a":47,"\u7684\u77e9\u9635":[46,62],"\u7684\u795e\u7ecf\u7f51\u7edc\u673a\u5668\u7ffb\u8bd1":89,"\u7684\u7a20\u5bc6\u5411\u91cf\u4f5c\u4e3a\u8f93\u5165":62,"\u7684\u7aef\u5230\u7aef\u7cfb\u7edf\u6765\u89e3\u51b3srl\u4efb\u52a1":87,"\u7684\u7b2ci\u4e2a\u503c":62,"\u7684\u7b2cj\u4e2a\u503c":62,"\u7684\u7cfb\u7edf":60,"\u7684\u7d22\u5f15\u6587\u4ef6\u5f15\u7528\u8bad\u7ec3":68,"\u7684\u7ed3\u6784\u5982\u4e0b":86,"\u7684\u7ef4\u5ea6":80,"\u7684\u7f29\u5199":28,"\u7684\u884c\u6570\u5e94\u8be5\u4e00\u81f4":89,"\u7684\u89c4\u8303":33,"\u7684\u89d2\u5ea6":21,"\u7684\u8ba1\u7b97\u4ee3\u7801":64,"\u7684\u8ba1\u7b97\u8fc7\u7a0b\u4e66\u5199\u66f4\u52a0\u7b80\u5355":63,"\u7684\u8bad\u7ec3\u6a21\u578b\u811a\u672c":84,"\u7684\u8bdd":46,"\u7684\u8def\u5f84\u4e2d":88,"\u7684\u8f93\u5165":58,"\u7684\u8f93\u51fa":67,"\u7684\u8f93\u51fa\u4fe1\u606f\u5165\u624b\u662f\u4e2a\u4e0d\u9519\u7684\u9009\u62e9":67,"\u7684\u8f93\u51fa\u51fd\u6570\u8fd4\u56de\u7684\u662f\u4e0b\u4e00\u4e2a\u65f6\u523b\u8f93\u51fa\u8bcd\u7684":59,"\u7684\u8f93\u51fa\u683c\u5f0f":56,"\u7684\u8f93\u51fa\u88ab\u7528\u4f5c":59,"\u7684\u8f93\u51fab\u662f\u4e00\u4e2a":46,"\u7684\u8fd0\u884c\u73af\u5883":60,"\u7684\u8fd4\u56de\u503c\u4e00\u81f4":85,"\u7684\u8fdc\u7a0b\u4ed3\u5e93\u7684\u540d\u5b57":61,"\u7684\u90e8\u5206":68,"\u7684\u914d\u7f6e":80,"\u7684\u914d\u7f6e\u5199\u5230\u914d\u7f6e\u6587\u4ef6\u4e2d":21,"\u7684\u96c6\u88c5\u7bb1\u6280\u672f":60,"\u7684\u9875\u9762\u5220\u9664\u8fdc\u7a0b\u4ed3\u5e93\u7684\u5206\u652f":61,"\u7684cpu":63,"\u7684docker\u955c\u50cf":49,"\u76ee\u524d":[58,87],"\u76ee\u524d\u4f7f\u7528":61,"\u76ee\u524d\u53ea\u652f\u6301":53,"\u76ee\u524d\u53ea\u8003\u8651\u52a8\u6001\u6269\u5bb9trainer\u6570\u91cf":20,"\u76ee\u524d\u5d4c\u5165python\u89e3\u91ca\u5668":33,"\u76ee\u524d\u5fc5\u987b\u8bbe\u7f6e\u6210":[53,54],"\u76ee\u524d\u6211\u4eec\u7528cephfs\u6765\u642d\u5efa":28,"\u76ee\u524d\u652f\u6301\u4e24\u79cd":55,"\u76ee\u524d\u652f\u6301fail":70,"\u76ee\u524d\u8be5\u53c2\u6570\u4ec5\u7528\u4e8eaucvalidationlayer\u548cpnpairvalidationlayer\u5c42":70,"\u76ee\u524d\u8fd8\u672a\u652f\u6301":58,"\u76ee\u524dpaddle\u7684\u8fdb\u7a0b\u6a21\u578b\u662fc":33,"\u76ee\u524dpaddlepaddle\u7684develop\u5206\u652f\u7684\u6587\u6863\u662f\u81ea\u52a8\u89e6\u53d1\u66f4\u65b0\u7684":65,"\u76ee\u5f55":[53,54,60,68,74,75,88],"\u76ee\u5f55\u4e0b":[34,49,62,84,89],"\u76ee\u5f55\u4e0b\u627e\u5230":84,"\u76ee\u5f55\u4e0b\u65b0\u589e\u7684":63,"\u76ee\u5f55\u4e0b\u7684\u751f\u6210\u6587\u4ef6\u7528\u4e8e\u6df1\u5ea6\u5b66\u4e60\u76f8\u5173android":53,"\u76ee\u5f55\u4e0b\u7684demo\u8bad\u7ec3\u51fa\u6765":4,"\u76ee\u5f55\u4e0b\u7684python\u5305":46,"\u76ee\u5f55\u4e2d":[68,86],"\u76ee\u5f55\u4e2d\u4f1a\u5305\u542b":[53,54],"\u76ee\u5f55\u4e2d\u7684":[67,68],"\u76ee\u5f55\u4e2dpaddl":75,"\u76ee\u5f55\u4f1a\u51fa\u73b0\u5982\u4e0b\u51e0\u4e2a\u65b0\u7684\u6587\u4ef6":87,"\u76ee\u5f55\u548c":[53,54],"\u76ee\u5f55\u5c31\u6210\u4e3a\u4e86\u5171\u4eab\u5b58\u50a8":75,"\u76ee\u5f55\u7ed3\u6784\u5982\u4e0b":89,"\u76ee\u5f55\u91cc\u63d0\u4f9b\u4e86\u8be5\u6570\u636e\u7684\u4e0b\u8f7d\u811a\u672c\u548c\u9884\u5904\u7406\u811a\u672c":84,"\u76ee\u6807":89,"\u76ee\u6807\u51fd\u6570\u662f\u6807\u7b7e\u7684\u4ea4\u53c9\u71b5":87,"\u76ee\u6807\u5411\u91cf":59,"\u76ee\u6807\u5b57\u5178":89,"\u76ee\u6807\u673a\u7248protobuf\u5e93":[53,54],"\u76ee\u6807\u67b6\u6784abi":53,"\u76f4\u5230\u8bad\u7ec3\u6536\u655b\u4e3a\u6b62":46,"\u76f4\u5230\u903c\u8fd1\u771f\u5b9e\u89e3":47,"\u76f4\u63a5\u4f7f\u7528c\u8bed\u8a00\u7684":33,"\u76f4\u63a5\u5220\u9664\u8fd9\u4e2a\u53c2\u6570\u5373\u53ef":34,"\u76f4\u63a5\u5bfc\u51fa\u5230c\u7684\u63a5\u53e3\u6bd4\u8f83\u56f0\u96be":33,"\u76f4\u63a5\u7f16\u8bd1raspberri":54,"\u76f4\u63a5\u8fd0\u884c":49,"\u76f4\u63a5\u8fd4\u56de\u8ba1\u7b97\u7ed3\u679c":4,"\u76f4\u63a5\u8fdb\u5165\u8bad\u7ec3\u6a21\u578b\u7ae0\u8282":84,"\u76f8\u5173\u6587\u6863":54,"\u76f8\u5173\u6587\u6863\u6240\u8ff0":54,"\u76f8\u5173\u6982\u5ff5\u662f":2,"\u76f8\u5173\u7684\u9e1f\u7c7b\u6570\u636e\u96c6\u53ef\u4ee5\u4ece\u5982\u4e0b\u5730\u5740\u4e0b\u8f7d":81,"\u76f8\u5173\u8bba\u6587":87,"\u76f8\u53cd":89,"\u76f8\u540c\u540d\u5b57\u7684\u53c2\u6570":46,"\u76f8\u5bf9":56,"\u76f8\u5bf9\u4e8epaddlepaddle\u7a0b\u5e8f\u8fd0\u884c\u65f6\u7684\u8def\u5f84":1,"\u76f8\u5bf9mnist\u800c\u8a00":2,"\u76f8\u5e94\u7684\u6570\u636e\u8bfb\u53d6\u811a\u672c\u548c\u8bad\u7ec3\u6a21\u578b\u811a\u672c":84,"\u76f8\u5e94\u7684\u6570\u636e\u8fed\u4ee3\u5668\u5982\u4e0b":87,"\u76f8\u5e94\u7684\u6807\u8bb0\u53e5\u5b50\u662f":87,"\u76f8\u5f53":56,"\u76f8\u6bd4":63,"\u770b\u5f53\u524dmpi\u96c6\u7fa4\u662f\u5426\u652f\u6301resourc":46,"\u77e9\u9635":69,"\u77e9\u9635\u4e58\u6cd5\u7684\u516c\u5f0f":63,"\u7814\u7a76\u4eba\u5458\u5206\u6790\u4e86\u51e0\u4e2a\u5173\u4e8e\u6d88\u8d39\u8005\u4fe1\u5fc3\u548c\u653f\u6cbb\u89c2\u70b9\u7684\u8c03\u67e5":88,"\u7814\u7a76\u751f":85,"\u786e\u4fdd\u7f16\u8bd1\u5668\u9009\u9879":61,"\u78c1\u76d8\u4e0d\u591f":60,"\u78c1\u76d8\u7a7a\u95f4\u4e0d\u8db3\u7b49":46,"\u793a":84,"\u793a\u4f8b":[46,82],"\u793a\u4f8b3\u5bf9\u4e8e\u5355\u5c42rnn\u548c\u53cc\u5c42rnn\u6570\u636e\u5b8c\u5168\u76f8\u540c":56,"\u793a\u4f8b3\u7684\u914d\u7f6e\u4f7f\u7528\u4e86\u5355\u5c42rnn\u548c\u53cc\u5c42rnn":56,"\u793a\u4f8b3\u7684\u914d\u7f6e\u5206\u522b\u4e3a":56,"\u793a\u4f8b\u4ee3\u7801\u5982\u4e0b":46,"\u793a\u4f8b\u5982\u4e0b":46,"\u793e\u533a\u53c2\u4e0e\u56f0\u96be":33,"\u793e\u533a\u8d21\u732e\u4ee3\u7801\u5b66\u4e60\u6210\u672c\u9ad8":33,"\u795e\u7ecf\u7f51\u7edc\u4e2d\u7684\u53c2\u6570":20,"\u795e\u7ecf\u7f51\u7edc\u4e5f\u9700\u8981\u4e00\u4e9b\u7279\u5b9a\u7684layer\u4f5c\u4e3a\u8f93\u5165\u63a5\u53e3":51,"\u795e\u7ecf\u7f51\u7edc\u53c2\u6570\u4ee5\u53ca\u8fed\u4ee3\u65b9\u7a0b":51,"\u795e\u7ecf\u7f51\u7edc\u5728\u8bad\u7ec3\u7684\u65f6\u5019":46,"\u795e\u7ecf\u7f51\u7edc\u672c\u8d28\u4e0a\u662f\u4e00\u4e2a\u8ba1\u7b97\u56fe":64,"\u795e\u7ecf\u7f51\u7edc\u673a\u5668\u7ffb\u8bd1":89,"\u795e\u7ecf\u7f51\u7edc\u7684\u7f51\u7edc\u7ed3\u6784\u4e2d\u5177\u6709\u6709\u5411\u73af\u7ed3\u6784":56,"\u795e\u7ecf\u7f51\u7edc\u7684\u8bad\u7ec3\u672c\u8eab\u662f\u4e00\u4e2a\u975e\u5e38\u6d88\u8017\u5185\u5b58\u548c\u663e\u5b58\u7684\u5de5\u4f5c":46,"\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e":47,"\u79bb":56,"\u79bb\u7ebf\u6279\u5904\u7406":21,"\u79d1\u5b66\u5bb6":85,"\u79d1\u5e7b\u7247":85,"\u79f0\u4e3a":[59,61],"\u79f0\u4e3a\u5f00\u53d1\u955c\u50cf":49,"\u79f0\u4e3a\u751f\u4ea7\u955c\u50cf":49,"\u79f0\u4e4b\u4e3a\u53cc\u5c42\u5e8f\u5217\u7684\u4e00\u4e2a\u5b50\u5e8f\u5217":55,"\u79f0\u4e4b\u4e3a\u96c6\u675f\u5927\u5c0f":70,"\u79f0\u4f5c\u6709kernel":63,"\u79f0\u4f5ckernel":63,"\u7a00\u758f\u6570\u636e\u7684\u683c\u5f0f":62,"\u7a00\u758f\u66f4\u65b0\u7684\u7aef\u53e3\u6570\u91cf":75,"\u7a00\u758f\u768401\u5411\u91cf":[2,51],"\u7a00\u758f\u7684\u5411\u91cf":[2,51],"\u7a00\u758f\u77e9\u9635\u7684\u4e58\u79ef\u5e94\u7528\u4e8e\u524d\u5411\u4f20\u64ad\u8fc7\u7a0b":72,"\u7a0b\u5e8f\u4ece\u6b64\u76ee\u5f55\u62f7\u8d1d\u6587\u4ef6\u5230\u5bb9\u5668\u5185\u8fdb\u884c\u8bad\u7ec3":75,"\u7a0b\u5e8f\u505c\u6b62":70,"\u7a0b\u5e8f\u5458":85,"\u7a0b\u5e8f\u76f4\u63a5\u9000\u51fa":70,"\u7a20\u5bc6\u5411\u91cf":62,"\u7a20\u5bc6\u66f4\u65b0\u7684\u7aef\u53e3\u6570\u91cf":75,"\u7a20\u5bc6\u7684\u6d6e\u70b9\u6570\u5411\u91cf":[2,51],"\u7a97\u6237":56,"\u7acb\u523b\u9000\u51fa":46,"\u7aef\u53e3":[46,68],"\u7aef\u53e3\u6570":68,"\u7aef\u53e3\u9644\u52a0\u5230\u4e3b\u673a\u540d\u4e0a":68,"\u7aef\u81ea\u5b9a\u4e49\u4e00\u4e2a":1,"\u7aef\u8bfb\u53d6\u6570\u636e":46,"\u7b2c":56,"\u7b2c\u4e00\u4e2a":61,"\u7b2c\u4e00\u4e2a\u53c2\u6570":63,"\u7b2c\u4e00\u4e2a\u53c2\u6570\u662fsettings\u5bf9\u8c61":2,"\u7b2c\u4e00\u4e2a\u65b9\u6cd5\u7684\u597d\u5904\u662f\u7b80\u5355\u5feb\u6377":49,"\u7b2c\u4e00\u4e2a\u6837\u672c\u540c\u65f6encode\u4e24\u6761\u6570\u636e\u6210\u4e24\u4e2a\u5411\u91cf":56,"\u7b2c\u4e00\u4e2apass\u4f1a\u4ecepython\u7aef\u8bfb\u53d6\u6570\u636e":2,"\u7b2c\u4e00\u4e2atag\u4e3a":42,"\u7b2c\u4e00\u5929":56,"\u7b2c\u4e00\u6b65\u662f\u8fd0\u884c":49,"\u7b2c\u4e00\u7ae0\u8282":51,"\u7b2c\u4e00\u884c\u4ece":88,"\u7b2c\u4e00\u884c\u5b58\u7684\u662f\u56fe\u50cf":82,"\u7b2c\u4e00\u884c\u662f":80,"\u7b2c\u4e00\u884c\u7684":89,"\u7b2c\u4e00\u90e8\u5206\u5b9a\u4e49\u4e86\u6570\u636e\u8f93\u5165":47,"\u7b2c\u4e00\u90e8\u5206\u662f\u56fe\u7247\u7684\u6807\u7b7e":2,"\u7b2c\u4e09":88,"\u7b2c\u4e09\u4e2a\u53c2\u6570":63,"\u7b2c\u4e09\u5217\u662f\u751f\u6210\u7684\u82f1\u8bed\u5e8f\u5217":89,"\u7b2c\u4e09\u6b65":82,"\u7b2c\u4e09\u6b65\u5b8c\u6210\u540e":42,"\u7b2c\u4e8c":88,"\u7b2c\u4e8c\u4e2a":46,"\u7b2c\u4e8c\u4e2a\u4e3a":42,"\u7b2c\u4e8c\u4e2a\u53c2\u6570":63,"\u7b2c\u4e8c\u4e2a\u65b9\u6cd5\u7684\u597d\u5904\u662fdockerfile\u53ef\u4ee5\u628a\u6574\u4e2a\u751f\u6210\u6d41\u7a0b\u63cf\u8ff0\u5f88\u6e05\u695a":49,"\u7b2c\u4e8c\u5217\u662f\u96c6\u675f\u641c\u7d22\u7684\u5f97\u5206":89,"\u7b2c\u4e8c\u6b65":[80,82],"\u7b2c\u4e8c\u6b65\u662f\u8fd0\u884c":49,"\u7b2c\u4e8c\u7c7b":46,"\u7b2c\u4e8c\u884c\u5b58\u7684\u662f\u56fe\u50cf":82,"\u7b2c\u4e8c\u90e8\u5206\u4e3b\u8981\u662f\u9009\u62e9\u5b66\u4e60\u7b97\u6cd5":47,"\u7b2c\u4e8c\u90e8\u5206\u662f28":2,"\u7b2c\u56db\u4e2a\u53c2\u6570":63,"\u7b2ci\u884c\u7b2cj\u5217\u7684\u6570\u503c":62,"\u7b49":[34,46,63],"\u7b49\u4e8e\u6837\u672c\u6570":46,"\u7b49\u5168\u90e8\u9759\u6001\u5e93\u4e2d\u7684\u76ee\u6807\u6587\u4ef6\u5168\u90e8\u6253\u5305\u540e\u4ea7\u751f\u7684\u6587\u4ef6":34,"\u7b49\u5176\u4ed6":89,"\u7b49\u53c2\u6570":75,"\u7b49\u591a\u79cd\u516c\u6709\u4e91\u73af\u5883":73,"\u7b49\u5f85\u8fd9\u4e2a\u7a0b\u5e8f\u6267\u884c\u6210\u529f\u5e76\u8fd4\u56de0\u5219\u6210\u529f\u9000\u51fa":73,"\u7b49\u6587\u4ef6":34,"\u7b49\u7b49":[84,89],"\u7b49\u90fd\u5c5e\u4e8e\u4e00\u4e2a\u547d\u540d\u7a7a\u95f4":73,"\u7b80\u4ecb":66,"\u7b80\u5199":63,"\u7b80\u5355\u4ecb\u7ecd\u9700\u8981\u7528\u5230\u57fa\u7c7b":63,"\u7b80\u5355\u603b\u7ed3op\u9700\u8981\u5305\u542b\u7684\u5185\u5bb9\u5982\u4e0b":63,"\u7b80\u5355\u6765\u8bf4":67,"\u7b80\u5355\u7684\u5168\u8fde\u63a5\u7f51\u7edc":46,"\u7b80\u5355\u7684\u57fa\u4e8e\u5b57\u6bcd\u7684\u8bcd\u5d4c\u5165":86,"\u7b80\u5355\u7684\u6027\u80fd\u5206\u6790":67,"\u7b80\u5355\u7684\u6574\u4e2a\u8bcd\u5d4c\u5165":86,"\u7b80\u5355\u7684pydataprovider2\u6837\u4f8b\u5c31\u8bf4\u660e\u5b8c\u6bd5\u4e86":2,"\u7b80\u5355\u7684yaml\u6587\u4ef6\u5982\u4e0b":74,"\u7b80\u76f4":56,"\u7b97\u6cd5":[46,47,59,88],"\u7b97\u6cd5\u4e2d\u7684beam\u5927\u5c0f":59,"\u7b97\u6cd5\u914d\u7f6e":88,"\u7ba1\u7406\u4eba\u5458":85,"\u7ba1\u7406\u5458":85,"\u7c7b\u4f3c":[34,55],"\u7c7b\u4f3c\u5730":87,"\u7c7b\u4f5c\u4e3a\u53c2\u6570\u7684\u62bd\u8c61":62,"\u7c7b\u522b\u4e2a\u6570":81,"\u7c7b\u522b\u4e2d\u7684\u53c2\u6570\u53ef\u7528\u4e8e\u6240\u6709\u573a\u5408":69,"\u7c7b\u522bid":84,"\u7c7b\u522bid\u548c\u6587\u672c\u4fe1\u606f\u7528":84,"\u7c7b\u540d\u548cc":33,"\u7c7b\u578b":[33,63,70,86],"\u7c7b\u578b\u4e3a":63,"\u7c7b\u578b\u4ecd\u7136\u4e3aeigenvector":64,"\u7c7b\u578b\u53ef\u4ee5\u662fpaddlepaddle\u652f\u6301\u7684\u4efb\u610f\u8f93\u5165\u6570\u636e\u7c7b\u578b":55,"\u7c7b\u578b\u540d\u4e3a":63,"\u7c7b\u578b\u662fnumpy\u7684ndarrai":46,"\u7c7b\u578b\u662fsparse_binary_vector":[2,51],"\u7c7b\u578b\u662fsparse_float_vector":[2,51],"\u7c7b\u578b\u6765\u8bbe\u7f6e":2,"\u7c7b\u578b\u7684":56,"\u7c7b\u578b\u7b49\u662f\u5426\u5408\u6cd5":63,"\u7c7b\u7684\u5b9a\u4e49\u5199\u5728":63,"\u7c7b\u7684\u6784\u9020\u51fd\u6570\u548c\u6790\u6784\u51fd\u6570":62,"\u7c7b\u91cd\u5199":63,"\u7c7b\u9700\u8981\u5b9e\u73b0\u521d\u59cb\u5316":62,"\u7cfb\u6570":63,"\u7cfb\u7edf\u4f1a\u5bf9\u65b0\u589e\u7684op\u81ea\u52a8\u7ed1\u5b9apython":63,"\u7cfb\u7edf\u4f1a\u6839\u636e\u6587\u4ef6\u540d\u81ea\u52a8\u6784\u5efaop\u548c\u5176\u5bf9\u5e94\u7684python\u6269\u5c55":63,"\u7ea2\u697c\u68a6":80,"\u7eaa\u5f55\u7247":85,"\u7eafcpu\u955c\u50cf\u4ee5\u53cagpu\u955c\u50cf\u90fd\u4f1a\u7528\u5230avx\u6307\u4ee4\u96c6":49,"\u7ebf\u6027\u56de\u5f52\u7684\u8f93\u5165\u662f\u4e00\u6279\u70b9":47,"\u7ebf\u6027\u56de\u5f52\u7684\u8f93\u51fa\u662f\u4ece\u8fd9\u6279\u70b9\u4f30\u8ba1\u51fa\u6765\u7684\u53c2\u6570":47,"\u7ebf\u6027\u8ba1\u7b97\u7f51\u7edc\u5c42":47,"\u7ebf\u7a0bid\u53f7":72,"\u7ec6\u8282\u63cf\u8ff0":71,"\u7ecf\u5e38\u4f1a\u6d88\u8017\u657010gb\u7684\u5185\u5b58\u548c\u6570gb\u7684\u663e\u5b58":46,"\u7ed3\u5408":73,"\u7ed3\u5c3e":63,"\u7ed3\u675f\u6807\u8bb0":59,"\u7ed3\u6784\u5982\u4e0b":88,"\u7ed3\u6784\u5982\u4e0b\u56fe":80,"\u7ed3\u679c\u4fdd\u5b58\u5728":87,"\u7ed3\u679c\u53d1\u73b0\u5b83\u4eec\u4e0e\u540c\u65f6\u671f\u7684twitter\u6d88\u606f\u4e2d\u7684\u60c5\u7eea\u8bcd\u9891\u7387\u76f8\u5173":88,"\u7ed3\u679c\u8f93\u51fa\u5230":60,"\u7ed3\u8bba":33,"\u7ed9":56,"\u7ed9\u4e2a\u7b80\u5355\u7684":61,"\u7ed9\u51fa\u56fe\u7247\u5c3a\u5bf8":81,"\u7ed9\u51fa\u8f93\u5165\u6570\u636e\u6240\u5728\u8def\u5f84":81,"\u7ed9\u5b9a\u52a8\u8bcd":87,"\u7ed9\u5b9a\u7684\u6587\u672c\u53ef\u4ee5\u662f\u4e00\u4e2a\u6587\u6863":88,"\u7ed9\u5b9aencoder\u8f93\u51fa\u548c\u5f53\u524d\u8bcd":58,"\u7edf\u4e00\u7528":21,"\u7ee7\u627f\u81ea":63,"\u7ee7\u627f\u81eaoperatorbas":63,"\u7ee7\u7eed\u8bad\u7ec3\u6216\u9884\u6d4b":2,"\u7ef4\u57fa\u767e\u79d1\u4e2d\u6587\u9875\u9762":56,"\u7ef4\u57fa\u767e\u79d1\u9875\u9762":56,"\u7ef4\u5ea6\u4e3aword_dim":84,"\u7ef4\u5ea6\u662f\u7c7b\u522b\u4e2a\u6570":84,"\u7ef4\u5ea6\u662f\u8bcd\u5178\u5927\u5c0f":84,"\u7ef4\u62a4":73,"\u7ef4\u7a7a\u95f4":59,"\u7ef4\u7a7a\u95f4\u5b8c\u6210":59,"\u7f13\u5b58\u6c60\u7684\u51cf\u5c0f":46,"\u7f13\u5b58\u8bad\u7ec3\u6570\u636e\u5230\u5185\u5b58":2,"\u7f16\u5199\u597d\u6570\u636e\u63d0\u4f9b\u811a\u672c\u540e":86,"\u7f16\u5199\u5b8cyaml\u6587\u4ef6\u540e":75,"\u7f16\u5199\u672c\u6b21\u8bad\u7ec3\u7684yaml\u6587\u4ef6":75,"\u7f16\u53f7":86,"\u7f16\u53f7\u4ece0\u5f00\u59cb":46,"\u7f16\u53f7\u5b57\u6bb5":86,"\u7f16\u7801\u5411\u91cf":59,"\u7f16\u7801\u5668\u8f93\u51fa":59,"\u7f16\u7801\u6e90\u5e8f\u5217":59,"\u7f16\u89e3\u7801\u6a21\u578b\u5c06\u4e00\u4e2a\u6e90\u8bed\u53e5\u7f16\u7801\u4e3a\u4e00\u4e2a\u5b9a\u957f\u7684\u5411\u91cf":89,"\u7f16\u8bd1":[49,60,61],"\u7f16\u8bd1\u548c\u5b89\u88c5paddlepaddl":54,"\u7f16\u8bd1\u548c\u5b89\u88c5paddlepaddle\u9884\u6d4b\u5e93":53,"\u7f16\u8bd1\u548c\u6267\u884c":63,"\u7f16\u8bd1\u5668":[53,54],"\u7f16\u8bd1\u5668\u6ca1\u6709":33,"\u7f16\u8bd1\u578b\u8bed\u8a00":33,"\u7f16\u8bd1\u5b8c\u6210\u4e4b\u540e":65,"\u7f16\u8bd1\u5de5\u5177\u94fe\u6240\u5728\u7684\u7edd\u5bf9\u8def\u5f84":54,"\u7f16\u8bd1\u6210\u529f\u540e":63,"\u7f16\u8bd1\u6210\u52a8\u6001\u5e93":70,"\u7f16\u8bd1\u6d41\u7a0b\u4e3b\u8981\u63a8\u8350\u9ad8\u7ea7\u7528\u6237\u67e5\u770b":50,"\u7f16\u8bd1\u751f\u6210":65,"\u7f16\u8bd1\u8fd9\u4e2a\u7248\u672c\u7684docker\u53d1\u884c\u955c\u50cf":42,"\u7f16\u8bd1\u8fd9\u4e2a\u7248\u672c\u7684python":42,"\u7f16\u8bd1\u8fd9\u4e2a\u7248\u672c\u7684ubuntu":42,"\u7f16\u8bd1\u9009\u9879":48,"\u7f16\u8bd1c":34,"\u7f16\u8bd1master\u5206\u652f\u7684docker\u53d1\u884c\u955c\u50cf":42,"\u7f16\u8bd1paddlepaddle\u548c\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5":66,"\u7f16\u8bd1ubuntu\u7684deb\u5305":42,"\u7f16\u8f91":73,"\u7f29\u653e\u53c2\u6570":82,"\u7f51\u7edc":[87,88],"\u7f51\u7edc\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf":73,"\u7f51\u7edc\u540d\u79f0":84,"\u7f51\u7edc\u5c42\u53ef\u4ee5\u6709\u591a\u4e2a\u8f93\u5165":62,"\u7f51\u7edc\u5c42\u7684\u6807\u8bc6\u7b26\u4e3a":62,"\u7f51\u7edc\u5c42\u7684\u7c7b\u578b":62,"\u7f51\u7edc\u5c42\u7684\u7ec6\u8282\u53ef\u4ee5\u901a\u8fc7\u4e0b\u9762\u8fd9\u4e9b\u4ee3\u7801\u7247\u6bb5\u6765\u6307\u5b9a":62,"\u7f51\u7edc\u5c42\u7684\u8f93\u51fa\u662f\u7ecf\u8fc7\u6fc0\u6d3b\u51fd\u6570\u4e4b\u540e\u7684\u503c":70,"\u7f51\u7edc\u5c42\u914d\u7f6e\u5305\u542b\u4ee5\u4e0b\u51e0\u9879":62,"\u7f51\u7edc\u63a5\u53e3\u5361":68,"\u7f51\u7edc\u6a21\u5757":82,"\u7f51\u7edc\u6a21\u578b\u5c06\u8f93\u51fa\u6807\u7b7e\u7684\u6982\u7387\u5206\u5e03":87,"\u7f51\u7edc\u7684\u8bad\u7ec3\u8fc7\u7a0b":88,"\u7f51\u7edc\u7ed3\u6784\u5982\u4e0b\u56fe\u6240\u793a":86,"\u7f51\u7edc\u901a\u4fe1":62,"\u7f51\u7edc\u914d\u7f6e":[68,84,88],"\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6":[82,87],"\u800c":[46,47,59,63,74,86],"\u800c\u4e0d\u4f1a\u6539\u53d8\u539f\u6709tensor\u7684shape\u4fe1\u606f":64,"\u800c\u4e0d\u4f7f\u7528imdb\u6570\u6910\u96c6\u4e2d\u7684imdb":88,"\u800c\u4e0d\u5fc5\u5728\u610fpaddl":34,"\u800c\u4e0d\u652f\u6301pypy\u89e3\u91ca\u5668":33,"\u800c\u4e0d\u662f\u4f7f\u7528\u540c\u6b65":68,"\u800c\u4e0d\u662f\u5728layer\u91cc\u5b9e\u73b0":46,"\u800c\u4e0d\u662f\u6e90\u7801\u76ee\u5f55\u91cc":46,"\u800c\u4e0d\u662f\u7279\u5f81\u7684\u96c6\u5408":56,"\u800c\u4e0d\u662f\u7ec4\u5408\u4e0a\u4e0b\u6587\u7ea7\u522b\u4fe1\u606f":88,"\u800c\u4e0d\u66b4\u9732\u6982\u5ff5\u7684\u5b9e\u73b0":34,"\u800c\u4e0d\u7528\u5173\u5fc3\u6570\u636e\u5982\u4f55\u4f20\u8f93":2,"\u800c\u4e14":89,"\u800c\u4e14\u5728\u4f20\u8f93\u7684\u8fc7\u7a0b\u4e2d\u4e5f\u53ef\u80fd\u51fa\u73b0\u7f51\u7edc\u4e0d\u7a33\u5b9a\u7684\u60c5\u51b5":28,"\u800c\u4e4b\u524d\u7684\u53c2\u6570\u5c06\u4f1a\u88ab\u5220\u9664":70,"\u800c\u4ece\u5e94\u7528\u7684\u89d2\u5ea6":67,"\u800c\u4f18\u5316\u6027\u80fd\u7684\u9996\u8981\u4efb\u52a1":67,"\u800c\u5176\u4ed6\u5c42\u4f7f\u7528cpu\u8ba1\u7b97":72,"\u800c\u51fa\u73b0\u9636\u6bb5\u6027\u7684\u8fd0\u884c\u505c\u6ede":20,"\u800c\u53cc\u5c42rnn\u662f\u53ef\u4ee5\u5904\u7406\u8fd9\u79cd\u8f93\u5165\u6570\u636e\u7684\u7f51\u7edc\u7ed3\u6784":56,"\u800c\u53ea\u9700\u8981\u83b7\u5f97recurr":46,"\u800c\u53f3\u56fe\u7684\u74f6\u9888\u8fde\u63a5\u6a21\u5757\u7528\u4e8e50\u5c42":82,"\u800c\u5728\u8ba1\u7b97\u7ed3\u675f\u4e4b\u540e":64,"\u800c\u5728cpp\u91cc\u9762\u5b9e\u73b0\u8fd9\u4e2ac\u7684\u63a5\u53e3":33,"\u800c\u591a\u8bed\u8a00\u63a5\u53e3\u9700\u8981\u76f4\u63a5\u8bfb\u53d6\u751f\u6210\u7684\u4e8c\u8fdb\u5236":33,"\u800c\u5927\u591a\u6570\u65b9\u6cd5\u53ea\u662f\u5229\u7528n":88,"\u800c\u5b89\u88c5\u5305":46,"\u800c\u5b89\u88c5\u5305\u662f":46,"\u800c\u5bf9\u4e8e\u53cc\u5c42\u5e8f\u5217":56,"\u800c\u5bf9\u4e8e\u6bcf\u4e00\u4e2a\u5185\u5c42\u7279\u5f81\u6570\u636e\u800c\u8a00":56,"\u800c\u5bf9\u4e8egolang":33,"\u800c\u5bf9\u4e8egolang\u9519\u8bef\u5904\u7406\u5e94\u8be5\u4f7f\u7528\u8fd4\u56de\u503c":33,"\u800c\u5c06\u8fd9\u4e2a\u6bb5\u843d\u7684\u6bcf\u4e00\u53e5\u8bdd\u7528lstm\u7f16\u7801\u6210\u4e00\u4e2a\u5411\u91cf":56,"\u800c\u5f53\u524d\u5df2\u7ecf\u67095":67,"\u800c\u662f\u76f4\u63a5\u4ece\u5185\u5b58\u7684\u7f13\u5b58\u91cc\u8bfb\u53d6\u6570\u636e":46,"\u800c\u662f\u76f4\u63a5\u4fee\u6539paddl":34,"\u800c\u662f\u76f4\u63a5\u7528api\u7684\u63a5\u53e3\u8fdc\u7a0b\u8bbf\u95ee":21,"\u800c\u66f4\u6df1\u5165\u7684\u5206\u6790":67,"\u800c\u6709\u4e9b\u53c2\u6570\u9700\u8981\u5728\u96c6\u7fa4\u591a\u673a\u8bad\u7ec3\u4e2d\u4f7f\u7528\u7b49":69,"\u800c\u6ca1\u6709\u77ed\u65f6\u8bb0\u5fc6\u7684\u635f\u5931":88,"\u800c\u6e90\u5e8f\u5217\u7684\u7f16\u7801\u5411\u91cf\u53ef\u4ee5\u88ab\u65e0\u8fb9\u754c\u7684memory\u8bbf\u95ee":59,"\u800c\u795e\u7ecf\u7f51\u7edc\u662f\u6211\u4eec\u8981\u642d\u5efa\u7684\u5b9d\u5854":51,"\u800c\u7a00\u758f\u66f4\u65b0\u5728\u53cd\u5411\u4f20\u64ad\u4e4b\u540e\u7684\u6743\u91cd\u66f4\u65b0\u65f6\u8fdb\u884c":72,"\u800c\u8ba1\u7b97\u8fc7\u7a0b\u662f\u7531":64,"\u800c\u8fd9\u4e00\u53e5\u8bdd\u5c31\u53ef\u4ee5\u8868\u793a\u6210\u8fd9\u4e9b\u4f4d\u7f6e\u7684\u6570\u7ec4":56,"\u800c\u8fd9\u4e2acontext\u53ef\u80fd\u4f1a\u975e\u5e38\u5927":2,"\u800c\u8fd9\u6bcf\u4e00\u4e2a\u6570\u7ec4\u5143\u7d20":56,"\u800c\u975e\u76f4\u63a5\u56de\u590d\u7684\u65b9\u5f0f":61,"\u800c\u975e\u9759\u6001\u52a0\u8f7dcuda\u52a8\u6001\u5e93":48,"\u800ceigenvector":64,"\u800cpaddlepaddle\u5219\u4f1a\u5e2e\u7528\u6237\u505a\u4ee5\u4e0b\u5de5\u4f5c":2,"\u800crnn\u662f\u6700\u6d41\u884c\u7684\u9009\u62e9":58,"\u800cswig\u53ea\u80fd\u7b80\u5355\u7684\u66b4\u9732c":33,"\u800ctrainer\u9700\u8981\u8bfb\u53d6\u8bad\u7ec3\u6570\u636e\u8fdb\u884c\u8bad\u7ec3":51,"\u800cweight":81,"\u800cy_predict\u662f\u63a5\u6536x\u4f5c\u4e3a\u8f93\u5165":51,"\u804c\u4e1a":85,"\u804c\u4e1a\u4ece\u4e0b\u9762\u6240\u5217\u4e2d\u9009\u62e9":85,"\u80fd\u591f\u5904\u7406\u53cc\u5c42\u5e8f\u5217":58,"\u80fd\u591f\u5bf9\u53cc\u5411\u5e8f\u5217\u8fdb\u884c\u5904\u7406\u7684\u6709":58,"\u80fd\u591f\u627e\u5230\u8fd9\u91cc\u4f7f\u7528\u7684\u6240\u6709\u6570\u636e":84,"\u80fd\u591f\u8bb0\u5f55\u4e0a\u4e00\u4e2asubseq":58,"\u80fd\u591f\u9488\u5bf9cpu\u548cgpu\u7684\u8ba1\u7b97\u505a\u66f4\u591a\u4f18\u5316":46,"\u80fd\u83b7\u53d6":68,"\u811a\u672c":[60,68,81,86],"\u811a\u672c\u4fdd\u5b58\u5728":81,"\u811a\u672c\u5f00\u59cb\u65f6":75,"\u811a\u672c\u63d0\u4f9b\u4e86\u4e00\u4e2a\u9884\u6d4b\u63a5\u53e3":88,"\u811a\u672c\u65f6\u9700\u8981\u52a0\u4e0a":88,"\u811a\u672c\u8fd0\u884c\u5b8c\u6210\u540e":81,"\u81ea\u52a8\u5173\u95ed\u5bf9\u5e94\u7684":61,"\u81ea\u52a8\u5730\u5c06\u8fd9\u4e9b\u9009\u9879\u5e94\u7528\u5230":68,"\u81ea\u52a8\u5b8c\u6210\u8fd9\u4e00\u8fc7\u7a0b":58,"\u81ea\u52a8\u6302\u8f7d\u5206\u5e03\u5f0f\u5b58\u50a8\u76ee\u5f55":20,"\u81ea\u52a8\u751f\u6210":65,"\u81ea\u52a8\u83b7\u53d6\u4e0a\u4e00\u4e2a\u751f\u6210\u7684\u8bcd":59,"\u81ea\u5e95\u5411\u4e0a\u6cd5":84,"\u81ea\u6b64":53,"\u81ea\u7136\u4e5f\u5c31\u6709\u7ba1\u7406\u5458\u6743\u9650":60,"\u81ea\u7136\u8bed\u8a00\u5904\u7406\u7b49":72,"\u81ea\u7531\u804c\u4e1a\u8005":85,"\u81f3\u4e8e\u4e3a\u4ec0\u4e48\u9700\u8981c":34,"\u81f3\u6b64":[2,56,61],"\u8212\u9002":56,"\u826f\u597d\u7684\u6587\u6863":33,"\u827a\u672f\u5bb6":85,"\u8282\u70b9\u4e2d\u7684":68,"\u82e5":62,"\u82e5\u5728paddlepaddle\u7f16\u8bd1\u65f6":46,"\u82e5\u5e0c\u671b\u6700\u5feb\u7684\u6267\u884c\u901f\u5ea6":[53,54],"\u82e5\u5e72\u4e2a\u53e5\u5b50\u6784\u6210\u4e00\u4e2a\u6bb5\u843d":55,"\u82e5\u6709\u4e0d\u4e00\u81f4\u4e4b\u5904":67,"\u82e5\u6709\u5fc5\u8981":62,"\u82e5\u6ca1\u6709\u663e\u5f0f\u8bbe\u7f6e":53,"\u82e5\u73af\u5883\u53d8\u91cf":[53,54],"\u82e5\u8981\u5bf9\u8fd9\u51e0\u4e2alayer\u4f7f\u7528dropout":46,"\u82e5\u8f93\u51fa\u662f\u5355\u5c42\u5e8f\u5217":55,"\u82e5\u8f93\u51fa\u662f\u53cc\u5c42\u5e8f\u5217":55,"\u82f1\u6587\u6587\u6863":65,"\u82f1\u6587\u6587\u6863\u76ee\u5f55":65,"\u82f1\u8bed":89,"\u8303\u56f4":72,"\u83b7\u53d6":61,"\u83b7\u53d6\u5229\u7528":84,"\u83b7\u53d6\u5b57\u5178\u7ef4\u5ea6":88,"\u83b7\u53d6\u5f53\u524d\u7cfb\u7edf\u652f\u6301\u7684python\u5305\u7684\u540e\u7f00":46,"\u83b7\u53d6\u6700\u65b0\u7684\u68c0\u67e5\u70b9\u7684\u6587\u4ef6uuid":20,"\u83b7\u53d6\u6e90\u7801":60,"\u83b7\u53d6\u8be5\u6761\u6837\u672c\u7c7b\u522bid":84,"\u83b7\u53d6\u901a\u8fc7":88,"\u83b7\u53d6trainer":75,"\u83b7\u5f97\u53c2\u6570\u5c3a\u5bf8":62,"\u865a\u62df\u673a\u4e0a":60,"\u867d\u7136":47,"\u867d\u7136\u4e0d\u9f13\u52b1\u8fd9\u6837":34,"\u867d\u7136\u6bcf\u4e2agenerator\u5728\u6ca1\u6709\u8c03\u7528\u7684\u65f6\u5019":2,"\u867d\u7136\u8fd9\u4e9b\u6587\u4ef6\u5e76\u975e\u90fd\u9700\u8981\u96c6\u7fa4\u8bad\u7ec3":68,"\u867d\u7136paddle\u770b\u8d77\u6765\u5305\u542b\u4e86\u4f17\u591a\u53c2\u6570":69,"\u884c":80,"\u884c\u4f18\u5148\u6b21\u5e8f\u5b58\u50a8":82,"\u884c\u5185\u4f7f\u7528":2,"\u884c\u653f\u5de5\u4f5c":85,"\u884c\u955c\u50cf\u5305\u62ec\u7eafcpu\u7248\u672c\u548cgpu\u7248\u672c\u4ee5\u53ca\u5176\u5bf9\u5e94\u7684\u975eavx\u7248\u672c":49,"\u8865\u5145\u4e0a\u6b21\u7684commit":61,"\u8868\u660e\u4e86\u8fd9\u4e9b\u884c\u7684\u6807\u53f7":62,"\u8868\u660e\u8fd9\u4e2a\u5c42\u7684\u4e00\u4e2a\u5b9e\u4f8b\u662f\u5426\u9700\u8981\u504f\u7f6e":62,"\u8868\u793a":63,"\u8868\u793a\u4e00\u4e2akubernetes\u96c6\u7fa4\u4e2d\u7684\u4e00\u4e2a\u5de5\u4f5c\u8282\u70b9":73,"\u8868\u793a\u4e3adeviceid":72,"\u8868\u793a\u5728\u96c6\u7fa4\u4f5c\u4e1a":68,"\u8868\u793a\u5973\u6027":85,"\u8868\u793a\u5c06\u5916\u5c42\u7684outer_mem\u4f5c\u4e3a\u5185\u5c42memory\u7684\u521d\u59cb\u72b6\u6001":56,"\u8868\u793a\u5f53\u524d\u96c6\u7fa4\u4f5c\u4e1a\u7684\u8282\u70b9":68,"\u8868\u793a\u6570\u636e\u7c7b\u578b":63,"\u8868\u793a\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":68,"\u8868\u793a\u751f\u6210\u6570\u636e\u7684\u5e8f\u5217id":89,"\u8868\u793a\u7528\u4e8e\u8bad\u7ec3\u6216\u9884\u6d4b":2,"\u8868\u793a\u7537\u6027":85,"\u8868\u793a\u7684\u6bcf\u4e2a\u5355\u8bcd":84,"\u8868\u793a\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u7684\u5206\u7c7b\u9519\u8bef":88,"\u8868\u793a\u8bad\u7ec3\u4e86xx\u4e2a\u6837\u672c":88,"\u8868\u793a\u8bad\u7ec3\u4e86xx\u4e2abatch":88,"\u8868\u793a\u8bbe\u5907\u7c7b\u578b":63,"\u8868\u793a\u8bfb\u8005\u6240\u4f7f\u7528\u7684docker\u955c\u50cf\u4ed3\u5e93\u5730\u5740":75,"\u8868\u793a\u8fc7\u4e8620\u4e2abatch":84,"\u8868\u793a\u8fc7\u4e862560\u4e2a\u6837\u672c":84,"\u8868\u793a\u8fd9\u4e2ajob\u7684\u540d\u5b57":75,"\u88ab":61,"\u88ab\u6269\u5c55\u4e3a\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":55,"\u88ab\u653e\u5728":62,"\u88ab\u79f0\u4e3a":59,"\u897f\u90e8\u7247":85,"\u8981\u4e0b\u8f7d\u548c\u89e3\u538b\u6570\u636e\u96c6":86,"\u8981\u4e0b\u8f7d\u89e3\u538b\u8fd9\u4e2a\u6a21\u578b":89,"\u8981\u4f7f\u7528\u547d\u4ee4\u884c\u5206\u6790\u5de5\u5177":67,"\u8981\u5728\u5df2\u6709\u7684kubernetes\u96c6\u7fa4\u4e0a\u8fdb\u884cpaddlepaddle\u7684\u5206\u5e03\u5f0f\u8bad\u7ec3":75,"\u8981\u5728\u6240\u6709\u8282\u70b9\u4e0a\u5b58\u5728":68,"\u8981\u5bf9\u4e00\u4e2a\u56fe\u7247\u7684\u8fdb\u884c\u5206\u7c7b\u9884\u6d4b":81,"\u8981\u5c06\u5b57\u6bb5\u914d\u7f6e\u6587\u4ef6\u8f6c\u5316\u4e3ameta\u914d\u7f6e\u6587\u4ef6":86,"\u8981\u6c42\u5355\u5c42\u5e8f\u5217\u542b\u6709\u5143\u7d20\u7684\u6570\u76ee":55,"\u8981\u751f\u6210\u7684\u76ee\u6807\u5e8f\u5217":58,"\u8981\u8c03\u7528":62,"\u89e3\u51b3\u529e\u6cd5\u662f":46,"\u89e3\u51b3\u65b9\u6848\u662f":46,"\u89e3\u538b":89,"\u89e3\u6790\u5668\u80fd\u901a\u8fc7\u6587\u4ef6\u7684\u6269\u5c55\u540d\u81ea\u52a8\u8bc6\u522b\u6587\u4ef6\u7684\u683c\u5f0f":86,"\u89e3\u6790\u6570\u636e\u96c6\u4e2d\u7684\u6bcf\u4e00\u4e2a\u5b57\u6bb5":86,"\u89e3\u6790\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":4,"\u89e3\u6790\u73af\u5883\u53d8\u91cf\u5f97\u5230":75,"\u89e3\u6790\u8bad\u7ec3\u6a21\u578b\u65f6\u7528\u7684\u914d\u7f6e\u6587\u4ef6":4,"\u89e3\u7801\u5668\u4f7f\u7528":59,"\u89e3\u7801\u5668\u57fa\u4e8e\u7f16\u7801\u6e90\u5e8f\u5217\u548c\u6700\u540e\u751f\u6210\u7684\u76ee\u6807\u8bcd\u9884\u6d4b\u4e0b\u4e00\u76ee\u6807\u8bcd":59,"\u89e3\u7801\u5668\u662f\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":59,"\u89e3\u7801\u5668\u6839\u636e\u4e0a\u4e0b\u6587\u5411\u91cf\u9884\u6d4b\u51fa\u4e00\u4e2a\u76ee\u6807\u5355\u8bcd":89,"\u89e3\u91ca":84,"\u89e3\u91ca\u578b\u8bed\u8a00\u53ea\u80fd\u8c03\u7528\u52a8\u6001\u5e93":33,"\u89e3\u91ca\u6027\u8bed\u8a00\u5b9e\u9645\u8fd0\u884c\u7684\u4e8c\u8fdb\u5236\u662f\u89e3\u91ca\u5668\u672c\u8eab":33,"\u8ba1\u7b97":59,"\u8ba1\u7b97\u504f\u7f6e\u7684\u68af\u5ea6":62,"\u8ba1\u7b97\u5355\u5143\u662f\u795e\u7ecf\u7f51\u7edc\u7684\u6838\u5fc3":47,"\u8ba1\u7b97\u53cd\u5411rnn\u7684\u7b2c\u4e00\u4e2a\u5b9e\u4f8b":59,"\u8ba1\u7b97\u53d8\u6362\u77e9\u9635\u7684\u5927\u5c0f\u548c\u683c\u5f0f":62,"\u8ba1\u7b97\u5f53\u524d\u5c42\u6743\u91cd\u7684\u68af\u5ea6":62,"\u8ba1\u7b97\u6548\u7387\u66f4\u9ad8":46,"\u8ba1\u7b97\u6bcf\u4e2a\u8bcd\u7684\u8bcd\u5411\u91cf":59,"\u8ba1\u7b97\u6fc0\u6d3b\u51fd\u6570\u7684\u68af\u5ea6":62,"\u8ba1\u7b97\u7684\u7ec6\u8282\u5c06\u5728\u4e0b\u9762\u7684\u5c0f\u8282\u7ed9\u51fa":62,"\u8ba1\u7b97\u8bef\u5dee\u51fd\u6570":47,"\u8ba1\u7b97\u8f6c\u6362\u77e9\u9635\u548c\u8f93\u5165\u7684\u68af\u5ea6":62,"\u8ba1\u7b97\u8f93\u5165\u548c\u53c2\u6570\u7684\u68af\u5ea6":62,"\u8ba1\u7b97\u8f93\u5165\u5c42\u7684\u504f\u5dee":62,"\u8ba1\u7b97\u8f93\u51fa":62,"\u8ba1\u7b97\u8fd9\u4e2a\u6587\u4ef6\u7684md5":20,"\u8ba1\u7b97\u9700\u8981\u7684\u6570\u636e\u5b58\u653e\u5728":64,"\u8ba9\u6a21\u578b\u80fd\u591f\u5f97\u5230\u8bad\u7ec3\u66f4\u65b0":84,"\u8ba9\u795e\u7ecf\u7f51\u7edc\u53ef\u4ee5\u8fdb\u884c\u8bad\u7ec3\u6216\u9884\u6d4b":1,"\u8ba9\u8fd9\u4e2a\u793a\u4f8b\u53d8\u5f97\u66f4\u597d":86,"\u8ba9paddle\u6838\u5fc3\u4e2d":34,"\u8bad\u7ec3":[69,88],"\u8bad\u7ec3\u4e0e\u5e94\u7528":0,"\u8bad\u7ec3\u4efb\u52a1\u7684\u8fd0\u884c\u53ef\u80fd\u4f1a\u5360\u6ee1trainer\u548cparamet":20,"\u8bad\u7ec3\u4f5c\u4e1a":68,"\u8bad\u7ec3\u53ca\u6d4b\u8bd5\u8bef\u5dee\u66f2\u7ebf\u56fe\u4f1a\u88ab":81,"\u8bad\u7ec3\u53ef\u4ee5\u8bbe\u7f6e\u4e3atrue":87,"\u8bad\u7ec3\u540e":87,"\u8bad\u7ec3\u548c\u7eaf\u4f7f\u7528":42,"\u8bad\u7ec3\u5931\u8d25\u65f6\u53ef\u4ee5\u68c0\u67e5\u9519\u8bef\u65e5\u5fd7":68,"\u8bad\u7ec3\u597d\u4e00\u4e2a\u6df1\u5c42\u795e\u7ecf\u7f51\u7edc\u901a\u5e38\u8981\u8017\u8d39\u975e\u5e38\u957f\u7684\u65f6\u95f4":67,"\u8bad\u7ec3\u5b8c\u6210\u540e":47,"\u8bad\u7ec3\u6570\u636e\u548c\u6d4b\u8bd5\u6570\u636e\u7684\u76ee\u5f55":89,"\u8bad\u7ec3\u6570\u636e\u662f":2,"\u8bad\u7ec3\u6570\u636e\u6709\u95ee\u9898":46,"\u8bad\u7ec3\u6570\u6910\u96c6":88,"\u8bad\u7ec3\u65f6":75,"\u8bad\u7ec3\u65f6\u6240\u9700\u8bbe\u7f6e\u7684\u4e3b\u8981\u53c2\u6570\u5982\u4e0b":84,"\u8bad\u7ec3\u65f6\u9ed8\u8ba4shuffl":2,"\u8bad\u7ec3\u6a21\u578b\u4e4b\u524d":88,"\u8bad\u7ec3\u6a21\u578b\u540e":59,"\u8bad\u7ec3\u6a21\u578b\u6b63\u786e\u6027":42,"\u8bad\u7ec3\u7684\u635f\u5931\u51fd\u6570\u9ed8\u8ba4\u6bcf\u969410\u4e2abatch\u6253\u5370\u4e00\u6b21":89,"\u8bad\u7ec3\u7684\u811a\u672c\u662f":87,"\u8bad\u7ec3\u7ed3\u675f\u540e\u67e5\u770b\u8f93\u51fa\u7ed3\u679c":75,"\u8bad\u7ec3\u811a\u672c":84,"\u8bad\u7ec3\u811a\u672c\u7b49\u7b49":84,"\u8bad\u7ec3\u81f3\u591a":86,"\u8bad\u7ec3\u8282\u70b9\u6570\u91cf":75,"\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u8ddd\u79bb":46,"\u8bad\u7ec3\u8f6e\u6b21":84,"\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u53c2\u6570\u6216\u8005\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u7684\u68af\u5ea6\u5c3a\u5ea6\u8fc7\u5927":46,"\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u6d4b\u8bd5test_period":69,"\u8bad\u7ec3\u8fc7\u7a0b\u662f\u5426\u4e3a\u672c\u5730\u6a21\u5f0f":70,"\u8bad\u7ec3\u8fc7\u7a0b\u662f\u5426\u4f7f\u7528gpu":70,"\u8bad\u7ec3\u914d\u7f6e\u4e2d\u7684\u8bbe\u5907\u5c5e\u6027\u5c06\u4f1a\u65e0\u6548":70,"\u8bad\u7ec3\u96c6\u548c\u6d4b\u8bd5\u96c6\u6587\u4ef6\u5217\u8868":88,"\u8bad\u7ec3\u96c6\u5df2\u7ecf\u505a\u4e86\u968f\u673a\u6253\u4e71\u6392\u5e8f\u800c\u6d4b\u8bd5\u96c6\u6ca1\u6709":88,"\u8bad\u7ec3\u96c6\u5df2\u7ecf\u968f\u673a\u6253\u4e71":88,"\u8bad\u7ec3\u96c6\u5e73\u5747\u503c":81,"\u8bad\u7ec3dot_period":69,"\u8bb0\u5f55\u4e0b\u6240\u6709\u5931\u8d25\u7684\u4f8b\u5b50":42,"\u8bb0\u5fc6\u6a21\u5757":59,"\u8bba\u6587":82,"\u8bbe\u4e3a\u5df2\u90e8\u7f72\u7684\u5de5\u4f5c\u7a7a\u95f4\u76ee\u5f55":68,"\u8bbe\u4e3a\u672c\u5730":68,"\u8bbe\u5b9a":46,"\u8bbe\u7f6e":[34,46],"\u8bbe\u7f6e\u4e3a":[54,62],"\u8bbe\u7f6e\u4e3a\u4e0d\u540c\u7684\u503c":46,"\u8bbe\u7f6e\u4e3atrue\u4f7f\u7528\u672c\u5730\u8bad\u7ec3\u6216\u8005\u4f7f\u7528\u96c6\u7fa4\u4e0a\u7684\u4e00\u4e2a\u8282\u70b9":70,"\u8bbe\u7f6e\u4e3atrue\u4f7f\u7528gpu\u6a21\u5f0f":70,"\u8bbe\u7f6e\u4e86\u76f8\u540c\u7684\u53d6\u503c":46,"\u8bbe\u7f6e\u4efb\u52a1\u7684\u6a21\u5f0f\u4e3a\u6d4b\u8bd5":89,"\u8bbe\u7f6e\u4fdd\u5b58\u6a21\u578b\u7684\u8f93\u51fa\u8def\u5f84":89,"\u8bbe\u7f6e\u5168\u5c40\u5b66\u4e60\u7387":88,"\u8bbe\u7f6e\u5176\u53c2\u6570\u5c5e\u6027":46,"\u8bbe\u7f6e\u5185\u5b58\u4e2d\u6682\u5b58\u7684\u6570\u636e\u6761\u6570":2,"\u8bbe\u7f6e\u5185\u5b58\u4e2d\u6700\u5c0f\u6682\u5b58\u7684\u6570\u636e\u6761\u6570":2,"\u8bbe\u7f6e\u53c2\u6570\u7684\u540d\u5b57":46,"\u8bbe\u7f6e\u547d\u4ee4\u884c\u53c2\u6570":[46,66,86],"\u8bbe\u7f6e\u5b57\u5178\u6587\u4ef6":88,"\u8bbe\u7f6e\u5b66\u4e60\u7387\u8870\u51cf\u56e0\u5b50\u5206\u6bb5\u51fd\u6570":46,"\u8bbe\u7f6e\u5de5\u4f5c\u6a21\u5f0f\u4e3a\u8bad\u7ec3":88,"\u8bbe\u7f6e\u5e73\u5747sgd\u7a97\u53e3":88,"\u8bbe\u7f6e\u6210":46,"\u8bbe\u7f6e\u6210\u4e00\u4e2a\u5c0f\u4e00\u4e9b\u7684\u503c":46,"\u8bbe\u7f6e\u6570\u636e\u904d\u5386\u6b21\u6570":87,"\u8bbe\u7f6e\u6807\u7b7e\u7c7b\u522b\u5b57\u5178":88,"\u8bbe\u7f6e\u6a21\u578b\u8def\u5f84":88,"\u8bbe\u7f6e\u7684\u547d\u4ee4\u884c\u53c2\u6570":88,"\u8bbe\u7f6e\u795e\u7ecf\u7f51\u7edc\u7684\u914d\u7f6e\u6587\u4ef6":89,"\u8bbe\u7f6e\u7c7b\u522b\u6570":88,"\u8bbe\u7f6e\u7ebf\u7a0b\u6570":[87,88],"\u8bbe\u7f6e\u7f51\u7edc\u914d\u7f6e":88,"\u8bbe\u7f6e\u8f93\u51fa\u7684\u5c3a\u5bf8":62,"\u8bbe\u7f6e\u8f93\u51fa\u8def\u5f84\u4ee5\u4fdd\u5b58\u8bad\u7ec3\u5b8c\u6210\u7684\u6a21\u578b":88,"\u8bbe\u7f6e\u8f93\u51fatensor\u7684\u5f62\u72b6":63,"\u8bbe\u7f6e\u8fd9\u4e2apydataprovider2\u8fd4\u56de\u4ec0\u4e48\u6837\u7684\u6570\u636e":2,"\u8bbe\u7f6e\u9ed8\u8ba4\u8bbe\u5907\u53f7\u4e3a0":72,"\u8bbe\u7f6ebatch":88,"\u8bbe\u7f6ecpu\u7ebf\u7a0b\u6570\u6216\u8005gpu\u8bbe\u5907\u6570":89,"\u8bbe\u7f6egpu":70,"\u8bbe\u7f6epass":88,"\u8bbe\u7f6epasses\u7684\u6570\u91cf":89,"\u8bbf\u95ee\u5230\u4e3b\u673a\u4e0a\u7684\u6587\u4ef6":49,"\u8bbf\u95eekubernetes\u7684\u63a5\u53e3\u6765\u67e5\u8be2\u6b64job\u5bf9\u5e94\u7684\u6240\u6709pod\u4fe1\u606f":75,"\u8bc4\u4ef7\u9884\u6d4b\u7684\u6548\u679c":47,"\u8bc4\u4f30\u8be5\u4ea7\u54c1\u7684\u8d28\u91cf":84,"\u8bc4\u5206":[85,86],"\u8bc4\u5206\u6587\u4ef6\u7684\u6bcf\u4e00\u884c\u4ec5\u4ec5\u63d0\u4f9b\u7535\u5f71\u6216\u7528\u6237\u7684\u7f16\u53f7\u6765\u4ee3\u8868\u76f8\u5e94\u7684\u7535\u5f71\u6216\u7528\u6237":86,"\u8bc4\u5206\u88ab\u8c03\u6574\u4e3a5\u661f\u7684\u89c4\u6a21":85,"\u8bc4\u5ba1\u4eba\u4e00\u822c\u4e0d\u505a\u8bc4\u5ba1":61,"\u8bc4\u5ba1\u4eba\u7684\u6bcf\u4e2a\u610f\u89c1\u90fd\u5fc5\u987b\u56de\u590d":61,"\u8bc4\u5ba1\u4eba\u9700\u8981\u9010\u4e00\u67e5\u770b\u6bcf\u4e2acommit\u624d\u80fd\u77e5\u9053\u505a\u4e86\u54ea\u4e9b\u4fee\u6539":61,"\u8bc4\u8bba\u6846\u4e2d\u52a0\u4e0a":61,"\u8bc6\u522b\u6570\u5b57":42,"\u8bcd\u5178\u7684key\u548c":63,"\u8bcd\u5411\u91cf":[42,80],"\u8bcd\u5411\u91cf\u6a21\u578b":83,"\u8bcd\u5411\u91cf\u6a21\u578b\u540d\u79f0":80,"\u8bcd\u672c\u8eab\u548c\u8bcd\u9891":80,"\u8bcd\u9891\u6700\u9ad8\u7684":89,"\u8bd5\u7740\u8ba9\u8f93\u51fa\u7684\u5206\u6790\u6570\u636e\u548c\u7406\u8bba\u503c\u5bf9\u5e94":67,"\u8be5":[68,87],"\u8be5\u51fd\u6570\u5177\u6709\u4e24\u4e2a\u53c2\u6570":2,"\u8be5\u51fd\u6570\u5728\u521d\u59cb\u5316\u7684\u65f6\u5019\u4f1a\u88ab\u8c03\u7528":2,"\u8be5\u51fd\u6570\u7684\u529f\u80fd\u662f":2,"\u8be5\u53c2\u6570\u5728\u7f51\u7edc\u914d\u7f6e\u7684output":70,"\u8be5\u53c2\u6570\u5728\u96c6\u7fa4\u63d0\u4ea4\u73af\u5883\u4e2d\u81ea\u52a8\u8bbe\u7f6e":70,"\u8be5\u53c2\u6570\u5df2\u7ecf\u5728\u96c6\u7fa4\u63d0\u4ea4\u73af\u5883\u4e2d\u5b8c\u6210\u8bbe\u7f6e":70,"\u8be5\u53c2\u6570\u5fc5\u987b\u80fd\u88abflag":70,"\u8be5\u53c2\u6570\u6307\u793a\u662f\u5426\u6253\u5370\u65e5\u5fd7\u622a\u65ad\u4fe1\u606f":70,"\u8be5\u53c2\u6570\u6307\u793a\u662f\u5426\u6253\u5370\u9519\u8bef\u622a\u65ad\u65e5\u5fd7":70,"\u8be5\u53c2\u6570\u7528\u4e8e\u6307\u5b9a\u52a8\u6001\u5e93\u8def\u5f84":70,"\u8be5\u53c2\u6570\u7684\u610f\u601d\u662f\u8bad\u7ec3num":70,"\u8be5\u53c2\u6570\u9ed8\u8ba4\u4e3anull":70,"\u8be5\u5bf9\u8c61\u5177\u6709\u4ee5\u4e0b\u4e24\u4e2a\u5c5e\u6027":2,"\u8be5\u5c42\u4ec5\u9700\u8981\u8fd9\u4e9b\u975e\u96f6\u6837\u672c\u4f4d\u7f6e\u6240\u5bf9\u5e94\u7684\u53d8\u6362\u77e9\u9635\u7684\u90a3\u4e9b\u884c":62,"\u8be5\u5c42\u795e\u7ecf\u5143\u4e2a\u6570":84,"\u8be5\u622a\u65ad\u4f1a\u5f71\u54cd":70,"\u8be5\u6279\u6b21\u7684\u8f93\u5165\u4e2d\u4ec5\u6709\u4e00\u4e2a\u5b50\u96c6\u662f\u975e\u96f6\u7684":62,"\u8be5\u63a5\u53e3\u4f7f\u7528\u591a\u7ebf\u7a0b\u8bfb\u53d6\u6570\u636e":2,"\u8be5\u63a5\u53e3\u53ef\u7528\u4e8e\u9884\u6d4b\u548c\u5b9a\u5236\u5316\u8bad\u7ec3":48,"\u8be5\u6570\u636e\u53ca\u6709\u5f88\u591a\u4e0d\u540c\u7684\u7248\u672c":85,"\u8be5\u6570\u636e\u96c6":80,"\u8be5\u6570\u636e\u96c6\u4e8e2003\u5e742\u6708\u53d1\u5e03":85,"\u8be5\u6570\u636e\u96c6\u5305\u542b\u4e00\u4e9b\u7528\u6237\u4fe1\u606f":85,"\u8be5\u6570\u76ee\u662f\u63d0\u524d\u5b9a\u4e49\u597d\u7684":70,"\u8be5\u6587\u4ef6\u53ef\u4ee5\u4ece\u5b57\u6bb5\u914d\u7f6e\u6587\u4ef6\u751f\u6210":86,"\u8be5\u6587\u4ef6\u662f\u7531cpickle\u4ea7\u751f\u7684":82,"\u8be5\u6587\u4ef6\u662fpython\u7684pickle\u5bf9\u8c61":86,"\u8be5\u6587\u4ef6\u8d1f\u8d23\u4ea7\u751f\u56fe\u7247\u6570\u636e\u5e76\u4f20\u9012\u7ed9paddle\u7cfb\u7edf":81,"\u8be5\u6a21\u578b\u4f9d\u7136\u4f7f\u7528\u903b\u8f91\u56de\u5f52\u5206\u7c7b\u7f51\u7edc\u7684\u6846\u67b6":84,"\u8be5\u6a21\u578b\u5728\u957f\u8bed\u53e5\u7ffb\u8bd1\u7684\u573a\u666f\u4e0b\u6548\u679c\u63d0\u5347\u66f4\u52a0\u660e\u663e":89,"\u8be5\u6a21\u578b\u7684\u7f51\u7edc\u914d\u7f6e\u5982\u4e0b":47,"\u8be5\u6a21\u578b\u7684\u8bf4\u660e\u5982\u4e0b\u56fe\u6240\u793a":59,"\u8be5\u6a21\u578b\u7f51\u7edc\u53ea\u662f\u7528\u4e8e\u8fdb\u884cdemo\u5c55\u793apaddle\u5982\u4f55\u5de5\u4f5c":86,"\u8be5\u793a\u4f8b\u5c06\u5c55\u793apaddle\u5982\u4f55\u8fdb\u884c\u8bcd\u5411\u91cf\u5d4c\u5165":86,"\u8be5\u793a\u4f8b\u7684\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6":86,"\u8be5\u7b97\u6cd5\u6bcf\u6279\u91cf":47,"\u8be5\u7c7b\u7684":63,"\u8be5\u7c7b\u7684\u5b9e\u73b0\u7ec6\u8282\u5728":62,"\u8be5\u7f16\u8bd1\u5de5\u5177\u94fe\u9700\u8981\u7cfb\u7edfglibc\u652f\u63012":54,"\u8be5\u811a\u672c\u4ec5\u4ec5\u662f\u5f00\u59cb\u4e00\u4e2apaddle\u8bad\u7ec3\u8fc7\u7a0b":86,"\u8be5\u811a\u672c\u4f1a\u751f\u6210\u4e00\u4e2adot\u6587\u4ef6":82,"\u8be5\u811a\u672c\u5c06\u8f93\u51fa\u9884\u6d4b\u5206\u7c7b\u7684\u6807\u7b7e":81,"\u8be5\u8bed\u53e5\u4f1a\u4e3a\u6bcf\u4e2a\u5c42\u521d\u59cb\u5316\u5176\u6240\u9700\u8981\u7684\u53d8\u91cf\u548c\u8fde\u63a5":62,"\u8be5github\u4ed3\u5e93\u4e2d\u5305\u542b\u82e5\u5e72\u4e2a\u9884\u7f16\u8bd1\u597d\u7684":54,"\u8be5layer\u662f\u901a\u8fc7\u53c2\u6570":46,"\u8be6\u7ec6\u4ecb\u7ecd\u53ef\u4ee5\u53c2\u8003":56,"\u8be6\u7ec6\u4ecb\u7ecd\u8bf7\u53c2\u8003\u8bbe\u8ba1\u6587\u6863":63,"\u8be6\u7ec6\u4fe1\u606f\u8bf7\u68c0\u67e5":68,"\u8be6\u7ec6\u5185\u5bb9\u8bf7\u53c2\u89c1":84,"\u8be6\u7ec6\u53ef\u53c2\u8003":61,"\u8be6\u7ec6\u5730\u5c55\u793a\u4e86\u6574\u4e2a\u7279\u5f81\u63d0\u53d6\u7684\u8fc7\u7a0b":82,"\u8be6\u7ec6\u6587\u6863\u53c2\u8003":46,"\u8be6\u7ec6\u7684\u53c2\u6570\u89e3\u91ca":84,"\u8be6\u7ec6\u7684cmake\u4f7f\u7528\u65b9\u6cd5\u53ef\u4ee5\u53c2\u8003":48,"\u8be6\u7ec6\u89c1":55,"\u8be6\u7ec6\u8bbe\u8ba1":28,"\u8bed\u4e49\u89d2\u8272\u6807\u6ce8":[83,87],"\u8bed\u610f\u89d2\u8272\u6807\u6ce8":42,"\u8bed\u8a00\u6a21\u578b":80,"\u8bf4\u660e":[20,48,75],"\u8bf4\u660e\u63d0\u4ea4\u7684\u4ee3\u7801\u5b58\u5728\u95ee\u9898":61,"\u8bf4\u660e\u6bcf\u4e2a\u7279\u5f81\u6587\u4ef6\u5177\u4f53\u5b57\u6bb5\u662f":86,"\u8bf4\u660e\u8fd9\u4e2a\u5c42\u7684\u8f93\u5165":62,"\u8bf7\u4e0d\u8981\u521b\u5efa\u7a7a\u7684":63,"\u8bf7\u4fdd\u8bc1travi":61,"\u8bf7\u5148\u4f7f\u7528":[53,54],"\u8bf7\u53c2\u7167\u7f51\u7edc\u914d\u7f6e\u7684\u6587\u6863\u4e86\u89e3\u66f4\u8be6\u7ec6\u7684\u4fe1\u606f":72,"\u8bf7\u53c2\u8003":[2,34,46,51,56,62,63,84],"\u8bf7\u53c2\u8003\u5982\u4e0b\u8868\u683c":84,"\u8bf7\u53c2\u8003\u9875\u9762":86,"\u8bf7\u53c2\u8003layer\u6587\u6863":81,"\u8bf7\u53c2\u89c1":61,"\u8bf7\u53c2\u9605":59,"\u8bf7\u53c2\u9605\u60c5\u611f\u5206\u6790\u7684\u6f14\u793a\u4ee5\u4e86\u89e3\u6709\u5173\u957f\u671f\u77ed\u671f\u8bb0\u5fc6\u5355\u5143\u7684\u66f4\u591a\u4fe1\u606f":87,"\u8bf7\u5728\u8be5pull":61,"\u8bf7\u60a8\u6bcf\u6b21\u63d0\u4ea4\u4ee3\u7801\u65f6":61,"\u8bf7\u60a8\u9075\u5b88\u4ee5\u4e0b\u7ea6\u5b9a":61,"\u8bf7\u628a\u6587\u6863\u91cc\u547d\u4ee4\u4e2d\u7684paddlepaddl":49,"\u8bf7\u6307\u5b9a\u8be5\u76ee\u5f55":70,"\u8bf7\u663e\u793a\u5730\u8c03\u7528":63,"\u8bf7\u67e5\u770b":80,"\u8bf7\u6ce8\u610f":[49,59,63,74,80],"\u8bf7\u6ce8\u610f\u6bcf\u4e2acommit\u7684\u540d\u79f0":61,"\u8bf7\u6ce8\u610f\u8fd9\u4e2a\u547d\u4ee4\u7ed3\u5c3e\u5904\u7684":60,"\u8bf7\u6ce8\u610fcommit\u7684\u6570\u91cf":61,"\u8bf7\u76f4\u63a5\u586b\u51450":46,"\u8bf7\u770b\u4e0b\u9762\u7684\u4f8b\u5b50":72,"\u8bf7\u786e\u4fdd":61,"\u8bf7\u7ed9\u51fa\u603b\u4f53\u7684\u4fee\u6539\u60c5\u51b5":61,"\u8bf7\u7ed9\u51fa\u60a8\u81ea\u5df1\u7684\u53cd\u9a73\u7406\u7531":61,"\u8bf7\u8bb0\u4f4f":68,"\u8bf7\u8fd0\u884c":49,"\u8bf7\u9009\u62e9\u5408\u9002\u7684\u8bcd\u6c47":61,"\u8bf7\u9009\u62e9\u6b63\u786e\u7684\u7248\u672c":46,"\u8bf7\u9075\u5b88":61,"\u8bf7\u91c7\u7528":61,"\u8bf8\u5982\u56fe\u50cf\u5206\u7c7b":72,"\u8bfb\u53d612\u4e2a\u91c7\u6837\u6570\u636e\u8fdb\u884c\u968f\u673a\u68af\u5ea6\u8ba1\u7b97\u6765\u66f4\u65b0\u66f4\u65b0":47,"\u8bfb\u53d6\u6570\u636e":2,"\u8bfb\u53d6\u6bcf\u4e00\u884c":2,"\u8bfb\u53d6volume\u4e2d\u7684\u6570\u636e\u8fdb\u884c\u8fd9\u6b21\u5206\u5e03\u5f0f\u8bad\u7ec3":75,"\u8bfb\u8005\u53ef\u4ee5\u67e5\u770b":75,"\u8bfb\u8005\u9700\u8981\u66ff\u6362\u6210\u81ea\u5df1\u4f7f\u7528\u7684\u4ed3\u5e93\u5730\u5740":75,"\u8c03\u7528":[62,63,81],"\u8c03\u7528\u4e00\u6b21":2,"\u8c03\u7528\u4e0a\u9762\u7684process\u51fd\u6570\u83b7\u5f97\u89c2\u6d4b\u6570\u636e":47,"\u8c03\u7528\u5bf9\u5e94":64,"\u8c03\u7528\u65b9\u6cd5\u89c1c":53,"\u8c03\u7528\u7684\u51fd\u6570\u662f\u5426\u652f\u6301\u4e0d\u540c\u8bbe\u5907":63,"\u8c03\u7528\u7684pydataprovider2\u662f":2,"\u8c03\u7528\u7b2c\u4e8c\u6b21\u7684\u65f6\u5019":2,"\u8c03\u7528\u8be5\u51fd\u6570\u540e":62,"\u8c03\u7528\u8fd9\u4e2apydataprovider2\u7684\u65b9\u6cd5":2,"\u8c13\u8bcd\u4e0a\u4e0b\u6587":87,"\u8d1f\u6837\u672c":84,"\u8d1f\u9762\u7684\u8bc4\u8bba\u7684\u5f97\u5206\u5c0f\u4e8e\u7b49\u4e8e4":88,"\u8d1f\u9762\u8bc4\u4ef7\u6837\u672c":88,"\u8d21\u732e\u6587\u6863":65,"\u8d44\u6e90\u5bf9\u8c61\u7684\u540d\u5b57\u662f\u552f\u4e00\u7684":73,"\u8d77":56,"\u8def\u5f84\u4e0b":[47,82],"\u8df3\u8f6c\u5230":61,"\u8df3\u8fc7":46,"\u8f6c\u4e3ajpeg\u6587\u4ef6\u5e76\u5b58\u5165\u7279\u5b9a\u7684\u76ee\u5f55":81,"\u8f6c\u6362\u751f\u6210\u7684\u6587\u4ef6\u540d\u4f1a\u662f\u4ee5\u4e0b\u683c\u5f0f":21,"\u8f6c\u6362\u8fc7\u6765\u7684":82,"\u8f6e":86,"\u8f83":56,"\u8f93\u5165":[55,59],"\u8f93\u5165\u548c\u8f93\u51fa\u90fd\u662f\u5355\u5c42\u5e8f\u5217":58,"\u8f93\u5165\u548c\u8f93\u51fa\u90fd\u662f\u53cc\u5c42\u5e8f\u5217":58,"\u8f93\u5165\u56fe\u7247\u7684\u9ad8\u5ea6\u53ca\u5bbd\u5ea6":81,"\u8f93\u5165\u5c42\u5c3a\u5bf8":82,"\u8f93\u5165\u5e8f\u5217\u4e2d\u5143\u7d20\u7684\u603b\u6570":46,"\u8f93\u5165\u6570\u636e\u4e3a\u4e00\u4e2a\u5b8c\u6574\u7684\u65f6\u95f4\u5e8f\u5217":56,"\u8f93\u5165\u6570\u636e\u4e3a\u5728\u5355\u5c42rnn\u6570\u636e\u91cc\u9762":56,"\u8f93\u5165\u6570\u636e\u6574\u4f53\u4e0a\u662f\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217":56,"\u8f93\u5165\u6570\u636e\u7684\u5b57\u5178\u7ef4\u6570\u662f1\u767e\u4e07":72,"\u8f93\u5165\u6570\u6910\u6240\u5728\u76ee\u5f55":88,"\u8f93\u5165\u6587\u672c":80,"\u8f93\u5165\u6587\u672c\u4e2d\u6ca1\u6709\u5934\u90e8":80,"\u8f93\u5165\u662f\u5426\u662f\u8f6c\u7f6e\u7684":62,"\u8f93\u5165\u662f\u7531\u4e00\u4e2alist\u4e2d\u7684\u7f51\u7edc\u5c42\u5b9e\u4f8b\u7684\u540d\u5b57\u7ec4\u6210\u7684":62,"\u8f93\u5165\u7279\u5f81\u56fe\u7684\u901a\u9053\u6570\u76ee":82,"\u8f93\u5165\u7684":80,"\u8f93\u5165\u7684\u539f\u59cb\u6570\u636e\u96c6\u8def\u5f84":89,"\u8f93\u5165\u7684\u540d\u5b57":62,"\u8f93\u5165\u7684\u5927\u5c0f":62,"\u8f93\u5165\u7684\u6587\u672c\u683c\u5f0f\u5982\u4e0b":80,"\u8f93\u5165\u7684\u6587\u672c\u8bcd\u5411\u91cf\u6a21\u578b\u540d\u79f0":80,"\u8f93\u5165\u7684\u7c7b\u578b":62,"\u8f93\u5165\u8bcd\u5178":63,"\u8f93\u5165\u95e8":88,"\u8f93\u5165\u9884\u6d4b\u6837\u672c":88,"\u8f93\u5165n\u4e2a\u5355\u8bcd":84,"\u8f93\u51fa":[55,59,63],"\u8f93\u51fa\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":58,"\u8f93\u51fa\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":58,"\u8f93\u51fa\u4e3an\u4e2aword_dim\u7ef4\u5ea6\u5411\u91cf":84,"\u8f93\u51fa\u51fd\u6570":59,"\u8f93\u51fa\u5e8f\u5217\u7684\u7c7b\u578b":55,"\u8f93\u51fa\u5e8f\u5217\u7684\u8bcd\u8bed\u6570\u548c\u8f93\u5165\u5e8f\u5217\u4e00\u81f4":58,"\u8f93\u51fa\u5e94\u8be5\u7c7b\u4f3c\u5982\u4e0b":86,"\u8f93\u51fa\u6587\u4ef6\u7684\u683c\u5f0f\u8bf4\u660e":80,"\u8f93\u51fa\u65e5\u5fd7\u4fdd\u5b58\u5728\u8def\u5f84":88,"\u8f93\u51fa\u65e5\u5fd7\u8bf4\u660e\u5982\u4e0b":88,"\u8f93\u51fa\u67092\u5217":80,"\u8f93\u51fa\u7279\u5f81\u56fe\u7684\u901a\u9053\u6570\u76ee":82,"\u8f93\u51fa\u7684\u4e8c\u8fdb\u5236\u8bcd\u5411\u91cf\u6a21\u578b\u540d\u79f0":80,"\u8f93\u51fa\u7684\u6587\u672c\u6a21\u578b\u540d\u79f0":80,"\u8f93\u51fa\u7684\u68af\u5ea6":70,"\u8f93\u51fa\u76ee\u5f55":82,"\u8f93\u51fa\u7ed3\u679c\u53ef\u80fd\u4f1a\u968f\u7740\u5bb9\u5668\u7684\u6d88\u8017\u800c\u88ab\u5220\u9664":74,"\u8fbe\u5230\u5bb9\u707e\u7684\u76ee\u7684":20,"\u8fc7\u4e86\u4e00\u4e2a\u5f88\u7b80\u5355\u7684recurrent_group":56,"\u8fc7\u5b8c\u6240\u6709\u8bad\u7ec3\u6570\u636e\u5373\u4e3a\u4e00\u4e2apass":46,"\u8fd0":49,"\u8fd0\u884c\u4e00\u4e2a":60,"\u8fd0\u884c\u4e0b\u9762\u547d\u4ee4\u5373\u53ef":86,"\u8fd0\u884c\u4e0b\u9762\u547d\u4ee4\u53ef\u4ee5\u8fdb\u884c\u7f16\u8bd1":63,"\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4\u6765\u4e0b\u8f7d\u8fd9\u4e2a\u811a\u672c":89,"\u8fd0\u884c\u4ee5\u4e0b\u6307\u4ee4":49,"\u8fd0\u884c\u4ee5\u4e0b\u7684\u547d\u4ee4\u4e0b\u8f7d\u548c\u83b7\u53d6\u6211\u4eec\u7684\u5b57\u5178\u548c\u9884\u8bad\u7ec3\u6a21\u578b":80,"\u8fd0\u884c\u4ee5\u4e0b\u7684\u547d\u4ee4\u4e0b\u8f7d\u6570\u636e\u96c6":80,"\u8fd0\u884c\u4ee5\u4e0b\u8bad\u7ec3\u547d\u4ee4":47,"\u8fd0\u884c\u4ee5\u53ca\u53d1\u5e03\u60a8\u7684ai\u7a0b\u5e8f":49,"\u8fd0\u884c\u5206\u5e03\u5f0f\u4f5c\u4e1a":68,"\u8fd0\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3":[46,66,84],"\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5":[49,60],"\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5\u6d4b\u65f6\u9700\u8981\u7f16\u8bd1\u6574\u4e2a\u5de5\u7a0b":63,"\u8fd0\u884c\u5931\u8d25":72,"\u8fd0\u884c\u5b8c\u4ee5\u4e0a\u547d\u4ee4":80,"\u8fd0\u884c\u5b8c\u6210\u540e":68,"\u8fd0\u884c\u5b8c\u811a\u672c":88,"\u8fd0\u884c\u6210\u529f\u4ee5\u540e":80,"\u8fd0\u884c\u6210\u529f\u540e\u76ee\u5f55":88,"\u8fd0\u884c\u65e5\u5fd7":68,"\u8fd0\u884c\u7684\u4e00\u4e9b\u53c2\u6570\u901a\u8fc7\u8fd9\u79cd\u65b9\u5f0f\u4f20\u9012\u5230\u5bb9\u5668\u5185":75,"\u8fd1":56,"\u8fd1\u671f\u63d0\u51fa\u7684nmt\u6a21\u578b\u901a\u5e38\u90fd\u5c5e\u4e8e\u7f16\u89e3\u7801\u6a21\u578b":89,"\u8fd4\u56de":[7,8,9,12,14],"\u8fd4\u56de0":2,"\u8fd4\u56de8\u4e2a\u7279\u5f81list\u548c1\u4e2a\u6807\u7b7elist":87,"\u8fd4\u56de\u4e00\u6761\u5b8c\u6574\u7684\u6837\u672c":2,"\u8fd4\u56de\u6570\u636e\u7684\u6bcf\u4e00\u6761\u6837\u672c\u7ed9":86,"\u8fd4\u56de\u65f6":2,"\u8fd4\u56de\u7684\u662f":[2,51],"\u8fd4\u56de\u7684\u987a\u5e8f\u9700\u8981\u548cinput_types\u4e2d\u5b9a\u4e49\u7684\u987a\u5e8f\u4e00\u81f4":2,"\u8fd4\u56de\u7b2c\u4e8c\u6b65":42,"\u8fd4\u56de\u7b2ci\u4e2a\u8f93\u5165\u77e9\u9635":62,"\u8fd4\u56de\u7c7b\u578b":[7,8,9,12,14],"\u8fd4\u56depython\u7aef\u7684\u8ba1\u7b97\u7ed3\u679c":63,"\u8fd8\u4f1a":56,"\u8fd8\u4f1a\u5f3a\u5236\u8bbe\u7f6e\u4e00\u4e9bpaddlepaddle\u53c2\u6570\u7684\u503c":53,"\u8fd8\u4f1a\u8f93\u51fa\u4e00\u4e2a":61,"\u8fd8\u53ef\u4ee5\u901a\u8fc7\u51cf\u5c0f\u5b66\u4e60\u5f8b\u6216\u8005\u5bf9\u6570\u636e\u8fdb\u884c\u5f52\u4e00\u5316\u5904\u7406\u6765\u89e3\u51b3\u8fd9\u7c7b\u95ee\u9898":46,"\u8fd8\u662f":56,"\u8fd8\u662f\u4ece":21,"\u8fd8\u662f\u865a\u62df\u673a":60,"\u8fd8\u6709":56,"\u8fd8\u80fd\u5904\u7406\u5176\u4ed6\u7528\u6237\u81ea\u5b9a\u4e49\u7684\u6570\u636e":88,"\u8fd8\u91c7\u7528\u4e86\u4e24\u4e2a\u5176\u4ed6\u7279\u5f81":87,"\u8fd8\u9700\u8981\u8fdb\u884c\u9884\u5904\u7406":81,"\u8fd8\u9700\u8981\u91cd\u5199":63,"\u8fd9":[46,56,84],"\u8fd98\u79cdlearning_rate_schedule\u53ca\u5176\u5bf9\u5e94\u5b66\u4e60\u7387\u8ba1\u7b97\u65b9\u5f0f\u5982\u4e0b":46,"\u8fd9\u4e00\u5757\u7684\u8017\u65f6\u6bd4\u4f8b\u771f\u7684\u592a\u9ad8":67,"\u8fd9\u4e00\u5c42\u8fdb\u884c\u5c01\u88c5":34,"\u8fd9\u4e00\u6982\u5ff5\u4e0d\u518d\u7410\u788e":34,"\u8fd9\u4e00\u8ba1\u7b97\u5355\u5143":46,"\u8fd9\u4e00\u8fc7\u7a0b\u5bf9\u7528\u6237\u662f\u5b8c\u5168\u900f\u660e\u7684":58,"\u8fd9\u4e09\u4e2a\u5206\u652f":42,"\u8fd9\u4e09\u4e2a\u6b65\u9aa4\u53ef\u914d\u7f6e\u4e3a":84,"\u8fd9\u4e0e\u672c\u5730\u8bad\u7ec3\u76f8\u540c":68,"\u8fd9\u4e24\u4e2a\u6587\u4ef6\u5939\u4e0b\u5404\u81ea\u670910\u4e2a\u5b50\u6587\u4ef6\u5939":81,"\u8fd9\u4e24\u4e2a\u6807\u51c6":87,"\u8fd9\u4e2a":[56,60,73],"\u8fd9\u4e2a\u4efb\u52a1\u7684\u914d\u7f6e\u4e3a":46,"\u8fd9\u4e2a\u4efb\u52a1\u7684dataprovider\u4e3a":46,"\u8fd9\u4e2a\u4f8b\u5b50\u6709\u4e24\u5904\u4e0d\u540c":63,"\u8fd9\u4e2a\u51fd\u6570\u7684":59,"\u8fd9\u4e2a\u51fd\u6570\u8fdb\u884c\u53d8\u6362":56,"\u8fd9\u4e2a\u51fd\u6570\u9700\u8981\u8bbe\u7f6e":59,"\u8fd9\u4e2a\u52a8\u6001\u5e93\u7684\u8fde\u63a5\u53c2\u6570\u4e0epaddle\u7684\u5176\u4ed6\u4e8c\u8fdb\u5236":34,"\u8fd9\u4e2a\u53c2\u6570\u4e5f\u4e0d\u4f1a\u4e00\u5e76\u5220\u9664":34,"\u8fd9\u4e2a\u5411\u91cf\u4e0e\u6e90\u4e2d\u641c\u7d22\u51fa\u7684\u4f4d\u7f6e\u548c\u6240\u6709\u4e4b\u524d\u751f\u6210\u7684\u76ee\u6807\u5355\u8bcd\u6709\u5173":89,"\u8fd9\u4e2a\u5730\u5740\u5219\u4e3a\u5b83\u7684\u7edd\u5bf9\u8def\u5f84\u6216\u76f8\u5bf9\u8def\u5f84":1,"\u8fd9\u4e2a\u5730\u5740\u6765\u8868\u793a\u6b64\u6b65\u9aa4\u6240\u6784\u5efa\u51fa\u7684\u955c\u50cf":75,"\u8fd9\u4e2a\u57fa\u7c7b":62,"\u8fd9\u4e2a\u5934\u6587\u4ef6\u4e0d\u5047\u8bbe\u5176\u4ed6\u6587\u4ef6\u7684\u5f15\u7528\u987a\u5e8f":34,"\u8fd9\u4e2a\u5b57\u5178\u662f\u6574\u6570\u6807\u7b7e\u548c\u5b57\u7b26\u4e32\u6807\u7b7e\u7684\u4e00\u4e2a\u5bf9\u5e94":88,"\u8fd9\u4e2a\u5e8f\u5217\u7684\u6bcf\u4e2a\u5143\u7d20\u53c8\u662f\u4e00\u4e2a\u5e8f\u5217":58,"\u8fd9\u4e2a\u63a5\u53e3\u9700\u8981\u505a\u5230":33,"\u8fd9\u4e2a\u6570\u636e\u4e5f\u88ab\u5355\u5c42rnn\u7f51\u7edc\u76f4\u63a5\u4f7f\u7528":56,"\u8fd9\u4e2a\u6570\u91cf\u79f0\u4e3abeam":89,"\u8fd9\u4e2a\u6587\u4ef6\u5177\u6709\u72ec\u7279\u7684\u8bed\u6cd5":33,"\u8fd9\u4e2a\u663e\u793a\u5668\u5f88\u68d2":84,"\u8fd9\u4e2a\u6a21\u578b\u5bf9\u4e8e\u7f16\u89e3\u7801\u6a21\u578b\u6765\u8bf4":89,"\u8fd9\u4e2a\u76ee\u5f55\u4e2d\u9664\u4e86":34,"\u8fd9\u4e2a\u795e\u7ecf\u7f51\u7edc\u5355\u5143\u5c31\u53ebmemori":56,"\u8fd9\u4e2a\u7a0b\u5e8f\u662f\u60a8\u5728\u5f00\u53d1\u673a\u4e0a\u4f7f\u7528\u5f00\u53d1\u955c\u50cf\u5b8c\u6210\u5f00\u53d1":49,"\u8fd9\u4e2a\u7c7b\u7684\u53c2\u6570\u5305\u62ec":62,"\u8fd9\u4e2a\u7c7b\u9700\u8981\u7ee7\u627f":62,"\u8fd9\u4e2a\u7cfb\u7edf\u5c06srl\u4efb\u52a1\u89c6\u4e3a\u5e8f\u5217\u6807\u6ce8\u95ee\u9898":87,"\u8fd9\u4e2a\u7ed3\u6784\u4f53\u4e2d\u7684\u53e6\u4e00\u4e2a\u9879\u76ee\u662f":34,"\u8fd9\u4e2a\u7ed3\u6784\u4f53\u5305\u542b\u4e24\u4e2a\u9879\u76ee":34,"\u8fd9\u4e2a\u811a\u672c\u8c03\u7528":60,"\u8fd9\u4e2a\u8282\u70b9\u53ef\u4ee5\u662f\u7269\u7406\u673a\u6216\u8005\u865a\u62df\u673a":73,"\u8fd9\u4e2a\u8868\u683c":73,"\u8fd9\u4e2a\u8f93\u5165\u4e0d\u53c2\u4e0e":63,"\u8fd9\u4e2a\u8fc7\u7a0b\u5bf9\u7528\u6237\u4e5f\u662f\u900f\u660e\u7684":58,"\u8fd9\u4e2a\u8fc7\u7a0b\u5c31\u662f\u6a21\u578b\u7684\u8bad\u7ec3\u8fc7\u7a0b":47,"\u8fd9\u4e2a\u8fc7\u7a0b\u9664\u4e86\u7f16\u8bd1paddlepaddle\u4e3a":61,"\u8fd9\u4e2a\u914d\u7f6e\u4e0e":80,"\u8fd9\u4e2a\u914d\u7f6e\u6587\u4ef6":73,"\u8fd9\u4e2a\u914d\u7f6e\u662f\u5426\u7528\u6765\u751f\u6210":89,"\u8fd9\u4e2a\u955c\u50cf\u5305\u542b\u4e86paddle\u76f8\u5173\u7684\u5f00\u53d1\u5de5\u5177\u4ee5\u53ca\u7f16\u8bd1\u548c\u8fd0\u884c\u73af\u5883":49,"\u8fd9\u4e2a\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u751f\u6210\u4e00\u7cfb\u5217\u6743\u91cd":59,"\u8fd9\u4e2a\u95ee\u9898\u662fpydataprovider\u8bfb\u6570\u636e\u65f6\u5019\u7684\u903b\u8f91\u95ee\u9898":2,"\u8fd9\u4e2a\u9759\u6001\u5e93\u5305\u542b\u4e86paddle\u7684\u5168\u90e8\u7b26\u53f7":34,"\u8fd9\u4e2adataprovider\u8f83\u590d\u6742":2,"\u8fd9\u4e2ainstance\u53ef\u4ee5\u662f\u5355\u4e2a\u503c":21,"\u8fd9\u4e2aissu":60,"\u8fd9\u4e2ajob\u624d\u7b97\u6210\u529f\u7ed3\u675f":75,"\u8fd9\u4e2alayer\u7684\u8f93\u51fa\u4f1a\u4f5c\u4e3a\u6574\u4e2a":58,"\u8fd9\u4e5f\u4f1a\u6781\u5927\u51cf\u5c11\u6570\u636e\u8bfb\u5165\u7684\u8017\u65f6":46,"\u8fd9\u4e9b":68,"\u8fd9\u4e9b\u51fd\u6570\u4f1a\u5c06\u5bf9\u5e94\u5185\u5bb9\u6dfb\u52a0\u5230":63,"\u8fd9\u4e9b\u5206\u5e03\u5f0f\u5b58\u50a8\u670d\u52a1\u901a\u5e38\u4f1a\u628a\u6570\u636e\u5207\u5272\u6210\u591a\u4e2a\u5206\u7247\u5206\u5e03\u5f0f\u7684\u5b58\u50a8\u5728\u591a\u4e2a\u8282\u70b9\u4e4b\u4e0a":21,"\u8fd9\u4e9b\u53c2\u6570\u7684\u5177\u4f53\u63cf\u8ff0":75,"\u8fd9\u4e9b\u53c2\u6570\u7684\u7b80\u77ed\u4ecb\u7ecd\u5982\u4e0b":86,"\u8fd9\u4e9b\u540d\u5b57\u5fc5\u987b\u8981\u5199\u5bf9":62,"\u8fd9\u4e9b\u6570\u636e\u4f1a\u88ab\u7528\u6765\u66f4\u65b0\u53c2\u6570":46,"\u8fd9\u4e9b\u6570\u636e\u4f7f\u7528\u7684\u5185\u5b58\u4e3b\u8981\u548c\u4e24\u4e2a\u53c2\u6570\u6709\u5173\u7cfb":46,"\u8fd9\u4e9b\u6587\u4ef6\u5c06\u4f1a\u88ab\u4fdd\u5b58\u5728":82,"\u8fd9\u4e9b\u6a21\u578b\u90fd\u662f\u7531\u539f\u4f5c\u8005\u63d0\u4f9b\u7684\u6a21\u578b":82,"\u8fd9\u4e9b\u7279\u5f81\u503c\u4e0e\u4e0a\u8ff0\u4f7f\u7528c":82,"\u8fd9\u4e9b\u7279\u5f81\u548c\u6807\u7b7e\u5b58\u50a8\u5728":87,"\u8fd9\u4e9b\u7279\u5f81\u6570\u636e\u4e4b\u95f4\u7684\u987a\u5e8f\u662f\u6709\u610f\u4e49\u7684":56,"\u8fd9\u4efd\u6559\u7a0b\u5c55\u793a\u4e86\u5982\u4f55\u5728paddlepaddle\u4e2d\u5b9e\u73b0\u4e00\u4e2a\u81ea\u5b9a\u4e49\u7684\u7f51\u7edc\u5c42":62,"\u8fd9\u4efd\u7b80\u77ed\u7684\u4ecb\u7ecd\u5c06\u5411\u4f60\u5c55\u793a\u5982\u4f55\u5229\u7528paddlepaddle\u6765\u89e3\u51b3\u4e00\u4e2a\u7ecf\u5178\u7684\u7ebf\u6027\u56de\u5f52\u95ee\u9898":47,"\u8fd9\u4f1a\u63d0\u793a\u5f53\u524d\u76ee\u5f55\u7684\u4e00\u4e9b\u53d8\u5316":61,"\u8fd9\u4f1a\u7ed9\u8bc4\u5ba1\u4eba\u5e26\u6765\u5f88\u5927\u56f0\u6270":61,"\u8fd9\u4f1a\u81ea\u52a8\u8fdb\u884c\u7f51\u7edc\u914d\u7f6e\u4e2d\u58f0\u660e\u7684\u6fc0\u6d3b\u64cd\u4f5c":62,"\u8fd9\u4f7f\u5f97nmt\u6a21\u578b\u5f97\u4ee5\u89e3\u653e\u51fa\u6765":89,"\u8fd9\u4fbf\u662f\u4e00\u79cd\u53cc\u5c42rnn\u7684\u8f93\u5165\u6570\u636e":56,"\u8fd9\u51e0\u4e2a\u7f16\u8bd1\u9009\u9879\u7684\u8bbe\u7f6e":48,"\u8fd9\u53e5\u8868\u793a\u4f7f\u7528\u57fa\u7c7b":63,"\u8fd9\u53ef\u4ee5\u5e2e\u60a8\u7701\u6389\u82b1\u4e00\u5c0f\u65f6\u5b89\u88c5\u548c\u914d\u7f6e\u5404\u79cd\u5f00\u53d1\u5de5\u5177":60,"\u8fd9\u53ef\u4ee5\u8ba9\u5176\u4ed6\u4eba\u77e5\u9053\u8fd9\u6b21\u63d0\u4ea4\u505a\u4e86\u54ea\u4e9b\u6539\u53d8":61,"\u8fd9\u53ef\u4ee5\u901a\u8fc7":61,"\u8fd9\u548c\u5355\u5c42rnn\u7684\u914d\u7f6e\u662f\u7b49\u4ef7\u7684":56,"\u8fd9\u56db\u4e2a\u7b80\u5355\u7684\u7279\u5f81\u662f\u6211\u4eec\u7684srl\u7cfb\u7edf\u6240\u9700\u8981\u7684":87,"\u8fd9\u56db\u6761\u6570\u636e\u540c\u65f6\u5904\u7406\u7684\u53e5\u5b50\u6570\u91cf\u4e3a":56,"\u8fd9\u5728\u5f88\u5927\u7a0b\u5ea6\u4e0a\u4f18\u4e8e\u5148\u524d\u7684\u6700\u5148\u8fdb\u7684\u7cfb\u7edf":87,"\u8fd9\u5728\u6784\u9020\u975e\u5e38\u590d\u6742\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u65f6\u662f\u6709\u7528\u7684":59,"\u8fd9\u5bf9\u4e8e\u901a\u5e38\u7684java\u7684\u5f00\u53d1\u8005\u6765\u8bf4":33,"\u8fd9\u5c06\u4f1a\u5bfc\u81f4\u5355\u5143\u6d4b\u8bd5\u51fa\u9519":63,"\u8fd9\u5c06\u4f1a\u5bfc\u81f4\u7f16\u8bd1\u51fa\u9519":63,"\u8fd9\u5c06\u82b1\u8d39\u6570\u5206\u949f\u7684\u65f6\u95f4":89,"\u8fd9\u5df2\u7ecf\u5728":88,"\u8fd9\u610f\u5473\u7740":59,"\u8fd9\u610f\u5473\u7740\u6a21\u578b\u5728\u8bad\u7ec3\u6570\u636e\u4e0a\u4e0d\u65ad\u7684\u6539\u8fdb":47,"\u8fd9\u610f\u5473\u7740\u9664\u4e86\u6307\u5b9adevic":72,"\u8fd9\u65f6":46,"\u8fd9\u65f6\u5728\u4f7f\u7528":46,"\u8fd9\u65f6\u8fdb\u884c\u77e9\u9635\u4e58\u6cd5\u8fd0\u7b97\u5c31\u53ef\u80fd\u5bfc\u81f4\u6d6e\u70b9\u6570\u6ea2\u51fa":46,"\u8fd9\u662f\u4e00\u4e2a\u57fa\u4e8e\u7edf\u8ba1\u7684\u673a\u5668\u7ffb\u8bd1\u7cfb\u7edf":88,"\u8fd9\u662f\u4e00\u79cd\u6309\u5df2\u8bad\u7ec3\u6837\u672c\u6570\u5206\u6bb5\u53d6\u503c\u7684\u5b66\u4e60\u7387\u9000\u706b\u65b9\u6cd5":46,"\u8fd9\u662f\u4e00\u79cd\u6309\u5df2\u8bad\u7ec3pass\u6570\u5206\u6bb5\u53d6\u503c\u7684\u5b66\u4e60\u7387\u9000\u706b\u65b9\u6cd5":46,"\u8fd9\u662f\u4e00\u79cd\u975e\u5e38\u7075\u6d3b\u7684\u6570\u636e\u7ec4\u7ec7\u65b9\u5f0f":55,"\u8fd9\u662f\u56e0\u4e3a":33,"\u8fd9\u662f\u56e0\u4e3a\u5b83\u53d1\u6398\u51fa\u4e86\u56fe\u7247\u7684\u4e24\u7c7b\u91cd\u8981\u4fe1\u606f":81,"\u8fd9\u662f\u5f00\u6e90\u793e\u533a\u7684\u57fa\u672c\u793c\u8c8c":61,"\u8fd9\u662f\u666e\u901a\u7684\u5355\u5c42\u65f6\u95f4\u5e8f\u5217\u7684dataprovider\u4ee3\u7801":56,"\u8fd9\u662f\u76ee\u524dcmake\u5bfb\u627epython\u7684\u903b\u8f91\u5b58\u5728\u7f3a\u9677":46,"\u8fd9\u662f\u96c6\u675f\u641c\u7d22\u7684\u7ed3\u679c":89,"\u8fd9\u6765\u81ea\u4e8epaddlepaddle\u7684\u5185\u5b58\u4e2d":89,"\u8fd9\u6837":[34,47,68],"\u8fd9\u6837\u4fdd\u8bc1":42,"\u8fd9\u6837\u4fdd\u8bc1\u8fd0\u884c\u7ed3\u675f\u4e4b\u540e\u7684":60,"\u8fd9\u6837\u505a\u53ef\u4ee5\u6781\u5927\u7684\u51cf\u5c11\u5185\u5b58\u5360\u7528":46,"\u8fd9\u6837\u53ef\u4ee5\u51cf\u5c0fgpu\u5185\u5b58":72,"\u8fd9\u6837\u5982\u679c\u9047\u5230\u95ee\u9898":60,"\u8fd9\u6837\u5bb9\u5668\u7684":75,"\u8fd9\u6837\u5c31\u4f1a\u751f\u6210\u4e24\u4e2a\u6587\u4ef6":86,"\u8fd9\u6837\u5c31\u53ef\u4ee5\u5728\u4e91\u7aef\u6267\u884c\u591a\u79cd\u6570\u636e\u7c7b\u8ba1\u7b97\u4efb\u52a1":21,"\u8fd9\u6837\u5df2\u7ecf\u4f20\u8f93\u6210\u529f\u7684\u90e8\u5206\u5c31\u4e0d\u7528\u91cd\u65b0\u4f20\u8f93\u4e86":28,"\u8fd9\u6837\u5f00\u53d1\u4eba\u5458\u53ef\u4ee5\u4ee5\u4e00\u81f4\u7684\u65b9\u5f0f\u5728\u4e0d\u540c\u7684\u5e73\u53f0\u4e0a\u5de5\u4f5c":49,"\u8fd9\u6837\u5f53\u8be5pull":61,"\u8fd9\u6837\u6211\u4eec\u53ef\u4ee5\u76f4\u63a5\u5728docker\u4e2d\u8fd0\u884c\u9700\u8981\u7684\u7a0b\u5e8f\u800c\u4e0d\u9700\u8981\u5b89\u88c5\u540e\u5728\u6267\u884c":49,"\u8fd9\u6837\u7684\u88c5\u9970\u5668":62,"\u8fd9\u6837\u7684\u8bdd":74,"\u8fd9\u6837\u7684\u8bdd\u6bcf\u4f4d\u7528\u6237\u5728\u6d4b\u8bd5\u6587\u4ef6\u4e2d\u5c06\u4e0e\u8bad\u7ec3\u6587\u4ef6\u542b\u6709\u540c\u6837\u7684\u4fe1\u606f":86,"\u8fd9\u6b63\u662f\u5b83\u4eec\u901f\u5ea6\u5feb\u7684\u539f\u56e0":67,"\u8fd9\u6bb5\u7b80\u77ed\u7684\u914d\u7f6e\u5c55\u793a\u4e86paddlepaddle\u7684\u57fa\u672c\u7528\u6cd5":47,"\u8fd9\u7528\u4e8e\u5728\u591a\u7ebf\u7a0b\u548c\u591a\u673a\u4e0a\u66f4\u65b0\u53c2\u6570":62,"\u8fd9\u79cd\u521d\u59cb\u5316\u65b9\u5f0f\u5728\u4e00\u822c\u60c5\u51b5\u4e0b\u4e0d\u4f1a\u4ea7\u751f\u5f88\u5dee\u7684\u7ed3\u679c":46,"\u8fd9\u79cd\u60c5\u51b5\u4e0b\u4e0d\u9700\u8981\u91cd\u5199\u8be5\u51fd\u6570":62,"\u8fd9\u79cd\u60c5\u51b5\u5e38\u5e38\u53d1\u751f\u5728":46,"\u8fd9\u79cd\u65b9\u5f0f\u5bf9\u5185\u5b58\u6d88\u8017\u8f83\u5927":46,"\u8fd9\u79cd\u65b9\u5f0f\u5fc5\u987b\u4f7f\u7528paddle\u5b58\u50a8\u7684\u6a21\u578b\u8def\u5f84\u683c\u5f0f":72,"\u8fd9\u79cd\u751f\u6210\u6280\u672f\u53ea\u7528\u4e8e\u7c7b\u4f3c\u89e3\u7801\u5668\u7684\u751f\u6210\u8fc7\u7a0b":59,"\u8fd9\u79cd\u7c7b\u578b\u7684\u8f93\u5165\u5fc5\u987b\u901a\u8fc7":58,"\u8fd9\u79cd\u96c6\u7fa4\u8282\u70b9\u7ba1\u7406\u65b9\u5f0f\u4f1a\u5728\u5c06\u6765\u4f7f\u7528":75,"\u8fd9\u7bc7\u6587\u6863":61,"\u8fd9\u7bc7\u6587\u6863\u4ecb\u7ecd\u57fa\u4e8e":60,"\u8fd9\u7bc7\u6587\u6863\u5c06\u4ee5linux":54,"\u8fd9\u7bc7\u6587\u7ae0":[60,89],"\u8fd9\u7ec4\u8bed\u4e49\u76f8\u540c\u7684\u793a\u4f8b\u914d\u7f6e\u5982\u4e0b":56,"\u8fd9\u901a\u8fc7\u83b7\u5f97\u53cd\u5411\u5faa\u73af\u7f51\u7edc\u7684\u7b2c\u4e00\u4e2a\u5b9e\u4f8b":59,"\u8fd9\u90fd\u9700\u8981\u8fd9\u4e2a\u63a5\u53e3\u6309\u7167\u7ea6\u5b9a\u4fd7\u6210\u7684\u89c4\u5219\u6765\u6ce8\u91ca\u5b8c\u5907":33,"\u8fd9\u91cc":[46,49,59,61,73,75,82,87],"\u8fd9\u91cc\u4e0d\u518d\u8d58\u8ff0":63,"\u8fd9\u91cc\u4ecb\u7ecd\u5982\u751f\u6210\u9020\u8fd9\u4e2a\u5f00\u53d1\u955c\u50cf":49,"\u8fd9\u91cc\u4ee5":84,"\u8fd9\u91cc\u4f7f\u7528\u4e00\u4e2a\u57fa\u4e8emomentum\u7684\u968f\u673a\u68af\u5ea6\u4e0b\u964d":47,"\u8fd9\u91cc\u4f7f\u7528\u4e86\u4e09\u79cd\u7f51\u7edc\u5355\u5143":47,"\u8fd9\u91cc\u4f7f\u7528\u4e86paddlepaddle\u7684python\u63a5\u53e3\u6765\u52a0\u8f7d\u6570\u6910":88,"\u8fd9\u91cc\u4f7f\u7528\u4e86paddlepaddle\u9884\u5b9a\u4e49\u597d\u7684rnn\u5904\u7406\u51fd\u6570":56,"\u8fd9\u91cc\u4f7f\u7528\u7b80\u5355\u7684":46,"\u8fd9\u91cc\u5229\u7528\u5b83\u5efa\u6a21\u53d8\u91cf\u4e4b\u95f4\u7684\u7ebf\u6027\u5173\u7cfb":47,"\u8fd9\u91cc\u53ea\u52a0\u8f7d":89,"\u8fd9\u91cc\u53ea\u7b80\u5355\u4ecb\u7ecd\u4e86\u5355\u673a\u8bad\u7ec3":84,"\u8fd9\u91cc\u5c06\u4ecb\u7ecdpaddlepaddle\u7684\u57fa\u672c\u4f7f\u7528\u6982\u5ff5":51,"\u8fd9\u91cc\u5c55\u793a\u5982\u4f55\u4f7f\u7528\u89c2\u6d4b\u6570\u636e\u6765\u62df\u5408\u8fd9\u4e00\u7ebf\u6027\u5173\u7cfb":47,"\u8fd9\u91cc\u6211\u4eec\u4f7f\u7528":86,"\u8fd9\u91cc\u6211\u4eec\u4f7f\u7528\u7684\u662f\u4e00\u4e2a\u5c0f\u7684vgg\u7f51\u7edc":81,"\u8fd9\u91cc\u6211\u4eec\u4f7f\u7528\u7684\u662fgpu\u6a21\u5f0f\u8fdb\u884c\u8bad\u7ec3":81,"\u8fd9\u91cc\u6211\u4eec\u5728movielens\u6570\u636e\u96c6\u63cf\u8ff0\u4e00\u79cd":86,"\u8fd9\u91cc\u6211\u4eec\u5c55\u793a\u4e00\u4efd\u7b80\u5316\u8fc7\u7684\u4ee3\u7801":62,"\u8fd9\u91cc\u6211\u4eec\u7528paddl":49,"\u8fd9\u91cc\u6211\u4eec\u901a\u8fc7\u5728kubernetes\u96c6\u7fa4\u4e0a\u542f\u52a8\u4e00\u4e2ajob\u6765\u4e0b\u8f7d\u5e76\u5207\u5272\u6570\u636e":75,"\u8fd9\u91cc\u6307\u5b9a\u8bcd\u5178":84,"\u8fd9\u91cc\u6570\u636e\u5c42\u6709\u4e24\u4e2a":47,"\u8fd9\u91cc\u662f\u4e00\u4e2a\u4f8b\u5b50":89,"\u8fd9\u91cc\u6709\u4e00\u4e9b\u4e0d\u540c\u7684\u53c2\u6570\u9700\u8981\u6307\u5b9a":89,"\u8fd9\u91cc\u6709\u4e24\u79cd\u6709\u6548\u7684\u89e3\u51b3\u65b9\u6cd5":46,"\u8fd9\u91cc\u68c0\u9a8c\u8fd0\u884c\u65f6\u95f4\u6a21\u578b\u7684\u6536\u655b":68,"\u8fd9\u91cc\u6bcf\u4e2a5\u4e2abatch\u6253\u5370\u4e00\u4e2a\u70b9":89,"\u8fd9\u91cc\u6bcf\u9694100\u4e2abatch\u663e\u793a\u4e00\u6b21\u53c2\u6570\u7edf\u8ba1\u4fe1\u606f":89,"\u8fd9\u91cc\u6bcf\u969410\u4e2abatch\u6253\u5370\u4e00\u6b21\u65e5\u5fd7":89,"\u8fd9\u91cc\u7684\u5217\u51fa\u7684\u548c":81,"\u8fd9\u91cc\u7684eigentensor\u4e4b\u95f4\u7684\u8fd0\u7b97\u53ea\u662f\u6539\u53d8\u4e86\u539f\u6709tensor\u4e2d\u7684\u6570\u636e":64,"\u8fd9\u91cc\u76f4\u63a5\u901a\u8fc7\u9884\u6d4b\u811a\u672c":84,"\u8fd9\u91cc\u7ed9\u51fa\u96c6\u4e2d\u5e38\u89c1\u7684\u90e8\u7f72\u65b9\u6cd5":73,"\u8fd9\u91cc\u8bbe\u7f6e\u4e3a\u4f7f\u7528cpu":89,"\u8fd9\u91cc\u8bbe\u7f6e\u4e3afals":89,"\u8fd9\u91cc\u8bbe\u7f6e\u4e3atrue":89,"\u8fd9\u91cc\u91c7\u7528adam\u4f18\u5316\u65b9\u6cd5":84,"\u8fd9\u91cc\u9700\u8981\u7528\u6237\u989d\u5916\u6ce8\u610f":20,"\u8fdb\u4e3b\u4ed3\u5e93\u540e":61,"\u8fdb\u5165":88,"\u8fdb\u5165\u5bb9\u5668":74,"\u8fdb\u5165\u5f00\u53d1\u955c\u50cf\u5e76\u5f00\u59cb\u5de5\u4f5c":49,"\u8fdb\u7a0b":49,"\u8fdb\u7a0b\u542f\u52a8\u7684\u5fc5\u8981\u53c2\u6570":75,"\u8fdb\u7a0b\u7684":68,"\u8fdb\u7a0b\u7684\u542f\u52a8\u53c2\u6570":75,"\u8fdb\u7a0b\u7684\u8fd0\u884c\u73af\u5883":75,"\u8fdb\u7a0b\u9700\u8981\u7684":75,"\u8fdb\u800c\u8fdb\u884c\u4ee3\u7801\u8bc4\u5ba1":42,"\u8fdb\u884c\u4e86":56,"\u8fdb\u884c\u4f7f\u7528":81,"\u8fdb\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3\u7684\u65b9\u6848":75,"\u8fdb\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3\u7684\u65b9\u6cd5":75,"\u8fdb\u884c\u56de\u590d":61,"\u8fdb\u884c\u5f00\u53d1":61,"\u8fdb\u884c\u62c6\u89e3":56,"\u8fdb\u884c\u6fc0\u6d3b\u64cd\u4f5c":62,"\u8fdb\u884c\u8bbe\u7f6e":63,"\u8fdb\u884c\u8bfb\u5165\u548c\u9884\u5904\u7406\u4ece\u800c\u5f97\u5230\u771f\u5b9e\u8f93\u5165":47,"\u8fdb\u884c\u9884\u6d4b":84,"\u8fdb\u9636\u6307\u5357":[51,78],"\u8fde\u63a5":58,"\u8fde\u63a5\u4e09\u4e2alstm\u9690\u85cf\u5c42":88,"\u9000\u4f11\u4eba\u5458":85,"\u9000\u51fa\u5bb9\u5668":74,"\u9002\u4e2d":56,"\u9002\u5408\u81ea\u5df1\u5b9e\u9a8c":49,"\u9009":56,"\u9009\u62e9":56,"\u9009\u62e9\u6d4b\u8bd5\u7ed3\u679c\u6700\u597d\u7684\u6a21\u578b\u6765\u9884\u6d4b":84,"\u9009\u62e9\u76ee\u6807\u5206\u652f":61,"\u9009\u62e9\u8def\u5f84\u6765\u52a8\u6001\u52a0\u8f7dnvidia":70,"\u9009\u62e9\u8fc7\u540e\u7684":89,"\u9009\u62e9\u9002\u5408\u60a8\u7684\u573a\u666f\u7684\u5408\u9002\u65b9\u6848":73,"\u9009\u81ea\u4e0b\u5217\u7c7b\u578b":85,"\u9009\u9879":[48,60,80],"\u900f\u4f20\u7528\u6237\u8eab\u4efd\u7684\u529e\u6cd5":28,"\u9012\u5f52\u795e\u7ecf\u7f51\u7edc":69,"\u901a\u5e38":[34,63,68,88],"\u901a\u5e38\u4f1a\u4f7f\u7528\u73af\u5883\u53d8\u91cf\u914d\u7f6ejob\u7684\u914d\u7f6e\u4fe1\u606f":75,"\u901a\u5e38\u4f7f\u7528\u7a00\u758f\u8bad\u7ec3\u6765\u52a0\u901f\u8ba1\u7b97\u8fc7\u7a0b":72,"\u901a\u5e38\u505a\u6cd5\u662f\u4ece\u4e00\u4e2a\u6bd4\u8f83\u5927\u7684learning_rate\u5f00\u59cb\u8bd5":46,"\u901a\u5e38\u540d\u5b57\u662f":61,"\u901a\u5e38\u60c5\u51b5\u4e0b":67,"\u901a\u5e38\u6211\u4eec\u4f1a\u5b89\u88c5ceph\u7b49\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u6765\u5b58\u50a8\u8bad\u7ec3\u6570\u636e":74,"\u901a\u5e38\u6307\u5c06\u4e00\u4e2a\u6574\u4f53\u62c6\u5206\u6210\u591a\u4efd\u7684\u5176\u4e2d\u7684\u4e00\u4efd":20,"\u901a\u5e38\u6bcf\u4e2ajob\u5305\u62ec\u4e00\u4e2a\u6216\u8005\u591a\u4e2apod":73,"\u901a\u5e38\u7684\u505a\u6cd5\u662f\u4f7f\u7528":59,"\u901a\u5e38\u7684\u505a\u6cd5\u662f\u5c06\u914d\u7f6e\u5b58\u4e8e":62,"\u901a\u5e38\u8981\u6c42\u65f6\u95f4\u6b65\u4e4b\u95f4\u5177\u6709\u4e00\u4e9b\u4f9d\u8d56\u6027":56,"\u901a\u5e38\u90fd\u4f1a\u4f7f\u7528\u4e0b\u9762\u8fd9\u4e9b\u547d\u4ee4\u884c\u53c2\u6570":72,"\u901a\u7528":69,"\u901a\u77e5":56,"\u901a\u8fc7":[46,56,61,62,63,68,84],"\u901a\u8fc7\u4e24\u4e2a\u5d4c\u5957\u7684":58,"\u901a\u8fc7\u4ea4\u53c9\u7f16\u8bd1\u7684\u65b9\u5f0f\u6765\u7f16\u8bd1":54,"\u901a\u8fc7\u4ea4\u66ff\u4f7f\u7528\u5377\u79ef\u548c\u6c60\u5316\u5904\u7406":81,"\u901a\u8fc7\u4f7f\u7528":48,"\u901a\u8fc7\u51fd\u6570":75,"\u901a\u8fc7\u5377\u79ef\u64cd\u4f5c\u4ece\u56fe\u7247\u6216\u7279\u5f81\u56fe\u4e2d\u63d0\u53d6\u7279\u5f81":81,"\u901a\u8fc7\u547d\u4ee4\u884c\u53c2\u6570":46,"\u901a\u8fc7\u5f15\u7528memory\u5f97\u5230\u8fd9\u4e2alayer\u4e0a\u4e00\u4e2a\u65f6\u523b\u7684\u8f93\u51fa":58,"\u901a\u8fc7\u5f15\u7528memory\u5f97\u5230\u8fd9\u4e2alayer\u4e0a\u4e00\u4e2a\u65f6\u523b\u8f93\u51fa":58,"\u901a\u8fc7\u6240\u6709\u5355\u5143\u6d4b\u8bd5":61,"\u901a\u8fc7\u6240\u6709\u8bad\u7ec3\u96c6\u4e00\u6b21\u79f0\u4e3a\u4e00\u904d":88,"\u901a\u8fc7\u67e5\u770b\u4e70\u5bb6\u5bf9\u67d0\u4e2a\u4ea7\u54c1\u7684\u8bc4\u4ef7\u53cd\u9988":84,"\u901a\u8fc7\u6a21\u578b\u63a8\u65adapi\u7684\u5b9e\u73b0\u4f5c\u4e3a\u4e00\u4e2a\u6837\u4f8b":34,"\u901a\u8fc7\u7ec4\u5408\u4e0d\u540c\u7684layer":51,"\u901a\u8fc7\u7f16\u8bd1\u4f1a\u751f\u6210py_paddle\u8f6f\u4ef6\u5305":4,"\u901a\u8fc7\u7f51\u7edc\u5c42\u7684\u6807\u8bc6\u7b26\u6765\u6307\u5b9a":62,"\u901a\u8fc7\u8be5\u53c2\u6570\u53ef\u83b7\u53d6\u5230\u8f93\u5165\u8f93\u51fa\u4ee5\u53ca\u5c5e\u6027":63,"\u901a\u8fc7\u8c03\u7528":4,"\u901a\u8fc7\u914d\u7f6e\u7c7b\u4f3c\u4e8e":84,"\u901a\u8fc7data":58,"\u901a\u8fc7volum":73,"\u903b\u8f91\u5212\u4e0a\u6587\u4ef6\u5206\u5757\u7684\u5355\u4f4d":28,"\u903b\u8f91\u56de\u5f52":84,"\u9047\u5230\u8be5\u9519\u8bef\u65f6":46,"\u9053\u6b49":56,"\u9069":56,"\u9075\u5b88\u4ee5\u4e0b\u7ea6\u5b9a":61,"\u9075\u5faa\u4ee5\u4e0b\u6d41\u7a0b":42,"\u9075\u5faa\u5982\u4e0b\u7684\u683c\u5f0f":85,"\u9075\u5faa\u6587\u7ae0":80,"\u90a3\u4e48":[34,58,62],"\u90a3\u4e480\u5c42\u5e8f\u5217\u5373\u4e3a\u4e00\u4e2a\u8bcd\u8bed":58,"\u90a3\u4e48\u53ef\u4ee5\u8ba4\u4e3a\u8bad\u7ec3\u4e0d\u6536\u655b":46,"\u90a3\u4e48\u5728":63,"\u90a3\u4e48\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u4e0d\u4f1a\u6267\u884c\u6d4b\u8bd5\u64cd\u4f5c":1,"\u90a3\u4e48\u5982\u4f55\u5224\u65ad\u8bad\u7ec3\u4e0d\u6536\u655b\u5462":46,"\u90a3\u4e48\u5e38\u6570\u8f93\u51fa\u6240\u80fd\u8fbe\u5230\u7684\u6700\u5c0fcost\u662f":46,"\u90a3\u4e48\u5f53check\u51fa\u6570\u636e\u4e0d\u5408\u6cd5\u65f6":2,"\u90a3\u4e48\u6211\u4eec\u4e5f\u5c31\u4e0d\u9700\u8981":60,"\u90a3\u4e48\u6211\u4eec\u53ef\u4ee5\u5224\u65ad\u4e3a\u8bad\u7ec3\u4e0d\u6536\u655b":46,"\u90a3\u4e48\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u6536\u96c6\u5e02\u573a\u4e0a\u623f\u5b50\u7684\u5927\u5c0f\u548c\u4ef7\u683c":47,"\u90a3\u4e48\u63a8\u8350\u4f7f\u7528":59,"\u90a3\u4e48\u63a8\u8350\u4f7f\u7528\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u65b9\u6cd5":59,"\u90a3\u4e48\u6536\u655b\u53ef\u80fd\u5f88\u6162":46,"\u90a3\u4e48\u6700\u597d\u5c06\u6570\u636e\u6587\u4ef6\u5728\u6bcf\u6b21\u8bfb\u53d6\u4e4b\u524d\u505a\u4e00\u6b21shuffl":46,"\u90a3\u4e48\u7528\u6237\u9700\u8981\u62c9\u53d6\u6240\u6709\u7684\u8fdc\u7a0b\u5206\u652f\u5230\u672c\u673a":46,"\u90a3\u4e48\u7f16\u8bd1\u8fc7\u7a0b\u53ea\u4f1a\u4ea7\u751f":60,"\u90a3\u4e48\u8bad\u7ec3\u6709\u53ef\u80fd\u4e0d\u6536\u655b":46,"\u90a3\u4e48\u8be5\u4f18\u5316\u7b97\u6cd5\u81f3\u5c11\u9700\u8981":46,"\u90a3\u4e48fc1\u548cfc2\u5c42\u5c06\u4f1a\u4f7f\u7528\u7b2c1\u4e2agpu\u6765\u8ba1\u7b97":72,"\u90a3\u4e48paddlepaddle\u4f1a\u6839\u636elayer\u7684\u58f0\u660e\u987a\u5e8f":2,"\u90a3\u4e5f\u5c31\u4e0d\u9700\u8981\u6025\u7740\u4f18\u5316\u6027\u80fd\u5566":67,"\u90a3\u4f30\u8ba1\u8fd9\u91cc\u7684\u6f5c\u529b\u5c31\u6ca1\u5565\u597d\u6316\u7684\u4e86":67,"\u90a3\u51cf\u5c11\u5b66\u4e60\u738710\u500d\u7ee7\u7eed\u8bd5\u9a8c":46,"\u90a3\u5bb9\u5668\u5c31\u662f\u8fd9\u4e2a\u7a0b\u5e8f\u8fd0\u884c\u65f6\u4ea7\u751f\u7684":49,"\u90a3\u6211\u4f1a\u671f\u671b\u5206\u6790\u5de5\u5177\u7edf\u8ba1\u5230\u901f\u5ea6\u662f100gb":67,"\u90a3\u7a0b\u5e8f\u5206\u6790\u5de5\u5177\u662f\u5fc5\u4e0d\u53ef\u5c11\u7684\u5229\u5668":67,"\u90ae\u7f16":85,"\u90e8\u7f72\u548c\u914d\u7f6e\u6bd4\u8f83\u7b80\u5355":73,"\u90fd":56,"\u90fd\u4e0d\u9700\u8981":60,"\u90fd\u4f1a\u4ea7\u751f\u5f53\u524d\u5c42\u72b6\u6001\u7684\u6240\u6709\u7ee7\u627f\u7ed3\u679c":70,"\u90fd\u4f7f\u7528\u968f\u673a\u503c\u521d\u59cb\u5316":47,"\u90fd\u53ea\u662f\u4ecb\u7ecd\u53cc\u5c42rnn\u7684api\u63a5\u53e3":56,"\u90fd\u53ef\u4ee5\u8fd0\u884c":60,"\u90fd\u662f\u4e94\u4f4d\u7684\u6570\u5b57":21,"\u90fd\u662f\u5bf9layer1\u5143\u7d20\u7684\u62f7\u8d1d":55,"\u90fd\u662f\u5c06\u6bcf\u4e00\u53e5\u5206\u597d\u8bcd\u540e\u7684\u53e5\u5b50":56,"\u90fd\u662fabi\u8c03\u7528\u6807\u51c6\u7684":33,"\u90fd\u7528":61,"\u90fd\u9700\u8981\u5199\u63d0\u4ea4\u8bf4\u660e":61,"\u90fd\u9700\u8981\u8c03\u7528\u4e00\u6b21":62,"\u914d\u7f6e":88,"\u914d\u7f6e\u4e86\u7f51\u7edc":86,"\u914d\u7f6e\u521b\u5efa\u5b8c\u6bd5\u540e":81,"\u914d\u7f6e\u5982\u4e0b":80,"\u914d\u7f6e\u6253\u5f00":67,"\u914d\u7f6e\u6587\u4ef6":84,"\u914d\u7f6e\u6587\u4ef6\u63a5\u53e3\u662ffc_layer":62,"\u914d\u7f6e\u6587\u4ef6\u91cc\u52a0\u4e24\u884c":60,"\u914d\u7f6e\u6a21\u578b\u6587\u4ef6":80,"\u914d\u7f6e\u7684\u65b9\u6cd5\u53c2\u8003":28,"\u914d\u7f6e\u7b49\u6587\u4ef6\u7684\u76ee\u5f55\u89c6\u4e3a":68,"\u914d\u7f6e\u7b80\u5355\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u4f8b\u5b50":59,"\u914d\u7f6e\u7f51\u7edc\u5c42\u7684\u8f93\u5165":62,"\u914d\u7f6eapi":55,"\u9152\u5e97":56,"\u91c7\u6837\u5c42":86,"\u91c7\u7528":87,"\u91c7\u7528\u53e6\u4e00\u79cd\u65b9\u6cd5\u6765\u5806\u53e0lstm\u5c42":87,"\u91c7\u7528\u5747\u5300\u5206\u5e03\u6216\u8005\u9ad8\u65af\u5206\u5e03\u521d\u59cb\u5316":70,"\u91c7\u7528multi":46,"\u91ca\u653e\u5bf9paramters\u5185\u5b58\u7684\u9501\u5b9a":20,"\u91cc":60,"\u91cc\u4ecb\u7ecd\u4e86\u7528paddle\u6e90\u7801\u4e2d\u7684\u811a\u672c\u4e0b\u8f7d\u8bad\u7ec3\u6570\u636e\u7684\u8fc7\u7a0b":74,"\u91cc\u53ef\u4ee5\u6807\u51c6\u5316\u7f16\u8bd1\u73af\u5883":60,"\u91cc\u5b8c\u6210":63,"\u91cc\u6240\u6709\u7684\u7b26\u53f7\u90fd\u5199\u5165\u81ea\u5df1\u7684\u7a0b\u5e8f\u7684\u4e8c\u8fdb\u5236\u6587\u4ef6\u91cc":33,"\u91cc\u6307\u5b9a\u56fe\u50cf\u6570\u636e\u5217\u8868":82,"\u91cc\u7684":60,"\u91cc\u7684\u65e5\u5fd7":68,"\u91cc\u7684\u6e90\u7801":60,"\u91cc\u8fd0\u884c\u7684\u7f16\u8bd1\u5de5\u5177\u5b9e\u9645\u4e0a\u90fd\u662f\u5728\u672c\u673a\u7684":60,"\u91cc\u9762":63,"\u91cc\u9762\u6db5\u76d6\u4e86":49,"\u91cc\u9762\u6db5\u76d6\u4e86paddlepaddle\u8fd0\u884c\u6240\u9700\u7684\u6240\u6709\u73af\u5883":49,"\u91cd\u547d\u540d\u6210":33,"\u91cd\u65b0\u7f16\u8bd1paddlepaddl":67,"\u9488\u5bf9\u4e0d\u540c\u5e73\u53f0\u7684\u7f16\u8bd1\u5de5\u5177":54,"\u9488\u5bf9\u4efb\u52a1\u8fd0\u884c\u5b8c\u6210\u540e\u5bb9\u5668\u81ea\u52a8\u9000\u51fa\u7684\u573a\u666f":74,"\u9488\u5bf9\u5185\u5b58\u548c\u663e\u5b58":46,"\u9488\u5bf9\u6587\u672c":86,"\u94fe\u63a5\u4f55\u79cdblas\u5e93\u7b49":48,"\u94fe\u63a5\u5230\u81ea\u5df1\u7684\u7a0b\u5e8f\u91cc":33,"\u94fe\u63a5\u5f85\u8865\u5145":84,"\u9500\u552e":85,"\u9519\u8bef\u5904\u7406":33,"\u9519\u8bef\u5904\u7406\u65b9\u5f0f\u662f\u8fd4\u56de\u503c":33,"\u9519\u8bef\u5904\u7406\u7684\u65b9\u5f0f\u4e5f\u4e0d\u5c3d\u76f8\u540c":33,"\u9519\u8bef\u7387":84,"\u9519\u8bef\u7684define_py_data_sources2\u7c7b\u4f3c":46,"\u952e\u6765\u542f\u52a8\u7f16\u8bd1\u4e86":60,"\u955c\u50cf":49,"\u955c\u50cf\u91cc\u6709":74,"\u957f\u5ea6":46,"\u95e8\u63a7\u5faa\u73af\u5355\u5143\u5355\u6b65\u51fd\u6570\u548c\u8f93\u51fa\u51fd\u6570":59,"\u95e8\u63a7\u5faa\u73af\u5355\u5143\u7684\u8f93\u51fa\u88ab\u7528\u4f5c\u8f93\u51famemori":59,"\u95ee\u9898":47,"\u95f4\u9694":84,"\u9650\u5236\u5957\u63a5\u5b57\u53d1\u9001\u7f13\u51b2\u533a\u7684\u5927\u5c0f":70,"\u9650\u5236\u5957\u63a5\u5b57\u63a5\u6536\u7f13\u51b2\u533a\u7684\u5927\u5c0f":70,"\u9664\u4e86":2,"\u9664\u4e86boot_lay":56,"\u9664\u53bbdata\u5c42":84,"\u9664\u6784\u9020\u67d0\u79cd\u7c7b\u578b\u7684\u51fd\u6570":34,"\u9664\u6b64\u4e4b\u5916":46,"\u9664\u8bcd\u5411\u91cf\u6a21\u578b\u5916\u7684\u53c2\u6570\u5c06\u4f7f\u7528\u6b63\u6001\u5206\u5e03\u968f\u673a\u521d\u59cb\u5316":80,"\u9664\u96f6\u7b49\u95ee\u9898":46,"\u968f\u540e\u53ef\u4ee5\u7528\u8fd9\u4e2a\u5f00\u53d1\u955c\u50cf\u5f00\u59cbbuild":61,"\u968f\u673a\u521d\u59cb\u4e0d\u5b58\u5728\u7684\u53c2\u6570":87,"\u968f\u673a\u6570\u7684\u79cd\u5b50":70,"\u968f\u673a\u6570seed":69,"\u968f\u7740\u8f6e\u6570\u589e\u52a0\u8bef\u5dee\u4ee3\u4ef7\u51fd\u6570\u7684\u8f93\u51fa\u5728\u4e0d\u65ad\u7684\u51cf\u5c0f":47,"\u9694\u5f00":82,"\u96c6":85,"\u96c6\u675f\u641c\u7d22\u4e2d\u7684\u6269\u5c55\u5e7f\u5ea6":89,"\u96c6\u675f\u641c\u7d22\u4f7f\u7528\u5e7f\u5ea6\u4f18\u5148\u641c\u7d22\u6765\u6784\u5efa\u641c\u7d22\u6811":89,"\u96c6\u675f\u641c\u7d22\u4f7f\u7528\u5e7f\u5ea6\u4f18\u5148\u641c\u7d22\u7684\u65b9\u5f0f\u6784\u5efa\u67e5\u627e\u6811":70,"\u96c6\u7fa4\u4e0a\u542f\u52a8\u4e00\u4e2a\u5355\u673a\u4f7f\u7528cpu\u7684paddle\u8bad\u7ec3\u4f5c\u4e1a":74,"\u96c6\u7fa4\u4f5c\u4e1a\u4e2d\u6240\u6709\u8fdb\u7a0b\u7684\u73af\u5883\u8bbe\u7f6e":68,"\u96c6\u7fa4\u4f5c\u4e1a\u5c06\u4f1a\u5728\u51e0\u79d2\u540e\u542f\u52a8":68,"\u96c6\u7fa4\u5de5\u4f5c":68,"\u96c6\u7fa4\u6d4b\u8bd5":69,"\u96c6\u7fa4\u8bad\u7ec3":69,"\u96c6\u7fa4\u8fdb\u7a0b":68,"\u96c6\u7fa4\u901a\u4fe1\u4fe1\u9053\u7684\u7aef\u53e3\u6570":68,"\u96c6\u7fa4\u901a\u4fe1\u901a\u9053\u7684":68,"\u96c6\u7fa4\u901a\u4fe1\u901a\u9053\u7684\u7aef\u53e3\u53f7":68,"\u9700\u52a0\u8be5\u6a21\u677f\u53c2\u6570":63,"\u9700\u5728nvvp\u754c\u9762\u4e2d\u9009\u4e0a\u624d\u80fd\u5f00\u542f":67,"\u9700\u8981":[21,60,63],"\u9700\u8981\u4e3a":63,"\u9700\u8981\u4f7f\u7528\u5176\u5236\u5b9a\u7684\u65b9\u5f0f\u6302\u8f7d\u540e\u5e76\u5bfc\u5165\u6570\u636e":75,"\u9700\u8981\u4fdd\u6301\u5f53\u524d\u5206\u652f\u76ee\u5f55":61,"\u9700\u8981\u4fee\u6539build":42,"\u9700\u8981\u5148\u6302\u8f7d\u5230\u670d\u52a1\u5668node\u4e0a\u518d\u901a\u8fc7kubernet":73,"\u9700\u8981\u5347\u7ea7pip\u7248\u672c\u5230\u6700\u65b0":46,"\u9700\u8981\u53ef\u4ee5\u8de8\u5e73\u53f0\u6267\u884c":28,"\u9700\u8981\u540c\u6b65\u539f\u4ed3\u5e93":61,"\u9700\u8981\u542f\u52a8\u7684\u8282\u70b9\u4e2a\u6570\u4ee5\u53ca":75,"\u9700\u8981\u548c\u8be5op\u7684\u540d\u5b57\u4e00\u6837":63,"\u9700\u8981\u5728":[63,68],"\u9700\u8981\u5728\u521b\u5efa\u5bb9\u5668\u524d\u6302\u8f7d\u5377\u4ee5\u4fbf\u6211\u4eec\u4fdd\u5b58\u8bad\u7ec3\u7ed3\u679c":74,"\u9700\u8981\u5728\u7cfb\u7edf\u91cc\u5148\u5b89\u88c5\u597ddocker\u5de5\u5177\u5305":65,"\u9700\u8981\u5728cmake\u7684\u65f6\u5019":34,"\u9700\u8981\u5b89\u88c5graphviz\u6765\u8f6c\u6362dot\u6587\u4ef6\u4e3a\u56fe\u7247":82,"\u9700\u8981\u5bf9":73,"\u9700\u8981\u5c06\u5176parameter\u8bbe\u7f6e\u6210":46,"\u9700\u8981\u5c06\u6807\u8bb0\u7684\u8bad\u7ec3\u548c\u6d4b\u8bd5\u6837\u672c\u79fb\u52a8\u5230\u53e6\u4e00\u4e2a\u8def\u5f84":88,"\u9700\u8981\u5c06bugfix\u7684\u5206\u652f\u540c\u65f6merge\u5230":42,"\u9700\u8981\u5f15\u7528":34,"\u9700\u8981\u5f3a\u8c03\u7684\u662f":60,"\u9700\u8981\u6307\u5b9a\u4e0e\u67d0\u4e00\u4e2a\u8f93\u5165\u7684\u5e8f\u5217\u4fe1\u606f\u662f\u4e00\u81f4\u7684":56,"\u9700\u8981\u6307\u5b9alayer\u7684\u8f93\u5165\u6765\u6e90":51,"\u9700\u8981\u660e\u786e\u6307\u5b9a":70,"\u9700\u8981\u6709\u4e00\u4e2a\u5916\u90e8\u7684\u5b58\u50a8\u670d\u52a1\u6765\u4fdd\u5b58\u8bad\u7ec3\u6240\u9700\u6570\u636e\u548c\u8bad\u7ec3\u8f93\u51fa":73,"\u9700\u8981\u6709\u7a33\u5b9a\u7684\u5bfc\u51fa\u7b26\u53f7":33,"\u9700\u8981\u6839\u636e\u4e0d\u540c\u7684\u5206\u5e03\u5f0f\u5b58\u50a8\u6765\u7ed1\u5b9a\u4e00\u4e2a":75,"\u9700\u8981\u6ce8\u610f":63,"\u9700\u8981\u6ce8\u610f\u7684\u662f":[42,46,70,86],"\u9700\u8981\u6ce8\u610f\u7684\u662f\u68af\u5ea6\u68c0\u67e5\u4ec5\u4ec5\u9a8c\u8bc1\u4e86\u68af\u5ea6\u7684\u8ba1\u7b97":62,"\u9700\u8981\u6ce8\u610f\u7684\u662fpaddlepaddle\u76ee\u524d\u53ea\u652f\u6301\u5b50\u5e8f\u5217\u6570\u76ee\u4e00\u6837\u7684\u591a\u8f93\u5165\u53cc\u5c42rnn":56,"\u9700\u8981\u7528\u5230\u7684\u7f16\u8bd1\u5de5\u5177\u548c\u7cfb\u7edf\u5e93":53,"\u9700\u8981\u7528\u6237\u663e\u5f0f\u8bbe\u5b9a":46,"\u9700\u8981\u88ab\u66b4\u9732\u5230\u5176\u4ed6\u8bed\u8a00":34,"\u9700\u8981\u8bf7\u7ba1\u7406\u5458\u5b89\u88c5\u548c\u914d\u7f6e\u597d":60,"\u9700\u8981\u9075\u5faa\u4ee5\u4e0b\u7ea6\u5b9a":58,"\u9700\u8981\u91cd\u547d\u540dwheel\u5305\u4e2dplatform\u76f8\u5173\u7684\u540e\u7f00":42,"\u9700\u8981\u989d\u5916\u6ce8\u610f\u7684\u662f":64,"\u975e\u5e38\u6570":62,"\u975e\u96f6\u6570\u5b57\u7684\u4e2a\u6570":62,"\u9762\u5411\u67b6\u6784\u4e3a32\u4f4darm\u67b6\u6784":53,"\u97f3\u4e50\u5267":85,"\u9879\u76ee\u5728\u52aa\u529b\u5f00\u59cb\u652f\u6301\u5176\u4ed6\u4e0d\u9700\u8981":60,"\u987a\u5e8f":56,"\u9884\u5904\u7406\u6570\u636e\u4e00\u822c\u7684\u547d\u4ee4\u4e3a":86,"\u9884\u5904\u7406\u811a\u672c":88,"\u9884\u5b9a\u4e49\u7f51\u7edc":88,"\u9884\u5b9a\u4e49\u7f51\u7edc\u5982\u56fe3\u6240\u793a":88,"\u9884\u63d0\u4ea4\u94a9\u5b50":61,"\u9884\u6d4b\u540e":87,"\u9884\u6d4b\u63a5\u53e3\u811a\u672c":88,"\u9884\u6d4b\u6982\u7387\u53d6\u5e73\u5747":82,"\u9884\u6d4b\u7a0b\u5e8f\u5c06\u8bfb\u53d6\u7528\u6237\u7684\u8f93\u5165":86,"\u9884\u6d4b\u7ed3\u679c\u4ee5\u6587\u672c\u7684\u5f62\u5f0f\u4fdd\u5b58\u5728":84,"\u9884\u6d4b\u811a\u672c\u662f":87,"\u9884\u6d4bid":84,"\u9884\u6d4bimdb\u7684\u672a\u6807\u8bb0\u8bc4\u8bba\u7684\u4e00\u4e2a\u5b9e\u4f8b\u5982\u4e0b":88,"\u9884\u8bad\u7ec3\u6a21\u578b\u4f7f\u7528\u7684\u5b57\u5178\u7684\u8def\u5f84":80,"\u9884\u8bad\u7ec3\u8bcd\u5411\u91cf\u5b57\u5178\u6a21\u578b\u7684\u8def\u5f84":80,"\u989c\u8272\u901a\u9053\u987a\u5e8f\u4e3a":82,"\u989d\u5916\u7684\u53c2\u6570":84,"\u9996\u5148":[2,47,56,59,62,80,82,84,87,88],"\u9996\u5148\u4e0b\u8f7dcifar":81,"\u9996\u5148\u5728\u7cfb\u7edf\u8def\u5f84":48,"\u9996\u5148\u5b89\u88c5\u5e76\u5728\u5f53\u524d\u76ee\u5f55\u8fd0\u884c\u5b83":61,"\u9996\u5148\u5b89\u88c5paddlepaddl":88,"\u9996\u5148\u5b9a\u4e49":63,"\u9996\u5148\u5bf9\u8f93\u5165\u505a\u4e00\u4e2a\u5c0f\u7684\u6270\u52a8":62,"\u9996\u5148\u6211\u4eec\u9700\u8981\u63a8\u5bfc\u8be5\u7f51\u7edc\u5c42\u7684":62,"\u9996\u5148\u662f\u6cd5\u8bed\u5e8f\u5217":89,"\u9996\u5148\u6784\u9020\u5934\u4fe1\u606f":46,"\u9996\u5148\u901a\u8fc7":61,"\u9996\u5148\u9700\u8981\u52a0\u8f7d\u76f8\u5e94\u7684python\u5e93":51,"\u9a71\u52a8":65,"\u9ad8\u4e2d\u6bd5\u4e1a\u751f":85,"\u9ad8\u4eae\u90e8\u5206":56,"\u9ad8\u53ef\u7528":73,"\u9ad8\u5ea6\u652f\u6301\u7075\u6d3b\u548c\u9ad8\u6548\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e":59,"\u9ad8\u65af\u5206\u5e03":46,"\u9ed1\u8272\u7535\u5f71":85,"\u9ed8\u8ba4":[2,70,89],"\u9ed8\u8ba4256k":28,"\u9ed8\u8ba4\u4e00\u4e2apass\u4fdd\u5b58\u4e00\u6b21\u6a21\u578b":84,"\u9ed8\u8ba4\u4e0d\u663e\u793a":70,"\u9ed8\u8ba4\u4e0d\u8bbe\u7f6e":58,"\u9ed8\u8ba4\u4e3a0":[70,72],"\u9ed8\u8ba4\u4e3a1":[2,72],"\u9ed8\u8ba4\u4e3a100":72,"\u9ed8\u8ba4\u4e3a4096mb":70,"\u9ed8\u8ba4\u4e3a\u4e0d\u4f7f\u7528":86,"\u9ed8\u8ba4\u4e3a\u7b2c\u4e00\u4e2a\u8f93\u5165":58,"\u9ed8\u8ba4\u4e3anull":70,"\u9ed8\u8ba4\u4f1a\u5c06a\u548cb":46,"\u9ed8\u8ba4\u4f7f\u7528concurrentremoteparameterupdat":70,"\u9ed8\u8ba4\u503c":[48,55,72],"\u9ed8\u8ba4\u503c\u4e3a":[53,54],"\u9ed8\u8ba4\u521d\u59cb\u72b6\u4e3a0":58,"\u9ed8\u8ba4\u60c5\u51b5\u4e0b":[46,49,68,88],"\u9ed8\u8ba4\u60c5\u51b5\u4e0b\u4f7f\u7528\u6b64\u7f51\u7edc":88,"\u9ed8\u8ba4\u60c5\u51b5\u4e0b\u6309\u7167float\u7cbe\u5ea6\u8ba1\u7b97":46,"\u9ed8\u8ba4\u6307\u5b9a\u7b2c\u4e00\u4e2a\u8f93\u5165":56,"\u9ed8\u8ba4\u7528\u6765\u5207\u5206\u5355\u8bb0\u548c\u6807\u70b9\u7b26\u53f7":88,"\u9ed8\u8ba4\u7684":74,"\u9ed8\u8ba4\u8bbe\u7f6e\u4e3a\u771f":72,"\u9ed8\u8ba4\u8bbe\u7f6e\u6210\u73af\u5883\u53d8\u91cf":[53,54],"\u9ed8\u8ba4\u8c03\u7528":60,"\u9ed8\u8ba4\u914d\u7f6e\u5982\u4e0b":68,"abstract":43,"adamax\u7b49":84,"amazon\u7535\u5b50\u4ea7\u54c1\u8bc4\u8bba\u6570\u636e":84,"android\u5e73\u53f0\u4e0a\u4f7f\u7528\u7684c":53,"android\u5e73\u53f0\u53ef\u9009\u914d\u7f6e\u53c2\u6570":53,"api\u4e2d\u4f7f\u7528":33,"api\u5bf9\u6bd4\u4ecb\u7ecd":57,"api\u5bfc\u51fa\u7684\u52a8\u6001\u5e93":34,"api\u5bfc\u51fa\u7684\u9759\u6001\u5e93":34,"api\u63a5\u53d7\u7684\u7c7b\u578b\u5168\u662f":34,"api\u63a5\u53e3":[28,73],"api\u63a5\u53e3\u751f\u6210":63,"api\u63a5\u53e3\u7684\u53c2\u6570\u8f6c\u53d1\u7ed9":34,"api\u63a5\u53e3\u7684\u751f\u6210":63,"api\u6587\u6863":53,"api\u65f6":34,"api\u65f6\u6240\u552f\u4e00\u9700\u8981\u5f15\u5165\u7684\u5934\u6587\u4ef6":34,"api\u662f\u591a\u8bed\u8a00api\u7684\u57fa\u7840\u90e8\u5206":34,"api\u66b4\u9732\u7684\u7c7b\u578b":34,"api\u6765\u9884\u6d4b":53,"api\u751f\u6210\u7684\u4e8c\u8fdb\u5236\u6587\u4ef6\u4f1a\u88ab\u5b89\u88c5\u5230":34,"api\u7684\u5934\u6587\u4ef6":[53,54],"api\u7684\u5b9e\u4f8b":34,"api\u7684\u5b9e\u73b0\u7ec6\u8282":34,"api\u7684\u63a5\u53e3":34,"api\u7684\u65f6\u5019\u63a8\u8350paddle\u4e0d\u5d4c\u5165python\u89e3\u91ca\u5668":34,"api\u7684\u7f16\u8bd1\u9009\u9879\u9ed8\u8ba4\u5173\u95ed":34,"api\u76ee\u5f55\u7ed3\u6784\u5982\u4e0a\u56fe\u8868\u6240\u793a":34,"api\u7ea7\u522b":53,"api\u7ea7\u522b\u4e0d\u5c0f\u4e8e21":53,"api\u7ea7\u522b\u4e3a21":53,"api\u83b7\u5f97\u4e86\u795e\u7ecf\u7f51\u7edc\u7684\u53c2\u6570\u5b9e\u4f8b":34,"app\u4e2d":53,"async_sgd\u8fdb\u884c\u8bad\u7ec3\u65f6":46,"atlas\u7684\u8def\u5f84":48,"avx\u7684\u955c\u50cf":49,"awselasticblockstore\u7b49":73,"batch\u4e2d\u5305\u542b":46,"batches\u4e2a\u6279\u6b21\u4fdd\u5b58\u4e00\u6b21\u53c2\u6570":70,"batches\u6b21":70,"bin\u548c\u8bc4\u5206\u6587\u4ef6":86,"blas\u7684\u8def\u5f84":48,"block\u6784\u6210\u4e00\u4e2amodel":20,"book\u4e00\u5b9a\u662f\u60a8\u6700\u597d\u7684\u9009\u62e9":49,"book\u4e2d\u6240\u6709\u7ae0\u8282\u529f\u80fd\u7684\u6b63\u786e\u6027":42,"book\u662f\u4e3a\u7528\u6237\u548c\u5f00\u53d1\u8005\u5236\u4f5c\u7684\u4e00\u4e2a\u4ea4\u4e92\u5f0f\u7684jupyt":49,"book\u7684":51,"book\u7684docker\u955c\u50cf":49,"bool\u578b\u53c2\u6570":2,"boolean":[8,29,32,33],"break":18,"bugfix\u5206\u652f\u4e5f\u662f\u5728\u5f00\u53d1\u8005\u81ea\u5df1\u7684fork\u7248\u672c\u5e93\u7ef4\u62a4":42,"bugfix\u5206\u652f\u9700\u8981\u5206\u522b\u7ed9\u4e3b\u7248\u672c\u5e93\u7684":42,"build\u6307\u4ee4\u6309\u7167dockerfile\u751f\u6210\u955c\u50cf":49,"build\u8fd9\u4e2a\u547d\u4ee4\u7684":49,"byte":[28,46],"c99\u662f\u76ee\u524dc\u6700\u5e7f\u6cdb\u7684\u4f7f\u7528\u6807\u51c6":33,"c\u6709\u6807\u51c6\u7684abi":33,"c\u8bed\u8a00\u662f\u6709\u5bfc\u51fa\u7b26\u53f7\u7684\u6807\u51c6\u7684":33,"caoying\u7684pul":89,"case":[8,22,32,34,38,39,41,67,79],"cc\u4e2d\u7684":64,"cells\u7b49":46,"char":24,"class":[5,6,7,8,9,10,11,12,14,15,17,30,31,33,39,41,43,44,46,62,63,64,88],"cmake\u4e2d\u5c06":67,"cmake\u53c2\u6570\u5982\u4e0b":54,"cmake\u5b98\u65b9\u5bf9android\u5e73\u53f0\u7684\u4ea4\u53c9\u7f16\u8bd1\u63d0\u4f9b\u4e86\u901a\u7528\u7684\u652f\u6301":53,"cmake\u627e\u5230\u7684python\u5e93\u548cpython\u89e3\u91ca\u5668\u7248\u672c\u53ef\u80fd\u6709\u4e0d\u4e00\u81f4\u73b0\u8c61":46,"cmake\u7cfb\u7edf\u5bf9\u4ea4\u53c9\u7f16\u8bd1\u63d0\u4f9b\u4e86\u652f\u6301":[53,54],"cmake\u7f16\u8bd1\u65f6":48,"cmake\u7f16\u8bd1\u7684\u76ee\u6807\u5e73\u53f0":[53,54],"cmake\u914d\u7f6e\u4e2d\u5c06":67,"cmake\u914d\u7f6e\u5b8c\u6210\u540e":[53,54],"com\u4e0b\u8f7d\u7684":49,"conf\u4f5c\u4e3a\u914d\u7f6e":89,"const":[17,22,24,36,43,44,62,63,64],"container\u4e2d":74,"core\u4e2d\u7684\u6a21\u578b\u8fd8\u5728\u4f7f\u7528\u8fd9\u4e2a\u53c2\u6570":34,"core\u4e2d\u8fd9\u4e00\u7c7b\u578b\u63a5\u53e3\u7684\u667a\u80fd\u6307\u9488":34,"core\u662f\u5426\u8fd8\u5728\u4f7f\u7528\u8fd9\u4e2a\u5b9e\u4f8b":34,"core\u6982\u5ff5":34,"cost\u63a5\u6536y_predict\u4e0ey\u4f5c\u4e3a\u8f93\u5165":51,"cost\u8fd8\u5927\u4e8e\u8fd9\u4e2a\u6570":46,"count\u4e2agpu\u4e0a\u4f7f\u7528\u6570\u636e\u5e76\u884c\u6765\u8ba1\u7b97\u67d0\u4e00\u5c42":72,"count\u548cgpu":72,"cuda\u5e93":70,"cudnn\u5e93":[48,70],"dat\u4e2d":86,"data\u5230\u5206\u5e03\u5f0f\u5b58\u50a8\u8865\u5145\u8bad\u7ec3\u6570\u636e":21,"data\u63a5\u53e3\u5206\u914d\u5b9e\u9645\u7684\u5185\u5b58":64,"data\u76ee\u5f55\u4e2d\u5b58\u653e\u5207\u5206\u597d\u7684\u6570\u636e":75,"dataprovider\u5171\u8fd4\u56de\u4e24\u4e2a\u6570\u636e":56,"dataprovider\u5171\u8fd4\u56de\u4e24\u7ec4\u6570\u636e":56,"dataprovider\u662fpaddlepaddle\u8d1f\u8d23\u63d0\u4f9b\u6570\u636e\u7684\u6a21\u5757":1,"dataprovider\u7684\u4ecb\u7ecd":[3,84],"dataprovider\u7f13\u51b2\u6c60\u5185\u5b58":46,"deb\u5305":42,"deb\u5305\u7f16\u8bd1\u95ee\u9898":42,"debian\u955c\u50cf":49,"decay\u5219\u4e3a0":81,"decoder\u5faa\u73af\u5c55\u5f00\u7684\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u4f1a\u5f15\u7528\u5168\u90e8\u7ed3\u679c":58,"decoder\u63a5\u53d7\u4e24\u4e2a\u8f93\u5165":58,"decoder\u6bcf\u6b21\u9884\u6d4b\u4ea7\u751f\u4e0b\u4e00\u4e2a\u6700\u53ef\u80fd\u7684\u8bcd\u8bed":58,"decoer\u67b6\u6784":58,"default":[6,7,8,9,11,12,14,15,17,18,32,35,43,44,45,72,74,75,79,88],"demo\u9884\u6d4b\u8f93\u51fa\u5982\u4e0b":4,"dictionary\u662f\u4ece\u7f51\u7edc\u914d\u7f6e\u4e2d\u4f20\u5165\u7684dict\u5bf9\u8c61":2,"dictionary\u7531\u89e3\u6790\u81ea\u52a8\u751f\u6210":86,"dir\u4e2d\u670916\u4e2a\u5b50\u76ee\u5f55":89,"docker\u5b58\u5728\u95ee\u9898":49,"docker\u5b89\u88c5\u8bf7\u53c2\u8003":65,"docker\u5bb9\u5668\u4e2d":49,"docker\u5bb9\u5668\u4f1a\u8fd0\u884c\u5728\u72ec\u7acb\u7684\u6587\u4ef6\u7cfb\u7edf\u7a7a\u95f4\u4e4b\u4e0a":49,"docker\u7684\u5b98\u7f51":65,"docker\u955c\u50cf\u6765\u670d\u52a1html\u4ee3\u7801":49,"dockerfile\u4f4d\u4e8epaddlepaddl":49,"dockers\u8bbe\u7f6e":49,"double\u7c7b\u578b\u65f6\u4e3a8":46,"dropout\u7684\u6bd4\u4f8b":62,"eigenscalar\u7684\u8f6c\u6362":64,"elec\u6d4b\u8bd5\u96c6":84,"embedding\u6a21\u578b\u9700\u8981\u7a0d\u5fae\u6539\u53d8\u63d0\u4f9b\u6570\u636e\u7684python\u811a\u672c":84,"encode\u6210\u7684\u6700\u540e\u4e00\u4e2a\u5411\u91cf":56,"encoder\u548cdecoder\u53ef\u4ee5\u662f\u80fd\u591f\u5904\u7406\u5e8f\u5217\u7684\u4efb\u610f\u795e\u7ecf\u7f51\u7edc\u5355\u5143":58,"encoder\u8f93\u51fa":58,"entropy\u4f5c\u4e3acost":46,"enum":[22,24,44,45],"evaluator\u7684\u503c\u4f4e\u4e8e0":89,"export":[46,49,81],"f\u4ee3\u8868\u4e00\u4e2a\u6d6e\u70b9\u6570":[2,51],"false\u7684\u60c5\u51b5":2,"fc1\u548cfc2\u5c42\u5728gpu\u4e0a\u8ba1\u7b97":72,"fc3\u5c42\u4f7f\u7528cpu\u8ba1\u7b97":72,"final":[8,9,16,35],"flatten\u65b9\u6cd5\u662f\u628apaddle\u4e2d\u7684\u4e00\u4e2atensor\u8fdb\u884creshape\u64cd\u4f5c":64,"float":[2,6,7,8,10,12,44,47,63,64,67,82],"float\u7b49":72,"from\u65b9\u6cd5\u662f\u628apaddle\u4e2d\u7684\u4e00\u7ef4tensor\u8f6c\u4e3aeigen\u7684\u4e00\u7ef4tensor":64,"from\u662feigentensor\u6a21\u677f\u63d0\u4f9b\u7684\u4e00\u4e2a\u63a5\u53e3":64,"full\u53c2\u6570\u63d0\u4ea4":46,"function":[8,9,10,12,15,17,19,23,24,25,27,31,35,36,37,38,39,40,41,43,44,59,79,88],"function\u4f7f\u7528":46,"gen\u6587\u4ef6\u5939\u4e2d\u7684\u6587\u4ef6\u5217\u8868":89,"generator\u4fbf\u4f1a\u5b58\u4e0b\u5f53\u524d\u7684\u4e0a\u4e0b\u6587":2,"generator\u81f3\u5c11\u9700\u8981\u8c03\u7528\u4e24\u6b21\u624d\u4f1a\u77e5\u9053\u662f\u5426\u505c\u6b62":2,"git\u6d41\u5206\u652f\u6a21\u578b":61,"github\u9996\u9875":61,"golang\u53ef\u4ee5\u4f7f\u7528":33,"golang\u7684":33,"gpu\u4e8c\u8fdb\u5236\u6587\u4ef6":48,"gpu\u5171\u4eabkernel\u5b9e\u73b0\u5728":63,"gpu\u5219\u8fd8\u9700\u8981\u9ad8\u5e76\u884c\u6027":67,"gpu\u53cc\u7f13\u5b58":2,"gpu\u548c\u975eavx\u533a\u5206\u4e86\u5982\u4e0b4\u4e2a\u955c\u50cf":49,"gpu\u5b9e\u73b0\u5171\u4eab\u540c\u4e00\u4e2a":63,"gpu\u5b9e\u73b0\u5728":63,"gpu\u6027\u80fd\u5206\u6790\u4e0e\u8c03\u4f18":66,"gpu\u6267\u884c":64,"gpu\u6838\u5728\u8bad\u7ec3\u914d\u7f6e\u4e2d\u6307\u5b9a":70,"gpu\u7248\u672c\u5e76\u60f3\u4f7f\u7528":87,"gpu\u7684\u4ee3\u7801\u53ef\u4ee5\u590d\u7528":63,"gpu\u7684docker\u955c\u50cf\u7684\u65f6\u5019":46,"gpu\u8ba1\u7b97\u7ed3\u679c":63,"gram\u7ea7\u522b\u7684\u77e5\u8bc6":88,"group\u6559\u7a0b":57,"group\u7684\u5b9e\u73b0\u65b9\u5f0f":46,"gru\u6216lstm":59,"gru\u6a21\u578b":84,"gru\u6a21\u578b\u914d\u7f6e":84,"h\u5e76\u4e0d\u56f0\u96be":33,"hot\u7a20\u5bc6\u5411\u91cf":86,"html\u5373\u53ef\u8bbf\u95ee\u672c\u5730\u6587\u6863":65,"i\u4ee3\u8868\u4e00\u4e2a\u6574\u6570":[2,51],"id\u4e3a0\u7684\u6982\u7387":84,"id\u4e3a1\u7684\u6982\u7387":84,"id\u6307\u5b9a\u4f7f\u7528\u54ea\u4e2agpu\u6838":70,"id\u6307\u5b9a\u7684gpu":72,"id\u65e0\u6548":70,"image\u91cc":74,"images\u6570\u636e\u96c6\u4e0a\u4f20\u5230\u4e91\u7aef\u7684":21,"imdb\u6570\u636e\u96c6\u5305\u542b25":88,"imdb\u6709\u4e24\u4e2a\u6807\u7b7e":88,"imdb\u7684\u6570\u6910\u96c6":88,"import":[2,4,7,8,15,17,18,32,35,40,43,45,47,51,63,80,81,82,88,89],"include\u4e0b\u9700\u8981\u5305\u542bcbla":48,"include\u4e0b\u9700\u8981\u5305\u542bmkl":48,"infer\u63a5\u53e3\u7684\u8fd4\u56de\u503c\u662f\u4e00\u4e2apython":46,"ingress\u9700\u8981\u628apfsclient\u7684\u8eab\u4efd\u4fe1\u606f\u4f20\u7ed9pfsserv":28,"init_hook\u53ef\u4ee5\u4f20\u5165\u4e00\u4e2a\u51fd\u6570":2,"instance\u4e0e\u751f\u6210\u6570\u636e\u96c6\u65f6":21,"instance\u5305\u6db5\u4e24\u4e2a\u503c":21,"instance\u662f\u4e00\u6a21\u4e00\u6837\u7684":21,"int":[2,6,7,8,9,12,17,22,23,24,27,33,34,37,38,44,56,62,64,72,87],"interface\u6587\u4ef6\u7684\u5199\u6cd5\u975e\u5e38":33,"issue\u7f16\u53f7":61,"job\u542f\u52a8\u540e\u4f1a\u521b\u5efa\u8fd9\u4e9bpod\u5e76\u5f00\u59cb\u6267\u884c\u4e00\u4e2a\u7a0b\u5e8f":73,"job\u6216\u8005\u5e94\u7528\u7a0b\u5e8f\u5728\u5bb9\u5668\u4e2d\u8fd0\u884c\u65f6\u751f\u6210\u7684\u6570\u636e\u4f1a\u5728\u5bb9\u5668\u9500\u6bc1\u65f6\u6d88\u5931":73,"job\u662f\u672c\u6b21\u8bad\u7ec3\u5bf9\u5e94\u7684job":75,"job\u7684\u540d\u5b57":75,"kernel\u5b9e\u73b0":63,"kernel\u6ce8\u518ccpu\u5b9e\u73b0\u5728":63,"kernel\u7684\u5b9e\u73b0\u57fa\u4e8eeigen":63,"kubernetes\u4e3a\u8fd9\u6b21\u8bad\u7ec3\u521b\u5efa\u4e863\u4e2apod\u5e76\u4e14\u8c03\u5ea6\u5230\u4e863\u4e2anode\u4e0a\u8fd0\u884c":75,"kubernetes\u5206\u5e03\u5f0f\u8bad\u7ec3":66,"kubernetes\u5355\u673a\u8bad\u7ec3":66,"kubernetes\u53ef\u4ee5\u5728\u7269\u7406\u673a\u6216\u865a\u62df\u673a\u4e0a\u8fd0\u884c":73,"kubernetes\u53ef\u4ee5\u901a\u8fc7yaml\u6587\u4ef6\u6765\u521b\u5efa\u76f8\u5173\u5bf9\u8c61":75,"kubernetes\u5c31\u4f1a\u521b\u5efa3\u4e2apod\u4f5c\u4e3apaddlepaddle\u8282\u70b9\u7136\u540e\u62c9\u53d6\u955c\u50cf":75,"kubernetes\u63d0\u4f9b\u4e86\u591a\u79cd\u96c6\u7fa4\u90e8\u7f72\u7684\u65b9\u6848":73,"kubernetes\u652f\u6301\u591a\u79cdvolum":73,"kubernetes\u6709job\u7c7b\u578b\u7684\u8d44\u6e90\u6765\u652f\u6301":74,"kubernetes\u96c6\u7fa4\u5c31\u662f\u7531node\u8282\u70b9\u4e0emaster\u8282\u70b9\u7ec4\u6210\u7684":73,"label\u662f\u539f\u59cb\u6570\u636e\u4e2d\u5bf9\u4e8e\u6bcf\u4e00\u53e5\u8bdd\u7684\u5206\u7c7b\u6807\u7b7e":56,"labels\u662f\u6bcf\u7ec4\u5185\u6bcf\u4e2a\u53e5\u5b50\u7684\u6807\u7b7e":56,"layer1\u5fc5\u987b\u662f\u4e00\u4e2a0\u5c42\u5e8f\u5217":55,"layer1\u5fc5\u987b\u662f\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":55,"layer\u4f5c\u4e3a\u4e00\u4e2a\u6574\u4f53\u6765\u5b9e\u73b0":46,"layer\u62ff\u5230\u7684\u7528\u6237\u8f93\u5165":58,"layer\u65f6":46,"layer\u662f\u6211\u4eec\u7684\u79ef\u6728":51,"layer\u7c7b\u53ef\u4ee5\u81ea\u52a8\u8ba1\u7b97\u4e0a\u9762\u7684\u5bfc\u6570":62,"layer\u8ba1\u7b97\u7684\u8f93\u51fa":46,"lib\u4e0b\u9700\u8981\u5305\u542bcblas\u548catlas\u4e24\u4e2a\u5e93":48,"lib\u4e0b\u9700\u8981\u5305\u542bcblas\u5e93":48,"lib\u4e0b\u9700\u8981\u5305\u542bopenblas\u5e93":48,"lib\u76ee\u5f55\u4e0b\u9700\u8981\u5305\u542bmkl_cor":48,"list\u4e2d\u7684\u6bcf\u4e00\u884c\u90fd\u4f20\u9012\u7ed9process\u51fd\u6570":2,"list\u4f5c\u4e3a\u68c0\u67e5\u5217\u8868":42,"list\u5199\u5165\u90a3\u4e2a\u6587\u672c\u6587\u4ef6\u7684\u5730\u5740":2,"list\u548ctest":1,"list\u5982\u4e0b\u6240\u793a":72,"list\u5b58\u653e\u5728\u672c\u5730":1,"list\u6307\u5b9a\u6d4b\u8bd5\u7684\u6a21\u578b\u5217\u8868":72,"long":[8,9,12],"lstm\u67b6\u6784\u7684\u6700\u5927\u4f18\u70b9\u662f\u5b83\u53ef\u4ee5\u5728\u957f\u65f6\u95f4\u95f4\u9694\u5185\u8bb0\u5fc6\u4fe1\u606f":88,"lstm\u6a21\u578b":84,"lstm\u6a21\u578b\u914d\u7f6e":84,"lstm\u7f51\u7edc\u7c7b\u4f3c\u4e8e\u5177\u6709\u9690\u85cf\u5c42\u7684\u6807\u51c6\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":88,"memory\u4e0d\u80fd\u72ec\u7acb\u5b58\u5728":58,"memory\u4e5f\u53ef\u4ee5\u5177\u6709":59,"memory\u4e5f\u53ef\u4ee5\u662f\u5e8f\u5217":59,"memory\u53ea\u80fd\u5728":58,"memory\u53ef\u4ee5\u7f13\u5b58\u4e0a\u4e00\u4e2a\u65f6\u523b\u67d0\u4e00\u4e2a\u795e\u7ecf\u5143\u7684\u8f93\u51fa":56,"memory\u6307\u5411\u4e00\u4e2alay":58,"memory\u662f\u5728\u5355\u6b65\u51fd\u6570\u4e2d\u5faa\u73af\u4f7f\u7528\u7684\u72b6\u6001":59,"memory\u662fpaddlepaddle\u5b9e\u73b0rnn\u65f6\u5019\u4f7f\u7528\u7684\u4e00\u4e2a\u6982\u5ff5":56,"memory\u7684":59,"memory\u7684\u521d\u59cb\u72b6\u6001":58,"memory\u7684\u65f6\u95f4\u5e8f\u5217\u957f\u5ea6\u4e00\u81f4\u7684\u60c5\u51b5":56,"memory\u7684\u66f4\u591a\u8ba8\u8bba\u8bf7\u53c2\u8003\u8bba\u6587":58,"memory\u7684\u8f93\u51fa\u5b9a\u4e49\u5728":59,"memory\u7684i":58,"memory\u9ed8\u8ba4\u521d\u59cb\u5316\u4e3a0":58,"mkl\u7684\u8def\u5f84":48,"mkl_sequential\u548cmkl_intel_lp64\u4e09\u4e2a\u5e93":48,"mnist\u662f\u4e00\u4e2a\u5305\u542b\u670970":2,"mode\u548cattent":89,"mode\u7684python\u51fd\u6570":89,"model\u505a\u5206\u652f\u7ba1\u7406":42,"model\u53ef\u4ee5\u901a\u8fc7":4,"model\u6765\u5b9e\u73b0\u624b\u5199\u8bc6\u522b\u7684\u9884\u6d4b\u4ee3\u7801":4,"movielens\u6570\u636e\u96c6":86,"name\u662f\u4f53\u88c1":86,"name\u662f\u5e74\u9f84":86,"name\u662f\u6027\u522b":86,"name\u662f\u7535\u5f71\u540d":86,"name\u662f\u804c\u4e1a":86,"name\u7ec4\u5408\u53ef\u4ee5\u627e\u5230\u672c\u6b21\u8bad\u7ec3\u9700\u8981\u7684\u6587\u4ef6\u8def\u5f84":75,"ndarray\u7c7b\u578b\u7684\u503c\u548c\u6574\u578b\u7684\u503c":21,"ndk\u4e2d\u5305\u542b\u4e86\u6240\u6709android":53,"new":[8,12,16,17,18,19,22,23,24,25,26,38,39,41,44,61,62,79],"nfs\u7684\u90e8\u7f72\u65b9\u6cd5\u53ef\u4ee5\u53c2\u8003":73,"nmt\u6a21\u578b\u53d7\u5236\u4e8e\u6e90\u8bed\u53e5\u7684\u7f16\u7801":89,"normalization\u5c42":82,"normalization\u5c42\u7684\u53c2\u6570":82,"note\u7684\u4e66\u5199":42,"notebook\u662f\u4e00\u4e2a\u5f00\u6e90\u7684web\u7a0b\u5e8f":49,"null":[8,62,70],"num_gradient_servers\u53c2\u6570":75,"num_samples_processed\u4e3a\u5df2\u8bad\u7ec3\u6837\u672c\u6570":46,"op\u4e0d\u9700\u8981\u5b9a\u4e49opprotomak":63,"op\u5b9a\u4e49":63,"op\u6709\u8ba1\u7b97\u51fd\u6570":63,"op\u6ce8\u518c\u5b9e\u73b0\u5728":63,"op\u8ba1\u7b97\u51fd\u6570\u7684\u57fa\u7c7b":63,"openblas\u7684\u8def\u5f84":48,"opprotomake\u5b9a\u4e49":63,"org\u76ee\u524d\u9075\u5faa":42,"out\u4e0b\u5305\u542b":81,"out\u7684\u6587\u4ef6\u5939":81,"outer_mem\u662f\u4e00\u4e2a\u5b50\u53e5\u7684\u6700\u540e\u4e00\u4e2a\u5411\u91cf":56,"output\u53ef\u4ee5\u662f\u4efb\u610f\u7ef4\u5ea6\u7684tensor":64,"output\u6587\u4ef6\u5939\u5b58\u653e\u8bad\u7ec3\u7ed3\u679c\u4e0e\u65e5\u5fd7":75,"output\u7684\u539f\u6709shape\u4fe1\u606f\u4e0d\u53d8":64,"packages\u91cc\u9762":46,"packages\u91cc\u9762\u7684python\u5305":46,"paddepaddle\u901a\u8fc7\u7f16\u8bd1\u65f6\u6307\u5b9a\u8def\u5f84\u6765\u5b9e\u73b0\u5f15\u7528\u5404\u79cdbla":48,"paddle\u4e00\u4e2a\u52a8\u6001\u5e93\u53ef\u4ee5\u5728\u4efb\u4f55linux\u7cfb\u7edf\u4e0a\u8fd0\u884c":33,"paddle\u4e0b\u8f7d":49,"paddle\u4e2d\u7684\u4e00\u6761pass\u8868\u793a\u8bad\u7ec3\u6570\u636e\u96c6\u4e2d\u6240\u6709\u7684\u6837\u672c\u4e00\u6b21":89,"paddle\u4e2d\u7ecf\u5e38\u4f1a\u5c06\u65f6\u95f4\u5e8f\u5217\u6210\u4e3a":56,"paddle\u4e8c\u8fdb\u5236\u5728\u8fd0\u884c\u65f6\u6355\u83b7\u4e86\u6d6e\u70b9\u6570\u5f02\u5e38":46,"paddle\u5185\u5d4c\u7684python\u89e3\u91ca\u5668\u548c\u5916\u90e8\u4f7f\u7528\u7684python\u5982\u679c\u7248\u672c\u4e0d\u540c":33,"paddle\u5185\u90e8\u7684\u7c7b\u4e3ac":33,"paddle\u662f\u4ece\u5b98\u65b9\u955c\u50cf\u6e90dockerhub":49,"paddle\u66ff\u6362\u6210dock":49,"paddle\u7684\u5404\u7248\u672c\u955c\u50cf\u53ef\u4ee5\u53c2\u8003":74,"paddle\u7684\u591a\u8bed\u8a00\u63a5\u53e3\u5b9e\u73b0\u5305\u62ec\u4e00\u4e0b\u51e0\u4e2a\u65b9\u9762":33,"paddle\u7684\u7c7b\u578b\u5168\u90e8\u9000\u5316\u6210":34,"paddle\u7684\u94fe\u63a5\u65b9\u5f0f\u6bd4\u8f83\u590d\u6742":33,"paddle\u7684c":34,"paddle\u7684dock":74,"paddle\u7684docker\u5f00\u53d1\u955c\u50cf\u5e26\u6709\u4e00\u4e2a\u901a\u8fc7":49,"paddle\u8bad\u7ec3\u4efb\u52a1":21,"paddle\u8def\u5f84\u4e0b":34,"paddle\u955c\u50cf":74,"paddle\u9700\u8981\u4e00\u4e2a\u591a\u8bed\u8a00\u63a5\u53e3":33,"paddle\u9700\u8981\u66b4\u9732\u7684api\u5f88\u591a":34,"paddle\u9759\u6001\u5e93\u94fe\u63a5\u590d\u6742":33,"paddle_\u7c7b\u578b\u540d":34,"paddle_\u7c7b\u578b\u540d_\u51fd\u6570\u540d":34,"paddlepaddle\u4e2d":[55,58],"paddlepaddle\u4e2d\u7684\u4e00\u4e2apass\u610f\u5473\u7740\u5bf9\u6570\u636e\u96c6\u4e2d\u7684\u6240\u6709\u6837\u672c\u8fdb\u884c\u4e00\u6b21\u8bad\u7ec3":88,"paddlepaddle\u4e2d\u7684\u8bb8\u591alayer\u5e76\u4e0d\u5728\u610f\u8f93\u5165\u662f\u5426\u662f\u65f6\u95f4\u5e8f\u5217":56,"paddlepaddle\u4e2d\u8fd8\u5305\u542b":46,"paddlepaddle\u4e2d\u901a\u8fc7reader\u6765\u52a0\u8f7d\u6570\u636e":51,"paddlepaddle\u4e3a\u4ea4\u53c9\u7f16\u8bd1\u63d0\u4f9b\u4e86\u5de5\u5177\u94fe\u914d\u7f6e\u6587\u6863":[53,54],"paddlepaddle\u4e3a\u6df1\u5ea6\u5b66\u4e60\u7814\u7a76\u4eba\u5458\u63d0\u4f9b\u4e86\u4e30\u5bcc\u7684api":51,"paddlepaddle\u4f1a\u5728\u8c03\u7528\u8bfb\u53d6\u6570\u636e\u7684python\u811a\u672c\u4e4b\u524d":84,"paddlepaddle\u4f1a\u81ea\u52a8\u8bbe\u5b9a":46,"paddlepaddle\u4f7f\u7528\u5747\u503c0":46,"paddlepaddle\u4f7f\u7528avx":46,"paddlepaddle\u4f7f\u7528git":42,"paddlepaddle\u4f7f\u7528swig\u5bf9\u5e38\u7528\u7684\u9884\u6d4b\u63a5\u53e3\u8fdb\u884c\u4e86\u5c01\u88c5":4,"paddlepaddle\u4fdd\u5b58\u7684\u6a21\u578b\u53c2\u6570\u6587\u4ef6\u5185\u5bb9\u753116\u5b57\u8282\u5934\u4fe1\u606f\u548c\u7f51\u7edc\u53c2\u6570\u4e24\u90e8\u5206\u7ec4\u6210":46,"paddlepaddle\u4fdd\u5b58\u7684\u6a21\u578b\u53c2\u6570\u6587\u4ef6\u524d16\u5b57\u8282\u4e3a\u5934\u4fe1\u606f":46,"paddlepaddle\u4fdd\u7559\u6dfb\u52a0\u53c2\u6570\u7684\u6743\u529b":2,"paddlepaddle\u5148\u4ece\u4e00\u4e2a\u6587\u4ef6\u5217\u8868\u91cc\u83b7\u5f97\u6570\u636e\u6587\u4ef6\u5730\u5740":47,"paddlepaddle\u53d1\u5e03\u65b0\u7248\u672c\u7684\u65f6\u5019\u90fd\u4f1a\u53d1\u5e03\u5bf9\u5e94\u7248\u672c\u7684\u751f\u4ea7\u955c\u50cf\u4ee5\u53ca\u5f00\u53d1\u955c\u50cf":49,"paddlepaddle\u53ef\u4ee5\u4f7f\u7528cudnn":48,"paddlepaddle\u53ef\u4ee5\u6267\u884c\u7528\u6237\u7684python\u811a\u672c\u7a0b\u5e8f\u6765\u8bfb\u53d6\u5404\u79cd\u683c\u5f0f\u7684\u6570\u636e\u6587\u4ef6":84,"paddlepaddle\u53ef\u4ee5\u6bd4\u8f83\u7b80\u5355\u7684\u5224\u65ad\u54ea\u4e9b\u8f93\u51fa\u662f\u5e94\u8be5\u8de8\u8d8a\u65f6\u95f4\u6b65\u7684":56,"paddlepaddle\u53ef\u4ee5\u901a\u8fc7\u8be5\u673a\u5236\u5224\u65ad\u662f\u5426\u5df2\u7ecf\u6536\u96c6\u9f50\u6240\u6709\u7684\u68af\u5ea6":62,"paddlepaddle\u5728\u5b9e\u73b0rnn\u7684\u65f6\u5019":56,"paddlepaddle\u5728\u6fc0\u6d3b\u51fd\u6570\u91cc\u5b9e\u73b0dropout":46,"paddlepaddle\u5b58\u7684\u662f\u6709\u503c\u4f4d\u7f6e\u7684\u7d22\u5f15":[2,51],"paddlepaddle\u5b9a\u4e49\u7684\u53c2\u6570":2,"paddlepaddle\u5c06\u4ee5\u8bbe\u7f6e\u53c2\u6570\u7684\u65b9\u5f0f\u6765\u8bbe\u7f6e":84,"paddlepaddle\u5c06\u5728\u89c2\u6d4b\u6570\u636e\u96c6\u4e0a\u8fed\u4ee3\u8bad\u7ec330\u8f6e":47,"paddlepaddle\u5c06\u6839\u636e":53,"paddlepaddle\u5c06\u6bcf\u4e2a\u6a21\u578b\u53c2\u6570\u4f5c\u4e3a\u4e00\u4e2anumpy\u6570\u7ec4\u5355\u72ec\u5b58\u4e3a\u4e00\u4e2a\u6587\u4ef6":47,"paddlepaddle\u5c06\u81ea\u52a8\u4e0b\u8f7d\u548c\u7f16\u8bd1\u6240\u6709\u7b2c\u4e09\u65b9\u4f9d\u8d56\u5e93":[53,54],"paddlepaddle\u5c06train":2,"paddlepaddle\u5f00\u53d1\u8fc7\u7a0b\u4f7f\u7528":42,"paddlepaddle\u5f00\u53d1\u955c\u50cf\u5c31\u88ab\u6784\u5efa\u5b8c\u6bd5\u4e86":49,"paddlepaddle\u63d0\u4f9b\u4e13\u7528\u7684":21,"paddlepaddle\u63d0\u4f9b\u4e86\u5f88\u591a\u4f18\u79c0\u7684\u5b66\u4e60\u7b97\u6cd5":47,"paddlepaddle\u63d0\u4f9bdocker\u955c\u50cf\u6765\u90e8\u7f72\u73af\u5883":50,"paddlepaddle\u652f\u6301\u4e0d\u540c\u7c7b\u578b\u7684\u8f93\u5165\u6570\u636e":51,"paddlepaddle\u652f\u6301\u4ee5\u4e0b\u4efb\u610f\u4e00\u79cdblas\u5e93":48,"paddlepaddle\u652f\u6301\u5927\u91cf\u7684\u8ba1\u7b97\u5355\u5143\u548c\u4efb\u610f\u6df1\u5ea6\u7684\u7f51\u7edc\u8fde\u63a5":47,"paddlepaddle\u652f\u6301\u975e\u5e38\u591a\u7684\u4f18\u5316\u7b97\u6cd5":46,"paddlepaddle\u652f\u6301sparse\u7684\u8bad\u7ec3":46,"paddlepaddle\u6587\u6863\u4e3b\u9898\u5728":65,"paddlepaddle\u6587\u6863\u4f7f\u7528":65,"paddlepaddle\u662f\u6e90\u4e8e\u767e\u5ea6\u7684\u4e00\u4e2a\u6df1\u5ea6\u5b66\u4e60\u5e73\u53f0":[47,51],"paddlepaddle\u6bcf\u6b21\u53d1\u5e03\u65b0\u7248\u672c\u90fd\u4f1a\u53d1\u5e03\u5bf9\u5e94\u7684\u5f00\u53d1\u955c\u50cf\u4f9b\u5f00\u53d1\u8005\u76f4\u63a5\u4f7f\u7528":49,"paddlepaddle\u6bcf\u6b21\u53d1\u65b0\u7684\u7248\u672c":42,"paddlepaddle\u6bcf\u6b21\u53d1\u7248\u672c\u9996\u5148\u8981\u4fdd\u8bc1paddlepaddl":42,"paddlepaddle\u7684\u4e3b\u7248\u672c\u5e93\u9075\u5faa":42,"paddlepaddle\u7684\u5185\u5b58\u5360\u7528\u4e3b\u8981\u5206\u4e3a\u5982\u4e0b\u51e0\u4e2a\u65b9\u9762":46,"paddlepaddle\u7684\u53c2\u6570\u4f7f\u7528\u540d\u5b57":46,"paddlepaddle\u7684\u5df2\u7ecf\u5b89\u88c5\u5b8c\u6210":53,"paddlepaddle\u7684\u6240\u6709layer\u90fd\u6709\u552f\u4e00\u7684nam":46,"paddlepaddle\u7684\u6570\u636e\u5305\u62ec\u56db\u79cd\u4e3b\u8981\u7c7b\u578b":2,"paddlepaddle\u7684\u6587\u6863\u5305\u62ec\u82f1\u6587\u6587\u6863":65,"paddlepaddle\u7684\u6587\u6863\u6784\u5efa\u6709\u4e24\u79cd\u65b9\u5f0f":65,"paddlepaddle\u7684\u6e90\u7801":61,"paddlepaddle\u7684\u7f16\u8bd1\u9009\u9879":50,"paddlepaddle\u7684bas":62,"paddlepaddle\u7684cmake\u7cfb\u7edf\u5c06\u6839\u636e\u8be5\u503c\u81ea\u52a8\u63a8\u5bfc\u548c\u8bbe\u7f6e\u9700\u8981\u4f7f\u7528\u7684\u4ea4\u53c9\u7f16\u8bd1\u5668":53,"paddlepaddle\u7684cmake\u7cfb\u7edf\u5c06\u6839\u636e\u8be5\u503c\u81ea\u52a8\u8bbe\u7f6e\u9700\u8981\u4f7f\u7528\u7684\u4ea4\u53c9\u7f16\u8bd1\u5668":54,"paddlepaddle\u7684cmake\u7cfb\u7edf\u624d\u8ba4\u4e3a\u5728\u662f\u5728\u4ea4\u53c9\u7f16\u8bd1raspberri":54,"paddlepaddle\u7684cmake\u7cfb\u7edf\u624d\u8ba4\u4e3a\u662f\u5728\u4ea4\u53c9\u7f16\u8bd1android\u7cfb\u7edf\u7684\u7248\u672c":53,"paddlepaddle\u7684docker\u5bb9\u5668\u4f7f\u7528\u65b9\u5f0f":50,"paddlepaddle\u7684docker\u955c\u50cf\u5c31\u5305\u542b\u4e86paddlepaddle\u7684python\u5e93\u4ee5\u53ca\u5176\u4f9d\u8d56\u7684\u591a\u4e2apython\u5e93":49,"paddlepaddle\u76ee\u524d\u53ea\u652f\u6301\u5728\u6bcf\u4e2a\u65f6\u95f4\u6b65\u4e2d":56,"paddlepaddle\u76ee\u524d\u552f\u4e00\u5b98\u65b9\u652f\u6301\u7684\u8fd0\u884c\u7684\u65b9\u5f0f\u662fdocker\u5bb9\u5668":49,"paddlepaddle\u76ee\u524d\u63d0\u4f9b\u4e24\u79cd\u53c2\u6570\u521d\u59cb\u5316\u7684\u65b9\u5f0f":46,"paddlepaddle\u76ee\u524d\u652f\u63018\u79cdlearning_rate_schedul":46,"paddlepaddle\u82e5\u68c0\u6d4b\u5230\u7528\u6237\u4f7f\u7528\u7684cmake\u7248\u672c\u4e0d\u4f4e\u4e8e3":53,"paddlepaddle\u8981\u6c42\u4f7f\u7528\u7684\u7f16\u8bd1\u5de5\u5177\u94fe\u6240\u652f\u6301\u7684andoid":53,"paddlepaddle\u8c03\u7528process\u51fd\u6570\u6765\u8bfb\u53d6\u6570\u636e":84,"paddlepaddle\u8d1f\u8d23\u5b8c\u6210\u4fe1\u606f\u548c\u68af\u5ea6\u5728\u65f6\u95f4\u5e8f\u5217\u4e0a\u7684\u4f20\u64ad":58,"paddlepaddle\u8d1f\u8d23\u5b8c\u6210\u4fe1\u606f\u548c\u8bef\u5dee\u5728\u65f6\u95f4\u5e8f\u5217\u4e0a\u7684\u4f20\u64ad":58,"paddlepaddle\u90fd\u53ef\u4ee5\u81ea\u52a8\u5224\u65ad\u786c\u4ef6\u662f\u5426\u652f\u6301avx":49,"paddlepaddle\u955c\u50cf\u9700\u8981\u63d0\u4f9b":75,"paddlepaddle\u9700\u8981\u7528\u6237\u5728\u7f51\u7edc\u914d\u7f6e":1,"paddlepaddle\u9700\u8981\u7684\u6240\u6709\u7f16\u8bd1\u5de5\u5177":49,"pass\u4e2a\u6a21\u578b\u5230\u7b2c":70,"pass\u5230":89,"pass\u5c06\u4e0d\u8d77\u4f5c\u7528":70,"pass\u8f6e\u5f00\u59cb\u8bad\u7ec3":70,"pass\u8f6e\u7684\u6a21\u578b\u7528\u4e8e\u6d4b\u8bd5":70,"passes\u8f6e":70,"patch\u53f7":42,"patch\u53f7\u52a0\u4e00":42,"path\u6307\u5b9a\u6d4b\u8bd5\u7684\u6a21\u578b":72,"period\u4e2a\u6279\u6b21\u5bf9\u6240\u6709\u6d4b\u8bd5\u6570\u636e\u8fdb\u884c\u6d4b\u8bd5":70,"period\u4e2a\u6279\u6b21\u6253\u5370\u65e5\u5fd7\u8fdb\u5ea6":70,"period\u4e2a\u6279\u6b21\u8f93\u51fa\u53c2\u6570\u7edf\u8ba1":70,"period\u4e2a\u6279\u6b21\u8f93\u51fa\u7b26\u53f7":70,"period\u4e2abatch\u5904\u7406\u7684\u5f53\u524d\u635f\u5931":88,"period\u4e2abatch\u7684\u5206\u7c7b\u9519\u8bef":88,"period\u6574\u9664":70,"period\u8f6e\u4fdd\u5b58\u8bad\u7ec3\u53c2\u6570":70,"pfsclient\u9700\u8981\u548cingress\u4e4b\u95f4\u505a\u53cc\u5411\u9a8c\u8bc1":28,"pfsclient\u9700\u8981\u5728\u4f20\u8f93\u5b8c\u6bd5\u6700\u540e\u4e00\u4e2achunk\u7684\u65f6\u5019\u68c0\u67e5destination\u6587\u4ef6\u7684md5\u503c\u662f\u5426\u548csource\u6587\u4ef6\u4e00\u81f4":28,"pfsserver\u63d0\u4f9brest":28,"pi\u5e73\u53f0\u4e0a\u9002\u7528\u7684paddlepaddle\u5e93":54,"pi\u5e73\u53f0\u4e0a\u9002\u7528\u7684paddlepaddle\u7684\u65b9\u6cd5\u548c\u6b65\u9aa4":54,"pi\u5e73\u53f0\u4f7f\u7528\u7684c":54,"pi\u5e73\u53f0\u53ef\u9009\u914d\u7f6e\u53c2\u6570":54,"pi\u7248\u672c\u7684\u5e93":54,"pi\u7248\u672cpaddlepaddle\u5e93\u65f6":54,"pi\u7cfb\u7edf":54,"pi\u7cfb\u7edf\u4e0a":54,"pi\u7cfb\u7edf\u7684\u7248\u672c":54,"pod\u4e2d\u7684\u5bb9\u5668\u5171\u4eabnet":73,"pod\u662fkubernetes\u7684\u6700\u5c0f\u8c03\u5ea6\u5355\u5143":73,"pooling\u5bf9\u7279\u5f81\u56fe\u4e0b\u91c7\u6837":81,"process\u51fd\u6570\u4f1a\u7528yield\u8bed\u53e5\u8f93\u51fa\u8fd9\u6761\u6570\u636e":84,"prod\u7684\u751f\u4ea7\u955c\u50cf":49,"public":[12,17,30,43,44,62,63,64,74,88],"py\u4e2d":42,"py_paddle\u91cc\u9762\u63d0\u4f9b\u4e86\u4e00\u4e2a\u5de5\u5177\u7c7b":4,"pydataprovider2\u4f1a\u5c3d\u53ef\u80fd\u591a\u7684\u4f7f\u7528\u5185\u5b58":2,"pydataprovider2\u63d0\u4f9b\u4e86\u4e24\u79cd\u7b80\u5355\u7684cache\u7b56\u7565":2,"pydataprovider2\u662fpaddlepaddle\u4f7f\u7528python\u63d0\u4f9b\u6570\u636e\u7684\u63a8\u8350\u63a5\u53e3":2,"pydataprovider2\u7684\u4f7f\u7528":[1,3,46,84,86],"pydataprovider\u4f7f\u7528\u7684\u662f\u5f02\u6b65\u52a0\u8f7d":46,"pypi\u4e0a\u7684package\u540d\u79f0\u4e3apaddlepaddle\u548cpaddlepaddl":42,"python\u4ee3\u7801\u5c06\u968f\u673a\u4ea7\u751f2000\u4e2a\u89c2\u6d4b\u70b9":47,"python\u53ef\u4ee5\u89e3\u9664\u6389\u5185\u90e8\u53d8\u91cf\u7684\u5f15\u7528":2,"python\u5b89\u88c5\u5305\u652f\u6301linux":46,"python\u5c01\u88c5\u7684\u5b9e\u73b0\u4f7f\u5f97\u6211\u4eec\u53ef\u4ee5\u5728\u914d\u7f6e\u6587\u4ef6\u4e2d\u4f7f\u7528\u65b0\u5b9e\u73b0\u7684\u7f51\u7edc\u5c42":62,"query\u6539\u5199":89,"rate\u4e3a0":89,"rate\u4e3a5":89,"rate\u88ab\u8bbe\u7f6e\u4e3a0":81,"reader\u7684\u4f7f\u7528\u65b9\u5f0f\u90fd\u662f\u4e00\u81f4\u7684":21,"reader\u8f93\u51fa\u7684data":21,"recommendation\u6587\u4ef6\u5939\u5185\u5b58\u653e\u8bad\u7ec3\u6587\u4ef6":75,"release\u9875\u9762":42,"repo\u7684\u6839\u76ee\u5f55":49,"request\u524d":61,"request\u7684":61,"request\u88ab\u5408\u5e76\u540e":61,"research\u5b9e\u9a8c\u5ba4\u641c\u96c6\u6574\u7406":85,"resnet\u6a21\u578b":83,"return":[2,6,8,9,11,12,14,15,16,17,21,22,24,27,30,32,35,39,43,44,45,46,47,51,56,59,62,64,75,82],"rnn\u5373\u65f6\u95f4\u9012\u5f52\u795e\u7ecf\u7f51\u7edc":56,"rnn\u5bf9\u4e8e\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u901a\u8fc7\u4e86\u4e00\u4e2alstm\u7f51\u7edc":56,"rnn\u603b\u662f\u5f15\u7528\u4e0a\u4e00\u65f6\u523b\u9884\u6d4b\u51fa\u7684\u8bcd\u7684\u8bcd\u5411\u91cf":58,"rnn\u6a21\u578b":84,"rnn\u76f8\u5173\u6a21\u578b":66,"rnn\u914d\u7f6e":57,"s3\u4e4b\u7c7b\u7684\u5206\u5e03\u5f0f\u5b58\u50a8\u4e4b\u4e0a":21,"search\u7684\u65b9\u6cd5":70,"sentences\u662f\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217\u7684\u6570\u636e":56,"seq\u53c2\u6570\u5fc5\u987b\u4e3afals":58,"server\u4e2a\u6279\u6b21\u6253\u5370\u65e5\u5fd7\u8fdb\u5ea6":70,"server\u4e4b\u4e0a":20,"server\u4e4b\u95f4\u7684\u7f51\u7edc\u5e26\u5bbd":20,"server\u4f1a\u6682\u505c\u53c2\u6570\u66f4\u65b0\u5e76\u7b49\u5f85":20,"server\u4f1a\u83b7\u53d6parameters\u5185\u5b58\u7684":20,"server\u5185\u5b58\u4e2d\u7684\u6a21\u578b\u6570\u636e\u7684\u5b8c\u6574\u955c\u50cf":20,"server\u540c\u6b65\u7684\u4fdd\u5b58\u4e00\u4e2a\u7279\u5b9a\u65f6\u95f4\u70b9\u7684\u5168\u5c40\u68c0\u67e5\u70b9":20,"server\u5728\u96c6\u7fa4\u4e2d\u542f\u52a8\u540e":20,"server\u6545\u969c\u540e\u88abkubernetes\u91cd\u65b0\u542f\u52a8":20,"server\u6b64\u65f6\u8fd8\u9700\u8981\u901a\u8fc7\u7f51\u7edc\u8bbf\u95ee\u5206\u5e03\u5f0f\u5b58\u50a8\u4ee5\u4fdd\u5b58\u5feb\u7167":20,"server\u751f\u6210\u4e00\u4e2auuid":20,"server\u7684\u5355\u70b9\u6216\u591a\u70b9\u540c\u65f6\u6545\u969c":20,"server\u7684\u6570\u636e\u5feb\u7167":20,"server\u7684\u68c0\u67e5\u70b9\u5404\u81ea\u72ec\u7acb\u4fdd\u5b58":20,"server\u7b2c\u4e00\u6b21\u542f\u52a8\u6216\u4efb\u610f\u65f6\u95f4paramet":20,"sh\u6765\u8bad\u7ec3\u6a21\u578b":81,"sh\u8c03\u7528\u4e86":82,"short":[8,9,43],"simd\u6307\u4ee4\u63d0\u9ad8cpu\u6267\u884c\u6548\u7387":46,"size\u4e3a1":89,"size\u4e3a50":89,"size\u4e3a512":70,"size\u53ef\u80fd\u4f1a\u5bf9\u8bad\u7ec3\u7ed3\u679c\u4ea7\u751f\u5f71\u54cd":46,"size\u5927\u5c0f\u4e3a128":88,"size\u662f3":89,"size\u672c\u8eab\u662f\u795e\u7ecf\u7f51\u7edc\u7684\u8d85\u53c2\u6570":46,"size\u7684\u503c":2,"softmax\u5c42":80,"softmax\u6fc0\u6d3b\u7684\u8f93\u51fa\u7684\u548c\u603b\u662f1":62,"sparse\u8bad\u7ec3\u9700\u8981\u8bad\u7ec3\u7279\u5f81\u662f":46,"srl\u4f5c\u4e3a\u5f88\u591a\u81ea\u7136\u8bed\u8a00\u5904\u7406\u4efb\u52a1\u4e2d\u7684\u4e2d\u95f4\u6b65\u9aa4\u662f\u5f88\u6709\u7528\u7684":87,"ssh\u65b9\u5f0f\u7684\u4e00\u4e2a\u4f18\u70b9\u662f\u6211\u4eec\u53ef\u4ee5\u4ece\u591a\u4e2a\u7ec8\u7aef\u8fdb\u5165\u5bb9\u5668":49,"ssh\u8fdb\u5165\u5bb9\u5668":49,"static":[14,24,34,43,44,79],"step\u51fd\u6570\u4e2d\u7684memori":58,"step\u51fd\u6570\u5185\u90e8\u53ef\u4ee5\u81ea\u7531\u7ec4\u5408paddlepaddle\u652f\u6301\u7684\u5404\u79cdlay":58,"subseq\u7684\u6bcf\u4e2a\u5143\u7d20\u662f\u4e00\u4e2a0\u5c42\u5e8f\u5217":55,"super":62,"swig\u652f\u6301\u7684\u8bed\u8a00\u6216\u8005\u89e3\u91ca\u5668\u6709\u5c40\u9650":33,"swig\u66b4\u9732\u7684\u63a5\u53e3\u4fdd\u7559\u4e86c":33,"swig\u751f\u6210\u7684\u4ee3\u7801\u4e0d\u80fd\u4fdd\u8bc1\u591a\u8bed\u8a00\u4ee3\u7801\u98ce\u683c\u7684\u4e00\u81f4\u6027":33,"swig\u76f4\u63a5\u8bfb\u53d6c":33,"swig\u9700\u8981\u5199\u4e00\u4e2ainterface\u6587\u4ef6":33,"swig_paddle\u4e2d\u7684\u9884\u6d4b\u63a5\u53e3\u7684\u53c2\u6570\u662f\u81ea\u5b9a\u4e49\u7684c":4,"switch":[17,34],"t\u6307\u5b9a\u4e86\u751f\u6210\u7684\u955c\u50cf\u7684\u540d\u5b57":49,"tag\u4e3a":42,"tensor\u5230\u5bf9eigentensor\u7684\u8f6c\u6362":64,"tensor\u5230eigentensor":64,"tensor\u5b9a\u4e49\u5728framework\u76ee\u5f55\u4e0b":64,"tensor\u662f\u4e00\u4e2a\u6b63\u5728\u5f00\u53d1\u4e2d\u7684\u6a21\u5757":64,"tensor\u6a21\u5757\u5bf9el":64,"tensor\u6a21\u5757\u6765\u5b9e\u73b0":63,"tensor\u6a21\u5757\u7684\u6587\u6863\u8f83\u5c11":64,"tensor\u6a21\u5757\u7684\u8be6\u7ec6\u4ecb\u7ecd\u8bf7\u53c2\u8003":64,"test\u548cgen\u8fd9\u4e09\u4e2a\u6587\u4ef6\u5939\u662f\u56fa\u5b9a\u7684":89,"tests\u7684paddlepaddl":61,"tflops\u4e86":67,"toolchain\u76ee\u5f55\u751f\u6210\u4e00\u5957\u7f16\u8bd1\u5de5\u5177\u94fe":53,"trainer\u63a5\u6536\u4e09\u4e2a\u53c2\u6570":51,"true":[6,7,8,9,11,12,14,15,17,22,32,38,44,46,56,59,62,72,75,82,86,87,88,89],"true\u8868\u793a\u53cd\u5411\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":59,"try":[10,18,19,22,23,24,38,43,46],"type\u5b57\u6bb5\u5747\u4e0d\u5c3d\u76f8\u540c":34,"type\u662fon":86,"ubuntu\u5b89\u88c5\u5305\u7684\u529f\u80fd\u6b63\u786e\u6027":42,"unit\u5728\u4e00\u4e2a\u65f6\u95f4\u6b65\u5185\u8ba1\u7b97\u5f97\u5230\u7684\u4e2d\u95f4\u503c":46,"unsupported\u6a21\u5757":63,"update\u53c2\u6570\u65f6\u624d\u6709\u6548":70,"utf8\u7f16\u7801":80,"uts\u7b49linux":73,"v1\u7248\u672c":46,"v2\u4e4b\u540e\u7684\u4efb\u4f55\u4e00\u4e2a\u7248\u672c\u6765\u7f16\u8bd1\u8fd0\u884c":48,"var":[17,30,31,32,37,41,45],"vector\u662frank\u4e3a1\u7684tensor":64,"vocab\u4e2d\u6bcf\u4e2a\u5207\u5206\u5355\u8bcd\u7684\u9884\u671f\u8bc4\u7ea7":88,"vocab\u505a\u4e3a\u5b57\u5178":88,"void":[17,22,24,30,33,34,37,44,62,63,64],"volume\u6302\u8f7d\u5230\u5bb9\u5668\u4e2d":73,"w0\u548c":82,"wbias\u662f\u9700\u8981\u5b66\u4e60\u7684\u53c2\u6570":82,"wheel\u5305":42,"while":[6,12,17,26,36,38,43,75,89],"wise\u8ba1\u7b97\u63d0\u4f9b\u4e86\u5f3a\u5927\u7684\u652f\u6301":64,"wmt14\u6570\u636e\u7684\u63d0\u4f9b\u6587\u4ef6\u5728":59,"words\u5373\u4e3a\u8fd9\u4e2a\u6570\u636e\u4e2d\u7684\u5355\u5c42\u65f6\u95f4\u5e8f\u5217":56,"words\u662f\u539f\u59cb\u6570\u636e\u4e2d\u7684\u6bcf\u4e00\u53e5\u8bdd":56,"x86_64\u548cmaco":46,"x\u4e0ey\u4e3a\u4e4b\u524d\u63cf\u8ff0\u7684\u8f93\u5165\u5c42":51,"x\u548cwindow":[49,53],"x\u548cwindows\u4e0a\u7684\u786c\u4ef6\u8d44\u6e90":49,"y\u8868\u793a\u8f93\u5165\u6570\u636e\u662f\u4e00\u4e2a\u7ef4\u5ea6\u4e3a1\u7684\u7a20\u5bc6\u5411\u91cf":51,"yaml\u6587\u4ef6\u4e2d\u5404\u4e2a\u5b57\u6bb5\u7684\u5177\u4f53\u542b\u4e49":75,"yaml\u6587\u4ef6\u63cf\u8ff0\u4e86\u8fd9\u6b21\u8bad\u7ec3\u4f7f\u7528\u7684docker\u955c\u50cf":75,"zero\u4e09\u79cd\u64cd\u4f5c":70,AGE:74,AWS:[21,73,76,77],And:[7,8,10,12,14,16,22,26,27,38,39,43],But:[8,9,12,16,30,43,46,79],EOS:8,For:[4,7,8,10,12,14,15,17,23,24,25,27,30,31,35,36,38,39,44,67,79],IDE:60,IDs:[12,26],IRs:39,Its:32,K8s:79,NFS:73,NMS:8,Not:[15,19,41,79],OPs:[39,40],One:[7,9,14,16,26,43],Ops:[41,44],PFS:28,QoS:74,TLS:[15,28],That:[8,12,38],The:[2,5,6,7,8,9,10,12,14,15,16,18,19,23,25,26,27,29,34,35,37,38,39,40,43,44,45,62,63,64,75,84,87,89],Their:[8,19],Then:[8,30,45],There:[7,8,12,14,15,17,18,19,24,26,27,29,35,36,43,45],These:[7,17,31,37,44],Use:[7,12,15,29,38,41],Used:[9,41],Using:[19,41,43,88],Will:[12,14],With:[8,9,37],YES:27,Yes:49,___embedding_0__:75,___embedding_1__:75,__doc__:44,__file__:27,__gradient_machines__:14,__impl__:44,__init__:[45,62],__main__:[4,82],__meta__:86,__metaclass__:63,__name__:[4,82],__param_conf__:14,__rnn_step__:59,__square_error_cost_0__:75,__tmp_params__:14,_binari:18,_block:45,_librari:18,_link:9,_name:45,_op:63,_proj:8,_recurrent_group:59,_res2_1_branch1_bn:82,_source_language_embed:[59,80],_target_language_embed:[59,80],_test:18,_var:45,_var_desc:45,a_op:63,aaaaa:21,abc:8,abl:[8,15,39,41,45,79],about:[8,9,17,18,27,29,35,39,43,44,87,89],abov:[2,7,8,15,17,18,19,23,30,31,35,37,39,44,67,79],abs:[8,9,16],abs_numerical_grad:16,acceler:20,accept:[6,8,12,15,38,41,87],access:[8,9,15,18,23,26,27,37,39],accord:[7,8,16,24,31,39,40,41,45],accordingli:[7,8],account:[41,79],accrodingli:22,accumul:[19,24],accur:[16,26],accuraci:7,acl:88,aclimdb:88,across:[8,12,35],act:[8,9,17,39,46,47,51,56,59],act_output:44,act_typ:84,activ:[9,13,18,39,44,46,51,59,84],activi:9,actual:[22,44],adadelta:[46,84],adagrad:84,adam:[15,24,46,84,88],adamoptim:[80,84,88,89],adapt:[7,10,14],add:[8,9,12,14,16,17,18,22,26,30,32,39,40,41,43,45,46,61,64],add_depend:18,add_execut:18,add_input:[35,62],add_memori:[17,35],add_output:35,add_test:[18,62],add_to:[8,46],add_two:[17,35],add_unittest_without_exec:62,add_var:45,addattr:[44,63],addbia:62,addcom:[44,63],added:[7,14,17,40,61],addinitialoper:45,addinput:[44,63],addit:[8,9],addop:30,addoutput:63,addr:19,address:[19,24,39,79],addrow:62,addtolay:8,addtyp:44,admin:79,administr:[26,79],advanc:16,advantag:16,adversari:38,affect:[8,17],afford:23,afi:2,aforement:18,after:[8,9,12,17,18,23,24,26,29,36,39,40,61],again:[15,19],age:[12,75,86],agg_level:[8,55,56],aggregatelevel:[55,56],ago:18,aircraft:89,airplan:81,aistat:8,alex:[8,88],alexnet_pass1:72,alexnet_pass2:72,algo_hrnn_demo:56,algorithm:[8,10,23,80,88,89],align:[8,9,12,89],all:[2,6,7,8,10,14,15,17,18,19,22,24,26,27,29,31,34,35,37,39,41,44,46,58,75,79,87,88],alloc:[6,24,27,64],allow:[15,24,39],allow_only_one_model_on_one_gpu:[69,70,72],alpha:18,alreadi:[18,19,39,40,43,46],also:[8,9,12,15,17,18,22,25,30,31,36,38,43,44,67,79,84],altogeth:79,alwai:[8,9,14,18,38,45,75],amazon:74,ambigu:38,amd64:73,amend:61,analysi:[87,88],ancestor:37,android:53,android_abi:53,android_arm_neon:53,android_native_api_level:53,android_standalone_toolchain:53,androideabi:53,ani:[8,9,12,15,18,19,24,26,27,38,39,40],annot:87,annual:87,anoth:[8,14,15,17,27,43,44],anroid_arm_mod:53,anyth:[12,38,87],api:[12,14,15,18,24,25,27,28,30,35,36,41,42,45,67,75,78,79,88],api_shar:18,api_test:18,api_trainer_config:86,api_trainer_config_helpers_data_sourc:86,api_trainer_config_helpers_lay:[59,84],api_trainer_config_helpers_layers_context_project:86,api_trainer_config_helpers_layers_cos_sim:86,api_trainer_config_helpers_layers_data_lay:86,api_trainer_config_helpers_layers_embedding_lay:86,api_trainer_config_helpers_layers_fc_lay:86,api_trainer_config_helpers_layers_pooling_lay:86,apiserv:73,apivers:[73,74,75],apo:89,appear:31,append:[2,14,38,56,59,75],append_gradient_machin:14,appleyard:67,appli:[8,31,43],applic:[31,41,67,74,79],applicaton:41,applyl1:22,approach:[8,36,39,40,79,84],apt:81,arbitrari:[8,39],arch:53,architectur:89,archiv:[12,33,34],arg:[2,7,9,44,46,47,63,75,81,82,84,87,88],arg_nam:8,argpars:75,args_ext:75,argument:[2,8,12,17,22,23,36,75,87],argumentpars:75,argv:82,arm64:53,arm:[53,54],arm_soft_fp_abi:53,armeabi:53,around:[2,8,26,79],arrai:[4,6,8,12,14,24,31,38,41,46,51,63,82],articl:[31,61],arxiv:[8,9,88],as_row_vector:8,ask:[19,26],assert:4,assign:[7,8,23,37,79],associ:[36,87],assum:[7,8,17,39],astyp:[38,63],asyc:19,async:[19,69],async_count:[69,70],async_lagged_grad_discard_ratio:70,async_lagged_ratio_default:[69,70],async_lagged_ratio_min:[69,70],asynchron:19,atla:48,atlas_root:48,attenion:9,attent:[8,9,89],attr1:8,attr2:8,attr:[6,8,9,17,37,44,46,59,63],attr_map:44,attrdesc:37,attribut:[8,9,13,17,37,40,41,43,44],attributemap:63,attrproto:44,attrtyp:[37,44,63],attrvalu:44,auc:69,author:[28,73],authorized_kei:68,auto:[17,22,33,41,43,62,63,64,67],automat:[8,15,24,30,39,40,41,44,89],automaticli:8,automobil:81,avail:[19,24,40,79],averag:[7,8,11,23,87],average_test_period:[69,70,87],averagepool:8,avg:[55,67,84],avgcost:[84,86,88,89],avgpool:[8,84],avoid:[16,17,19,36,67],avx:49,await:74,awar:[15,35],aws:28,axi:46,aync:40,azur:73,b2t:80,b363:74,b8561f5c79193550d64fa47418a9e67ebdd71546186e840f88de5026b8097465:74,back:[14,19,39],background:[7,8],background_id:[7,8],backpropag:16,backward:[5,8,9,16,17,22,24,36,62,63],backward_first:59,backward_op:16,backwardactiv:62,baidu:74,bake:39,balanc:40,balasubramanyan:88,bank:87,bare:[73,74,79],barrierstatset:67,base:[7,8,11,12,15,23,37,39,60],baseactiv:9,baseev:14,basematrix:62,basenam:7,basepoolingtyp:[8,9],basestr:[6,7,8,9,11,14],bash:[49,60,61,74,75],basic:[8,14,37,41],batch:[8,9,10,12,14,15,17,19,21,22,35,47,51,68,74,75,81,82,84,86,88,89],batch_0:82,batch_id:[14,51],batch_norm_lay:9,batch_norm_typ:8,batch_read:[21,38],batch_siz:[12,46,47,51,68,80,81,84,86,88,89],batchsiz:[8,62],bazel:18,bbbbb:21,bbox:[7,8],bcd:8,bcebo:12,bcm2708:54,bdist_wheel:42,beam:[8,59,89],beam_gen:[8,59],beam_search:[14,58,59],beam_siz:[8,59,69,70,72],beamsiz:89,becaus:[7,8,12,15,17,18,19,24,36,38,39,43,45,56],becom:[40,43],been:[18,23,87],befor:[8,9,19,26,29,31,36,38,46,63,79],begin:[7,8,22,24,29,31],beginiter:[14,15],beginn:59,beginpass:[14,15],begintrain:15,being:[26,38],belong:[8,39,43],below:[8,17,19,24,36,38,39,40,45],benefit:[9,26,27],bengio:8,bertolami:88,besid:[8,12,39],best:[8,18,86],best_model_path:87,besteffort:74,beta1:10,beta2:10,beta:82,better:[8,9,18,79],between:[7,8,10,14,18,19,24,34,36,40,41,43,89],bgr:82,bi_gru:9,bi_lstm:9,bia:[8,9,10,62,82],bias:8,bias_attr:[8,9,46,47,56,59],bias_param:46,bias_param_attr:9,biases_:62,biasparameter_:62,biassiz:62,bidi:74,bidirect:[8,9,87,88],bidirectional_lstm:46,bidirectional_lstm_net:88,big:[40,79],bigger:19,bilinear:8,bilinear_interpol:8,bilinearfwdbwd:67,bin:[49,68,73,74,75,86],binari:[7,8,12,18,27,39],bind:[30,43],bird:81,bitext:89,blank:8,bleu:89,bloack:32,block:[8,20,22,24,32,35,37,39,40,45,64],block_i:8,block_x:8,blockdesc:[17,31,37,41],bn_bias_attr:9,bn_layer_attr:9,bn_param_attr:9,bollen:88,book:[12,41,59],bool:[6,7,8,9,11,12,14,17,44,45,62,70,72,88],boot:[8,58,59,79],boot_bia:8,boot_bias_active_typ:8,boot_lay:[56,59],boot_with_const_id:8,bootstrapp:79,bos_id:[8,59],both:[5,6,8,9,15,17,18,19,26,39,40,45],bottom:[14,84],bound:8,boundari:39,boundri:7,box:8,brace:[17,31],brain:26,branch:[8,15,17,18,32,39,42,61],brendan:88,brief:[18,24,64],broadcast:[19,41,79],broken:61,browser:49,bryan:88,bsd:60,buf:22,buf_siz:12,buffer:[12,22,38,43],buffer_s:12,buffered_read:38,bug:61,build:[8,12,18,27,31,39,42,44,49,53,54,60,61,65,75,76,77,89],build_dict:12,build_doc:65,built:[18,39,44,67,79],bunk:88,c11:33,c703c041:61,c99:34,cach:[39,46,87],cache_pass_in_mem:[2,46,87],cachetyp:[2,46,87],cacul:9,caff:17,caffe2:17,caffe_poli:46,calc_batch_s:[2,87],calcul:[7,8,9,10,16,19,24],call:[7,8,9,14,15,16,17,22,23,24,25,27,31,35,37,39,41,43,44,67,75,84],callabl:[6,8,12],callback:62,caller:16,calrnn:56,caltech:81,can:[6,7,8,9,12,14,15,16,17,18,19,22,23,26,27,30,31,32,35,37,38,39,40,41,44,45,67,79],can_over_batch_s:[2,87],cancel:26,candid:8,cannot:[41,43,46],capabl:36,capi:33,capi_prvi:34,caption:89,card:68,care:[9,27,38,79],caret:14,cast_to_op_attr:44,cat:[49,75,81,82,88],categori:[8,12,19],categorig:12,categoryfil:74,caus:[19,29],cc_:18,cc_binari:18,cc_test:18,ccb2_pc30:89,cclient:25,cde:8,cdn:12,cdot:8,ceil:8,ceil_mod:8,cell:[8,9],cento:79,ceph:[21,73],cephf:[21,27,28],certain:43,certif:[15,28,73],cffi:33,cfg:74,cgo:33,chain:[12,31],challeng:[8,19],chanc:15,chang:[8,12,18,23,27,36,37,38,43,61,88],channel:[8,9,67],chapter:35,chapter_data:35,chapter_out:35,check:[2,12,17,18,41,46,61,62,70],check_align:12,check_attr:44,check_eq:62,check_fail_continu:2,check_grad:[16,63],check_l:62,check_sparse_distribution_batch:[69,70],check_sparse_distribution_in_pserv:[69,70],check_sparse_distribution_ratio:[69,70],check_sparse_distribution_unbalance_degre:[69,70],checker:41,checkgrad:70,checkgrad_ep:70,checkmark:79,checkout:61,checkpoint:40,checksum:28,child:17,choic:18,chunk:[23,28],chunk_schem:7,chunktyp:7,cifar:81,cifar_vgg_model:81,circl:31,claimnam:75,clang:[33,49,61],clarifi:7,class1:88,class2:88,class_dim:88,classic:8,classif:[8,31,84,88,89],classifi:[8,82],classification_cost:[39,46,56,81,84],classification_error_evalu:[7,84,88,89],classification_evalu:7,clean:[17,18,36,41,46,61],clear:[7,18,36,43],clearer:36,clearli:43,client:[22,25,41,73],clip:[6,9,70,84],clock:8,clone:[8,49,54,60],close:[2,38,61],cloud:[18,19,27,28,29,41,79],cloud_read:12,cluster:[12,14,15,17,19,24,39,68,73,75],cluster_train:68,cmake:[34,46,48,53,54,60,61,63,65,67],cmake_build_typ:[53,54],cmake_c:[53,54],cmake_system_nam:[53,54],cmakelist:[18,62],cmatrix:[33,34],cna:8,cnn:[8,74,84],coars:30,code:[2,4,8,12,15,18,26,30,31,36,38,39,40,41,44,45,49,62,74,86],codebas:41,coded_stream:46,codedinputstream:46,coeff:8,coeffici:8,collabor:19,collect:[8,12,14],collectbia:62,color:[81,82],colour:12,column:[7,8,31,38],column_evalu:7,com:[8,9,12,18,49,53,54,60,61,73,74,79,82],combin:[7,8,9,12,14,43],comma:[14,24],command:[18,22,27,29,60,62,72,74,75,76,77],commandlin:[67,75],comment:[10,18,44,56,75],commit:[18,74],common:21,common_util:68,commonli:29,commun:[19,24,25,39,40],compar:[16,18,41],compare_grad:63,comparison:18,compat:30,compelet:41,compil:[8,18,37,39,44,45,53,54,60],complaint:18,complet:[8,9,10,12,14,17,19,23,24,28,31,41,74,75,79],complex:[9,26,38,41],complic:[8,30,39],compon:39,compos:[12,15,30,35,41],composenotalign:12,composit:30,compress:23,comput:[8,9,15,16,19,39,40,43,45,63,64,87,88],computaiton:41,computation:8,conat:8,concat:[59,89],concaten:[9,35,46],concentr:41,concept:[7,15,30,35,36,43,45],concern:15,concurr:[19,26,40],cond:[17,32],condit:[8,23,39,74],conf:[4,8,46,56,68,75,80,82,89],conf_paddle_gradient_num:75,conf_paddle_n:75,conf_paddle_port:75,conf_paddle_ports_num:75,conf_paddle_ports_num_spars:75,confid:8,confidence_threshold:8,config:[6,8,21,29,47,51,62,68,69,70,72,73,74,75,79,81,84,86,87,88,89],config_:[22,70],config_arg:[69,70,72,82,84,87,88],config_bas:[7,8,14],config_fil:87,config_gener:[68,86],config_lay:62,config_len:24,config_pars:[4,62],config_proto:24,configprotostr:46,configur:[8,14,22,24,26,27,40,62,79,80,82,89],confirm:29,conflict:[43,61],confront:89,conll05st:87,conll:[12,87],connect:[9,27,39,40,46,74,79,84],connectionist:[8,88],connor:88,consequ:[8,9],consid:[7,8,10,79],consider:9,consist:[7,8,12,23,38,44],consolid:17,const_cast:63,constant:46,constraint:[39,43],construct:[7,15,35,41,43,44,45],constructbackwardgraph:31,constructoptimizationgraph:31,constructor:[41,43,44],consum:19,contact:26,contain:[2,7,8,9,11,12,14,15,17,23,36,41,43,44,60,74,75],content:[24,29],content_len:24,context:[2,8,9,12,43,46,59,63,64,73],context_attr:9,context_len:[8,9,84],context_proj_layer_nam:9,context_proj_param_attr:9,context_project:9,context_start:[8,9,84],continu:[7,19],contrast:8,contributor:41,control:[6,17,74,79,89],conv:9,conv_act:9,conv_batchnorm_drop_r:9,conv_bias_attr:9,conv_filter_s:9,conv_layer_attr:9,conv_num_filt:9,conv_op:8,conv_pad:9,conv_param_attr:9,conv_strid:9,conv_with_batchnorm:9,conveni:[15,44],convent:24,convers:39,convert:[2,4,12,21,38],convlay:8,convolut:[8,9,12],convoper:8,convtran:8,convtranslay:8,cool:61,coordin:[19,24],copi:[14,15,23,26,29,31,35,86],core:[6,34,36],coreo:79,corner:41,corpora:89,corpu:[12,87],correct:[8,16],correctli:[7,12],correspoind:15,correspond:[15,17,18,35,41,44,46],corss_entropi:15,cos:[8,44],cosin:[8,44],cosineop:44,cosineopproto:44,cosineopprotomak:44,cost:[10,14,15,31,37,39,47,51,86,88,89],cost_id:8,cost_val:39,could:[8,12,14,15,16,23,35,36,37,38,39,40,41],count:[7,19,27,38,67,70,72,74,86,87,88,89],counter:[19,23,31],cours:[7,27],cpp:[16,22,30,33,34,36,40,45,46,56,62,67,75,84,86,89],cpu:[2,6,8,16,27,36,39,41,42,49,60,63,64,67,72,74,75,87],cpuinfo:49,cpuplac:[63,64],cpusparsematrix:34,crash:[19,67],creat:[6,8,12,14,15,16,17,19,24,28,29,30,31,35,36,45,46,51,61,62,74,75,79],create_bias_paramet:62,create_doc_str:44,create_ifelseop:[17,32],create_input_paramet:62,create_op:63,create_oper:30,create_python_ops_creatation_funct:44,create_rnn:17,create_rnn_op:35,createfromconfigproto:4,createop:44,createoper:17,createvar:45,createvari:17,creation:30,creator:[12,21,41],credenti:29,crf:87,critic:88,crlf:61,crop:82,crop_siz:82,cross:[8,46,53,54,84],cross_entropi:[8,15],cross_entropy_with_selfnorm:8,crt:[28,73],csc:62,cslm:89,csr:62,csv:46,ctc:7,ctc_evalu:7,ctest:[49,60,61,63],ctrl:[60,68,86],ctx:[63,64,87],ctx_0:87,ctx_0_slot:87,ctx_n1:87,ctx_n1_slot:87,ctx_n2:87,ctx_n2_slot:87,ctx_p1:87,ctx_p1_slot:87,ctx_p2:87,ctx_p2_slot:87,cub:81,cuda:[18,60,67,68,70],cuda_dir:[69,70],cuda_so:[46,49],cuda_visible_devic:46,cudaconfigurecal:67,cudadevicegetattribut:67,cudaeventcr:67,cudaeventcreatewithflag:67,cudafre:67,cudagetdevic:67,cudagetdevicecount:67,cudagetdeviceproperti:67,cudagetlasterror:67,cudahostalloc:67,cudalaunch:67,cudamalloc:67,cudamemcpi:67,cudaprofilerstart:67,cudaprofilerstop:67,cudaprofilestop:67,cudaruntimegetvers:67,cudasetdevic:67,cudasetupargu:67,cudastreamcr:67,cudastreamcreatewithflag:67,cudastreamsynchron:67,cudeviceget:67,cudevicegetattribut:67,cudevicegetcount:67,cudevicegetnam:67,cudevicetotalmem:67,cudnn:[8,11,18],cudnn_batch_norm:8,cudnn_conv:8,cudnn_conv_workspace_limit_in_mb:[69,70],cudnn_convt:8,cudnn_dir:[69,70],cudnnv5:48,cudrivergetvers:67,cuinit:67,cumul:8,curl:73,curli:[17,31],current:[2,8,17,18,19,22,24,35,36,37,40,43,45,73],current_block:37,current_oper:37,current_word:[46,59],currentcost:[84,86,88,89],currentev:[84,86,88,89],curv:15,custom:[15,27,41],custom_batch_read:38,cut:40,cutoff:12,cxx_compil:[53,54],cxx_flags_minsizerel:[53,54],cycl:19,cyclic:8,cython:33,dalla:2,dan:87,dandroid_abi:53,dandroid_arm_mod:53,dandroid_arm_neon:53,dandroid_standalone_toolchain:53,darwin:73,dat:[21,68,86],data:[2,7,10,14,15,16,17,21,22,23,28,31,35,36,37,40,41,43,44,45,46,49,51,56,59,64,68,69,74,75,76,80,81,82,84,86,87,88,89],data_batch:46,data_config:4,data_dir:[80,81,88,89],data_fil:47,data_lay:[2,22,46,47,56,81,84,87],data_read:[12,38],data_reader_creator_random_imag:38,data_typ:[12,45,51,59],databas:12,datacent:[21,29],datacenter1:21,datacenter2:21,datacenter_1:21,datacenter_2:21,datacenter_nam:21,datadim:8,dataprovid:[1,46,47,68,75,84,86,87],dataprovider_:84,dataprovider_bow:84,dataprovider_emb:84,dataproviderconvert:4,datasci:8,dataset:[21,27,38,51,59,82,85,88,89],datatyp:45,date:87,db_lstm:87,dcmake_build_typ:65,dcmake_install_prefix:[53,54],dcmake_system_nam:[53,54],dcudnn_root:48,ddim:64,dead:19,deal:79,deb:61,debian:49,debug:[16,29,39,65],decai:10,decar:12,decayr:22,decent:23,decid:[15,26,38],declar:[8,17,35],decod:[8,9,58,59,89],decoder_boot:59,decoder_group_nam:59,decoder_input:[46,59],decoder_mem:59,decoder_prev:9,decoder_s:[46,59],decoder_st:[9,59],deconv:8,deconvolut:8,decor:[2,12],deduc:41,deep:[8,26,31,67,81,82],deer:81,def:[2,4,8,12,15,16,21,27,30,35,38,44,45,46,47,51,56,59,62,63,75,82,84,87],defalut:8,default_decor:75,default_devic:72,default_valu:[32,72],defect:36,defer:26,defin:[2,8,9,12,14,15,17,18,19,26,30,35,37,38,40,41,43,44,46,51,63,84],define_py_data_sources2:[2,46,47,81,82,84],defini:89,definit:[17,19,23,39,44,80],definiton:30,degre:8,delet:[27,28,61],delimit:[7,46,86],deliv:79,delta:[8,16],delv:8,demand:19,demo:[4,8,12,41,68,74,76,80,81,82,84,86,88,89],dens:[8,12,24,25,86],dense_arrai:12,dense_vector:[2,4,12,47,51],dense_vector_sequ:12,dep:18,depend:[17,18,19,27,39,40,45,79],deploi:[8,79],deploy:[41,79],deprec:8,depth:17,dequeu:40,deriv:[5,15],desc:[17,44],desc_:17,descent:[8,19],describ:[15,17,18,23,36,37,41,44,45,74],descript:[7,17,18,45,75],deseri:[14,36,41],desgin:31,design:[8,12,22,33,79],desir:[19,74],destin:[24,29],destroi:17,destruct:43,det_output:7,detail:[6,7,8,9,10,16,23,27,29,35,39,43,79],detect:61,detection_evalu:7,detection_output:7,determin:[8,12,17,41],dev:[41,46,49,60,61,79,81,89],dev_ctx:17,develop:[18,36,42,61,89],deviat:6,devic:[6,36,39,41,46,49,64,72],device_context:63,device_context_:63,devicecontext:[17,63],deviceid:72,devid:8,dez:88,dhcp:79,diagram:35,dict:[2,7,12,14,46,56,75,86,88,89],dict_dim:[46,56,88],dict_fil:[7,56,87],dict_siz:[12,22],dictionari:[2,7,8,12,14,15,16,46,84,89],dictrionari:84,dictsiz:89,did:36,diff:61,diff_mat:16,differ:[7,8,14,17,18,19,24,26,39,40,43],differenti:30,difficult:[7,16],dig:41,digit:8,dilat:8,dilation_i:8,dim0:63,dim1:63,dim:[8,12,22,35,41,45,62,63,64,80,88],dim_:64,dimens:[5,8,11,12,41,46,64],dimes:8,dir:[53,68,75,82,86,87,88,89],direct:[8,9],directli:[9,18,25,27,36,39,44],directori:[8,18,21,26,28,29,61,67,74],disabl:46,discard:[12,19,23,61,70],discexp:46,discount:8,discov:19,discuss:[15,17,23,24,25,39],dispatch:[36,39],displai:[27,29],disput:89,dist:[42,46],dist_train:[15,27],distanc:7,distinguish:18,distribut:[8,17,23,24,25,26,40,41,45,70,76,77,79],distribute_test:[69,70],distributedli:39,disucss:15,divid:[10,44,45],diy_beam_search_prob_so:[69,70],dmkl_root:48,do_forward_backward:38,doc:[4,12,35,49,63,65,75],doc_cn:65,doc_them:65,docker:[42,46,49,61,74,75,76,77,79],docker_build:15,docker_push:15,dockerfil:[49,60,61,75],dockerhub:49,document:[9,16,28,31,35,41],documentari:2,doe:[9,19,23,24,26,27,36,38,39,40,45],doesn:[6,8,12,15,17,38],dog:[81,82],doing:[22,26,31,39,41],don:[9,15,18,30,31,38],done:[7,8,9,18,19,23,24,39,61,67,75],dot:[63,89],dot_period:[70,72,75,81,86,88,89],dotmuloper:8,dotmulproject:8,doubl:[8,31,39,63,70],down:67,download:[12,18,19,22,26,28,74,79],download_cifar:81,doxygen:61,dozen:18,dpython_execut:46,dpython_include_dir:46,dpython_librari:46,drive:43,drop:9,drop_fc:46,drop_rat:[6,46],dropout:[6,9,46],dropout_r:[8,46],drpi_arm_neon:54,drpi_toolchain:54,drwxr:74,dst:24,dtoh:67,dtype:[4,46,47,82],dubai:89,due:[23,26],dummi:[14,23],duplic:40,durat:23,dure:[2,8,9,17,19,23,26,27,41,79,89],dwith_c_api:[34,53,54],dwith_doc:65,dwith_gpu:[48,54,65],dwith_mkldnn:65,dwith_mklml:65,dwith_profil:67,dwith_python:[34,54],dwith_swig_pi:[34,53,54],dwith_test:63,dwith_tim:67,dynam:[2,24,34,38],dynamic_cast:62,e2e:79,each:[2,7,8,9,11,12,14,16,17,18,19,22,23,24,26,27,31,35,36,37,38,39,40,41,43,44,79,86],each_feature_vector:5,each_pixel_str:2,each_time_step_output:5,each_word:2,eaqual:8,eas:[12,38],easi:[16,38,41],easier:[15,38,40],easili:[15,38,43],ec2:73,echo:[46,49,86,88],edit:7,edu:[12,74,81],effici:[8,39],effort:39,efg:8,eigen:[41,63],eigen_use_gpu:63,eigenmatrix:64,eigentensor:64,eigenvector:64,either:[8,12,14,15,35],electron:74,elem_dim:8,element:[7,8,9,12,14,16,23,32,38,40,41,63],element_typ:24,elementori:41,elementwis:41,elif:[15,44],els:[15,22,27,39,40,43,44,49,56,62,82],emac:60,emailweixu:18,emb1:[22,56],emb2:[22,56],emb:[46,56,74,84],emb_group:56,emb_para:46,emb_param_fil:46,emb_sum:46,embed:[15,17,22,40,46,59,80,86,88],embedding_lay:[22,46,56,84],embedding_nam:[8,59],embedding_s:[8,59],empir:8,emplace_back:62,emploi:44,empti:[7,12,19,47],enabl:[6,8,17,18,23,40,60,67],enable_grad_shar:[69,70],enable_parallel_vector:70,enc_proj:[9,59],enc_seq:9,enc_vec:59,encapsul:24,encod:[9,23,56,89],encoded_proj:[9,59],encoded_sequ:[9,59],encoded_vector:59,encoder1:56,encoder1_expand:56,encoder1_rep:56,encoder2:56,encoder2_rep:56,encoder_last:8,encoder_s:59,encount:22,encourag:39,end2end:79,end:[7,8,14,17,36,38,43,59,61,87,88],end_pass:15,end_po:8,enditer:[14,15,51],endpass:[14,15],endpoint:[12,21],endtrain:15,engin:27,english:[8,89],enlarg:63,enough:17,enqueu:40,ensembl:9,ensur:[19,43],enter:17,enterpris:41,entir:[8,9,24,26],entiti:[7,17,43],entri:[12,23,27],entropi:[8,84],entry_point:27,enueu:40,enumer:[5,46],env:[46,75],environ:[15,46,67,74],eol:61,eos_id:[8,59],epsilon:10,equal:[8,9,19,56,63],equat:[7,8,9,10,63],equival:[8,15,17,44,79],error:[6,7,8,9,10,15,16,23,29,43,46,63,70,84,86,88,89],error_clipping_threshold:[6,46],especi:9,essenc:15,essenti:[8,15],estim:[8,15,40],eta:74,etc:[7,12,17,38,39,43,79,89],etcd:[12,14,19,23,24,26],etcd_addr:24,etcd_endpoint:12,eth0:[68,75],eval:[7,17,41,45,84,86,88,89],eval_bleu:89,evalu:[8,13,14,26,39,67,68,86,88,89],evaluate_difficult:7,evaluate_pass:88,even:[15,38],evenli:24,event:[51,74],event_handl:[14,15,51],everi:[7,8,9,12,15,19,23,24,26,31,40,43,44],everyth:[39,40],evid:36,exactli:[8,9,46],exampl:[7,8,9,10,12,14,17,27,29,30,31,35,36,38,40,41,82],exc_path:46,exceed:8,except:[8,12,26,31],exchang:[36,41],excluded_chunk_typ:7,exconv:8,exconvt:8,exdb:12,exe:73,exec:49,execut:[18,19,23,27],executioncontext:[63,64],exist:[15,17,19,29,38,44,64,88],exit:[24,29,74],exp:46,expand:56,expand_a:[8,55,56],expand_lay:56,expand_level:[8,55],expandconvlay:8,expandlevel:55,expans:8,expect:8,expert:18,expir:19,explain:[7,19,30,31],explan:[8,16,27,39,43],explicit:62,explicitli:15,explor:8,exponenti:5,expos:25,express:[15,40],ext:8,extend:[7,40],extens:[26,40],extent:34,extern:[18,33,34,41],external_librari:18,extra:[6,8,9,39,79],extra_lay:14,extraattr:[6,72],extraattribut:8,extraattributenon:8,extract:[7,8,36,39,80,82,87],extract_fea_c:82,extract_fea_pi:82,extralayerattribut:[6,9,46],extralayeroutput:9,extrem:8,f1205:46,f120da72:74,fa0wx:74,fabric:68,face:18,facotr:8,factor:[6,8,10],factori:33,fail:[19,23,46,70,74],failur:[19,24],fake_imag:38,falloc:28,fals:[6,7,8,9,10,12,16,17,32,35,38,46,47,51,56,59,62,63,72,74,80,84,87,88,89],false_block:[17,32],false_label:38,false_read:38,faq:78,fast:[8,23,67],faster:[8,9,19],fault:[14,23,41],fbd1f2bb71f4:74,fc1:[45,62,72],fc2:[45,72],fc3:72,fc4:72,fc_act:9,fc_attr:9,fc_bias_attr:9,fc_layer:[44,46,47,56,72,84,86],fc_layer_nam:9,fc_mat:14,fc_op:44,fc_out:17,fc_output:44,fc_param:46,fc_param_attr:9,fc_without_b:17,fclayer:62,fcop:30,fdata:[56,87],fea:82,fea_output:82,feat:88,featur:[2,5,8,12,61,82,86,87],feature_a:46,feature_b:46,feature_map:86,feed:[9,14,15,31,35,39,51],feeder:12,fernan:88,festiv:2,fetch:[12,19,22,39,46,61],few:[18,19,38],fewer:8,fg0:8,field1:14,field2:14,field:[8,14,17,44,46,86],fifth:31,figur:[15,18,35,40,80,89],file1:89,file2:89,file:[2,7,8,12,14,15,18,19,21,23,24,26,27,28,29,34,38,39,45,61,63,79,82,85,88,89],file_list:2,file_nam:[46,47,56,82,84,87],filenam:[2,21,46,86],fileoffset:28,filer:8,filesystem:[26,27,28,39],fill:[8,19,23],fill_zero_grad:41,filter:[8,9],filter_s:[8,9],filter_size_i:8,find:[8,10,17,19,26,43],find_var:16,findop:17,findvar:[17,43],fine:[6,23,30],finish:[19,23,26,27,44,74],first:[8,14,15,17,19,23,26,27,29,31,35,37,39,41,44,63,64,79,84],first_seq:59,firstli:[7,8],firstn:12,firstseen:74,fit:[12,41],five:37,fix:[6,33,39,61],flag:12,flatten:64,flatten_result:46,flatten_s:45,flexiabl:38,flexibl:[8,9,15,24,31,35,39],flight:89,float32:[4,12,30,38,46,47,63,82],floor:[8,46],flow:[17,35,42],fmt:46,fname:46,folder:[18,21,27,29,89],follow:[7,8,9,10,12,14,15,16,17,18,19,23,27,30,31,35,37,38,40,41,43,44,76,77,79],forbid:15,force_load:33,forest:17,forget:[10,15],fork:8,form:[8,9,10],format:[7,12,14,16,23,61,62],former:[15,18],formula:[8,9,16],formular:8,forward:[5,8,9,16,17,22,24,36,41,62],forward_op:16,forwardactiv:62,forwardtest:4,found:[8,43],four:[7,31],fp16:[41,45],fp32:[41,45],fp64:45,fparam:46,frame:[7,41],framework:[15,17,41,43,44,45,63],free:[12,79],french:89,frequenc:12,frequent:[23,38],fresh:26,friend:43,frog:81,from:[2,4,7,8,9,12,14,16,17,18,19,21,22,23,24,29,30,31,32,35,36,38,39,40,41,43,46,47,49,58,63,64,67,74,79,80,81,84,86,87,88,89],from_no_sequ:[8,55],from_sequ:[8,55],from_tar:14,fromfil:[38,46,47,82],fromstr:46,fulfil:67,full:[8,19,26,35,79],full_matrix_project:[9,56,59],fullfil:45,fulli:[39,40,79,84],fullmatrixproject:8,fullsiz:22,fully_matrix_project:9,fullyconnectedlay:62,func:[2,12,23],funciton:9,functor:30,fundament:[40,41],further:[8,44,79],futur:[8,26,39,41],fvs:44,gain:8,gamma:82,gan:15,gangliao:18,gate:[8,9],gate_act:[8,9,56],gate_attr:8,gate_bias_attr:8,gate_param_attr:8,gate_recurr:8,gather:[8,36,63],gauss:6,gcc:[33,41,49,53,54,60],gce:73,gcepersistentdisk:73,gcreators_:44,gen:[8,89],gen_conf:89,gen_data:89,gen_proto_pi:65,gen_rand_param:46,gen_result:89,gender:[12,75,86],gener:[2,7,8,9,12,14,15,16,17,18,19,21,23,24,26,30,32,37,38,39,40,41,44,67,72,75,80,89],generated_word_embed:8,generatedinput:[8,58,59],genr:[75,86],geq:8,gereat:7,get:[2,7,8,12,14,16,17,18,19,23,24,26,27,28,35,41,43,44,45,62,74,81,86,87,88],get_all_op_proto:44,get_batch_s:87,get_best_pass:88,get_config_arg:[72,88],get_data:[74,84,87],get_default_block:45,get_dict:12,get_dim:16,get_embed:12,get_float_el:16,get_grad:14,get_imdb:88,get_input_lay:62,get_model:82,get_movie_title_dict:12,get_numeric_gradi:16,get_numerical_gradi:16,get_output:16,get_sample_from_lin:46,get_shap:14,get_support:46,get_tensor:16,get_vari:17,get_word_dict:12,getbatchs:62,geteigendevic:64,getenv:[15,27,75],gethostbynam:75,gethostnam:75,getidmap:75,getinput:62,getinputgrad:62,getinputvalu:62,getlayeroutput:14,getmat:22,getoptconfig:22,getoutputgrad:62,getoutputvalu:62,getparam:22,getparameterconfig:22,getparameterptr:62,getparameterspars:22,getparametersremot:22,getplac:[63,64],getpodlist:75,getsiz:62,gettask:23,gettranspos:62,getw:62,getweight:62,getwgrad:62,gildea:87,gist:9,git:[42,46,49,54,60,61],github:[9,18,49,54,60,61,82],give:[2,19,35,41],given:[8,12,14,24,26,30,38,40],glide:18,global:[6,15,17,18,19,36,41,43,44,60,67],global_learning_r:6,globalstat:67,globalstatinfo:67,globe:2,glusterf:73,gnueabihf:54,go_librari:18,go_test:18,goal:[31,41],gob:23,godep:18,godoc:33,goe:[8,9,19,43],going:[30,79],golang:18,good:[8,38,79],goodfellow13:8,googl:[15,41,46,53],googleapi:73,got:43,gprotos_:44,gpu:[6,8,11,16,27,36,39,41,42,46,49,60,63,67,72,79,82,87,88,89],gpu_id:[46,70,72],gpu_per_train:39,gpugpu_id:69,gpukernel:41,gpuplac:63,grab:19,grad:[16,24,70],grad_op_class:41,grad_op_typ:41,grad_share_block_num:[69,70],grad_var_nam:16,gradient:[6,7,8,10,14,19,23,31,36,41,70,75,84],gradient_check:63,gradient_clipping_threshold:[6,46,84,88],gradient_evalu:7,gradient_flat:16,gradient_machin:[14,34],gradientcheck:63,gradientmachin:[4,14,34,36,75,86,89],gradientmachine_:22,grain:30,gram:80,grandient:14,graph:[8,14,17,18,19,35,37,39,45],grave:88,great:[40,79],greater:8,greaterthan:44,grep:[49,88],groudtruth:59,ground:[7,8,89],group:[9,23,79],group_input1:59,group_input2:59,group_input:[56,59],grouplen:[12,85],grpc:79,gru:[8,84],gru_bias_attr:9,gru_decod:59,gru_decoder_with_attent:59,gru_encoder_decod:[80,89],gru_layer_attr:9,gru_memori:9,gru_siz:84,gru_step:59,gru_step_lay:9,grumemori:[9,46,59],gserver:[8,62],gsizex:67,guard:22,guid:[28,74],gzip:[23,74],hadoop:15,hand:41,handi:18,handl:[12,15,27,36,38,39,43],handler:[14,17],handwrit:88,happen:[23,44],hard:[39,40],harvest:84,has:[7,8,9,10,12,15,16,17,18,19,23,24,26,32,39,40,41,44,67,79,87],has_kei:14,hassubseq:56,have:[8,9,12,15,16,17,18,19,23,24,26,27,30,31,32,35,36,38,39,40,43,79],hdf:21,head:[63,88],header:[24,34,46,47,82],heavili:41,height:[8,12,17,33,38,46,62,63],held:19,hello:15,help:[4,17,29,41,45,61],helper:[8,39],henc:43,here:[6,7,8,9,12,15,18,19,25,29,31,38,44,79],heterogen:[39,40],heurist:[8,40,89],hidden:[8,9,39,46],hidden_a:46,hidden_b:46,hidden_dim:56,hidden_out:17,hidden_s:9,hierach:58,hierarch:[8,41,56],hierarchi:41,high:[6,79],higher:30,highest:[12,17],highli:12,him:15,hint:[4,40],histor:30,hl_get_sync_flag:62,hold:[15,19,23,25,39,43,44,64],holder_:64,home:[21,29,39,68,74,75],honor:23,hook2:56,hook:[6,56,87],hookattr:6,hookattribut:6,horizont:8,hors:81,horst:88,host:[18,27,68,74],host_c:[53,54],hostnetwork:75,hostpath:[73,74,75],hot:84,hous:[2,12],how:[6,8,15,17,19,23,29,30,36,44],howardjohnson:56,howev:[8,9,16,26,36,38,39,44],howto:75,hpp:33,html:[12,81],htod:67,http:[8,9,12,18,27,49,53,54,60,61,73,74,79,81,82,85,89],huber:8,huge:8,huina:88,human:8,hyper:8,hyperplan:12,i0601:86,i0706:89,i0719:89,i1116:75,i1117:67,ib0:68,ics:12,icwsm:88,id_input:7,idea:[8,18,38],identityoffsetproject:8,identityproject:8,idmap:75,ids:[7,8,46,86],idx:[23,62],ieee:88,ies:29,ifelseop:[32,37],ifop:32,ight:8,ignor:[2,8],iil:46,ijcnlp:88,illustr:[7,19,24,30,35,39],ilsvrc:82,imag:[11,12,15,31,37,38,45,49,60,61,74,75,76,77,79,81,82,89],image_a:38,image_b:38,image_classif:81,image_fil:38,image_lay:38,image_list_provid:82,image_nam:15,image_path:38,image_provid:81,image_reader_cr:38,image_s:82,imagenet:[8,21],imagepullpolici:75,imageri:8,images_reader_cr:38,imagin:31,imdber:88,img:[2,8,9,39,81],img_conv_lay:9,img_pool_lay:9,img_siz:81,imgsiz:67,imgsizei:67,imgsizex:67,immutable_paramet:15,imperfect:41,implement:[8,9,10,12,17,23,24,25,26,27,30,33,34,36,39,40,41,43,44],implemet:22,impli:18,implicitli:37,imposs:79,improv:[8,40,41],inarg:[4,22],inc_path:46,includ:[7,8,9,12,15,17,18,24,27,33,34,35,39,41,44,53,54,61,63,67],incorpor:8,incorrect:8,increas:[19,23,46],increment:31,incupd:62,inde:[12,38],independ:[8,16,24,43,79],index:[7,8,11,12,14,16,17,19,23,37,56,86],indexslot:8,indic:[7,8,17,24,35,37],individu:19,industri:[19,79],ineffici:36,infer:[15,17,19,33,41],inferior:26,infershap:[17,37,41,45,63,64],infershapecontext:[63,64],infiniband:68,info:[7,8,12,35,51,56,62,68,75,79],inform:[8,12,14,17,27,29,35,39,43],inherit:[17,41],ininst:15,init:[6,14,17,35,39,45,51,62,75,87],init_from_tar:14,init_hook:[56,84,87],init_model_path:[69,70,72,80,84,87],initi:[2,6,8,9,12,14,18,23,31,35,39,40,44,45,70,84],initial_max:[6,46],initial_mean:[6,8,46],initial_min:[6,46],initial_std:[6,8,46],initpaddl:4,initrd:79,inlcud:9,inlin:64,inner:[46,56],inner_:56,inner_mem:56,inner_param_attr:9,inner_rnn_output:56,inner_rnn_st:56,inner_rnn_state_:56,inner_step:56,inner_step_impl:56,inproj_attr:8,inproj_bias_attr:8,inproj_param_attr:8,input0:64,input1:[8,9,64],input2:8,input:[2,5,7,8,9,11,12,14,16,17,22,26,30,31,32,35,36,38,39,40,41,43,44,45,46,47,51,55,56,58,59,62,63,64,72,75,80,81,84,86,87,89],input_conf:8,input_data:62,input_data_target:62,input_featur:5,input_fil:[47,87],input_hassub_sequence_data:62,input_id:8,input_imag:[9,81],input_index:62,input_label:62,input_lay:62,input_loc:8,input_nam:15,input_proj_bias_attr:9,input_proj_layer_attr:9,input_seq:8,input_sequence_data:62,input_sequence_label:62,input_sparse_float_value_data:62,input_sparse_non_value_data:62,input_t:62,input_to_check:16,input_typ:[46,47,56,84],input_valu:16,input_var:16,inputbuff:22,inputdef:62,inputlayers_:62,inputs_to_check:16,inputtyp:12,insert:[41,61],insid:[7,8,19,36,38,40,41],instal:[8,27,42,46,53,54,61,68,74,81,86],instanc:[8,16,17,19,21,25,32,35,39,41],instanti:19,instead:[8,9,11,18,22,27,31,38,39],instruct:[17,31],int16:45,int32:[37,45,70],int64:[28,45],integ:[2,7,8,12,23,27,33,37,51,84],integer_sequ:46,integer_valu:[2,12,46,51,56,84],integer_value_sequ:[2,12,56,59,84,87],integer_value_sub_sequ:56,integr:[7,79,87],inter:[8,39],intercept:8,interchang:31,interfac:[6,8,9,14,17,23,27,29,36,41,68,79],intergr:8,intermedi:[29,39],intern:[8,9,14],internet:[18,19,79],interpret:7,interv:8,introduc:[8,17,19,41,43,44],intuit:[26,41],invalid:[38,43],invoc:[17,18,30,41],invok:[2,8,14,36,39,40,41,44,67],iob:7,ioe:7,ip_str:75,ipc:73,ips:75,ipt:[8,44,46,56,59],ipx:79,ipython:15,is_gener:[80,89],is_kei:86,is_loc:14,is_predict:[84,88],is_seq:[8,59],is_stat:[6,46],is_tensor:44,is_test:[82,87,88],is_train:2,isinst:[4,51],ispodallrun:75,isspars:62,issu:18,issue_numb:61,item:[8,12,14,26,38,75,79],iter:[2,8,9,10,12,14,15,19,38,39],its:[2,8,9,14,15,16,17,19,23,31,35,36,41,43,44,67],itself:[19,26,43,45],ivs:44,java:[17,33,37,41],jeremi:67,jie:[87,88],jmlr:8,job:[12,26,39,41,68,69,70,72,73,75,82,84,87,88,89],job_dispatch_packag:68,job_id:12,job_mod:80,job_nam:[27,75],job_namespac:75,job_path:75,job_path_output:75,job_workspac:68,jobnam:75,jobpath:75,jobselector:75,jobserv:27,johan:88,join:[19,56],joint:89,jointli:[9,89],journal:[87,88],jpg:82,json:[68,74,86],jth:9,jupyt:[27,49],just:[5,7,8,9,12,18,23,24,35,36,39,41,43],jypyt:15,k8s:[75,79],k8s_data:75,k8s_job:15,k8s_token:15,k8s_train:75,k8s_user:15,kafka:21,kaim:8,kaimingh:82,kebilinearinterpbw:67,kebilinearinterpfw:67,keep:[8,12,19,31,43,44,79],keep_top_k:8,kei:[2,12,14,16,17,19,21,23,28,41,44,60,61,67,73,75,86],kept:8,kernel:[8,16,63,64],key1:70,key2:70,keyword:75,kill:19,kind:[15,16,19,25,31,39,73,74,75],kingsburi:87,know:[15,23],known:[17,30,35],kriz:[12,81],ksimonyan:9,kube:73,kube_cluster_tl:15,kube_ctrl_start_job:15,kube_list_containers_in_job_and_return_current_containers_rank:15,kubeadm:73,kubectl:[73,74,75],kuberent:19,kubernet:[15,19,41,66,68,75,76,77,79],kubernetes_service_host:15,kwarg:[2,9,10,12,44,56,84,87],l1_rate:6,l2_rate:6,l2regular:[46,81,84,88],l93:22,label:[2,7,8,10,12,14,31,37,38,39,40,46,47,51,56,74,81,82,84,87,88],label_dict:87,label_dim:[8,56,84],label_fil:[38,87],label_lay:38,label_list:87,label_path:38,label_slot:87,labeledbow:88,labelselector:75,lag:70,lake:2,lambdacost:8,lambdarank:8,languag:[8,12,31,41,43,80],larg:[11,12,39,40,61,89],larger:[6,7,8],last:[7,8,9,35,45,55,56],last_seq:56,last_time_step_output:8,lastseen:74,latenc:[8,39],later:[18,41,64],latest:[8,17,18,19,26,46,74,75],launcher:15,layer1:[8,9,46,55],layer2:[8,46,55],layer3:8,layer:[4,6,7,9,11,12,13,14,17,22,31,32,37,38,39,40,41,44,45,51,55,58,59,62,82,84,87],layer_0:62,layer_att:46,layer_attr:[8,46,59,72],layer_num:[72,82],layer_s:8,layer_typ:8,layerbas:62,layerconfig:62,layergradutil:62,layermap:62,layeroutout:8,layeroutput:9,layers_test:46,lbl:[7,81],ld_library_path:68,learn:[6,7,8,9,10,12,15,24,26,31,38,39,40,49,67,81,82,87,88,89],learnabl:14,learning_method:[47,80,81,84,88,89],learning_r:[6,24,39,46,47,80,81,84,88,89],learning_rate_arg:46,learning_rate_decay_a:46,learning_rate_decay_b:46,learning_rate_schedul:46,leas:19,least:[7,8,19],leav:17,lecun:12,left:[8,17],legal:44,leman:89,len:[2,8,24,28,56,62,75,84,87],length:[8,9,12,24,35,41,74],leq:8,less:[8,15,79],less_than:15,let02:74,let:[7,8,15,17,26,30,31,35],level:[6,8,30,35,58],lgtest:18,lgtest_main:18,lib64:[46,49,68,70],lib:[34,48,53,54],lib_path:46,libapi:18,libari:34,libcuda:[46,49],libjpeg:81,libnvidia:[46,49],libpaddl:[33,34,41,61],libpaddle_capi:34,libpaddle_gserv:34,libpaddle_math:34,libprotobuf:46,librari:[8,18,25,34,39,68,70],life:19,lifecycl:79,lifetim:43,lightweight:30,like:[7,8,12,17,18,19,22,27,30,31,37,38,39,41,43,79,82],limit:[8,12,41,46,67],linaro:54,line:[2,7,12,18,22,27,29,31,37,41,46,56,61,72,87],line_count:46,linear:[8,46,51],linearactiv:47,linguist:87,link:[8,9,18,28,29,41,43,58,79,88],linux:[28,49,53,54,60,73],linux_x86_64:[42,46],lipeng:80,lipton:88,list:[1,2,7,8,9,12,14,15,17,18,23,27,29,32,36,43,46,47,68,72,81,82,84,86,87,88,89],listen:19,littl:24,lium:89,liwicki:88,load:[8,15,19,39,47,75,82,87,88,89],load_data_arg:4,load_featur:82,load_feature_c:82,load_feature_pi:82,load_missing_parameter_strategi:[69,70,72,80,87],load_paramet:46,loadparamet:4,loadsave_parameters_in_pserv:[22,69,70],loc:7,local:[6,14,16,17,19,25,26,31,35,41,48,68,69,70,75],local_scop:16,localhost:[49,73],localip:75,localpath:29,locat:[8,14,18],lock:[18,19,23,24,40],lod:35,lod_level:45,lod_tensor:[35,45],lodtensor:[41,45],lodtesnordesc:45,log:[23,29,39,46,62,68,70,74,75,81,86,87,88,89],log_barrier_abstract:[69,70],log_barrier_lowest_nod:[69,70],log_barrier_show_log:[69,70],log_clip:[69,70],log_error_clip:[69,70],log_period:[70,72,74,75,81,84,86,87,88,89],log_period_serv:[69,70],logarithm:5,logger:[2,56],logic:[26,36,39,40,43],logist:84,longer:[19,39],look:[2,7,17,27,31,39],lookahead:8,loop:[16,17,38,43],loss:[8,84],lot:[39,79],low:8,low_rnn:35,lower:[8,35],lower_level_rnn:35,lpaddle_capi_shar:34,lpaddle_capi_whol:34,lstm:[8,56,59,74,84],lstm_bias_attr:9,lstm_cell_attr:9,lstm_group:[9,56],lstm_group_input:56,lstm_input:56,lstm_last:56,lstm_layer_attr:9,lstm_nest_group:56,lstm_output:56,lstm_size:84,lstm_step:9,lstmemori:[9,46,56,59],lstmemory_group:[8,46,56],lstmemory_unit:46,ltr:8,mac:[34,49,53],machin:[8,9,12,14,39,40,46,58,79,88,89],machine_transl:59,maco:60,macro:30,made:[19,24],mai:[8,9,16,17,38,39,40,41,43],main:[4,41],mainli:25,maintain:[8,17,23,41],majel:18,major:[39,89],make:[2,7,8,15,17,18,19,23,24,26,31,35,36,38,39,41,53,54,60,61,62,63,65,67,79,88],make_ddim:64,make_function_oper:30,make_vari:44,maker:41,man:28,manag:[14,19,24,25,29,39,43,45],mandarin:8,mani:[8,9,18,23,36,41,43,44],manner:8,manual:[39,46,79],manufactur:89,manylinux1_x86_64:[42,46],mao:88,map:[7,8,12,14,15,17,23,43,44,51,79],map_read:12,mapper:12,mapreduc:15,marcu:88,mark:[31,35,40,43,45,79,87],mark_slot:87,market:88,martha:87,mask:[6,8],master:[15,26,41,42,73,88],mastermind:18,mat:[33,34],mat_cache_row:22,mat_norm:22,mat_normal_shar:22,mat_param_attr:9,mat_sparse_row:22,mat_sparse_row_auto_grow:22,mat_sparse_row_id:22,mat_sparse_row_prefetch:22,mat_sparse_row_prefetch_full_s:22,mat_value_shar:22,match:[8,18,46],matchbox:79,math:[9,33,41,62,63,67],matirx:8,matmul:[17,35,63],matplotlib:81,matrix:[7,8,9,12,14,22,33,34,62,63],matrixptr:62,matrixtyp:34,mattyp:22,max:[6,8,12,16,46,55,67,72,86],max_diff:16,max_id:[8,14],max_job_id:12,max_length:[8,59],max_movie_id:12,max_relative_error:[16,63],max_sort_s:8,max_user_id:12,maxframe_evalu:7,maxid:7,maxid_evalu:7,maxim:8,maximum:[7,8,12,17,24],maxinum:11,maxpool:8,mayb:[8,17],md5:[12,20],mean:[6,7,8,9,10,11,12,14,18,35,38,39,43,46,70,79,82,86],mean_img_s:81,mean_meta:82,mean_meta_224:82,mean_valu:82,mean_var_nam:8,mechan:[8,9,25],meet:87,mem:[8,17,27,56],member:[12,15,31,36,43],memcpi:[36,67],memori:[9,17,22,23,27,41,59,64,67,74],memory_boot:9,memory_nam:[8,46],memory_test:60,memory_threshold_on_load_data:[69,70],mention:[18,23,40],mere:9,merg:[8,24,26,35,36,61],mergedict:[80,89],messag:[17,31,37,41,45,61,74],meta:[68,81,82,86],meta_config:[68,86],meta_gener:[68,86],meta_path:81,metadata:[28,74,75],metal:[73,79],metaphor:31,metaplotlib:15,method:[2,8,10,14,16,17,26,31,40,41,43,45,89],mfs:75,might:[8,17,18,37,40],million:12,min:[6,8,67,72,86],min_block:17,min_count:40,min_desc:17,min_pool_s:[2,46],min_word_freq:12,mini:[8,12,14,17,19,35],mini_batch:38,minibatch:[8,17,31],minikub:73,minim:[10,17,40,41],minimum:8,minsizerel:[53,54],minut:[19,26],mip:53,mirror:18,mislead:24,miss:87,mix:[9,59],mixed_lay:[9,56,87],mixed_layer_attr:9,mixedlayertyp:8,mkdir:[29,65],mkl:[41,48],mkl_root:48,ml_data:[68,86],mnist:[2,4,21,31,37,38,39,41],mnist_model:4,mnist_provid:2,mnist_random_image_batch_read:38,mnist_train:[2,38],mnist_train_batch_read:38,mnt:75,mobil:41,mod:87,mode:[8,14,61,75,88],model:[8,9,12,14,17,19,20,31,39,40,41,72,80,81,84,86,87,88,89],model_config:4,model_list:[70,72,87,88],model_output:88,model_path:72,model_zoo:[80,82],modifi:[8,39,61],modul:[2,9,12,14,30,47,63,81,82,84],modulo:8,momentum:[6,43,46,51,84],momentumoptim:[47,81],mon:74,mono:8,month:[18,89],mood:88,moosef:73,more:[7,8,9,12,15,16,18,19,23,26,27,29,30,31,35,38,39,40,41,46,67,79],morin:8,mose:[88,89],moses_bleu:89,mosesdecod:88,most:[8,12,15,18,26,31,37,38,39,45,79],mostli:79,motiv:41,mount:27,mountpath:[74,75],move:[8,19,23,29,79],movi:[2,12,86],movie_categori:12,movie_id:[75,86],movie_info:12,movie_review:12,movieinfo:12,movielen:85,moving_average_fract:8,mpi:68,mse:[31,37],much:[8,19,38],mul:[30,41,62,63],mul_grad:63,mul_op:63,mulgradkernel:63,mulkernel:63,mulop:[30,63],mulopgrad:63,mulopmak:63,multi:[8,36,79,82,89],multi_binary_label_cross_entropi:8,multi_crop:82,multigradientmachin:36,multinomi:8,multipl:[7,8,9,12,14,15,16,23,24,26,30,39,40,45],multipli:[7,8],multiprocess:12,must:[5,7,8,9,12,24,32,38,41,44,46,62,63,64],mutabl:64,mutable_data:[63,64],mxnet:17,my_cost:46,my_standalone_toolchain:53,mypaddl:[74,75],name:[2,6,7,8,9,11,14,15,16,17,19,21,22,24,27,30,34,41,42,44,45,47,49,51,56,59,62,67,72,73,74,75,76,77,79,81,84,86,89],name_prefix:21,namespac:[17,33,62,63,73,74,75],nativ:8,natur:[23,26,40],nchw:8,ndarrai:[14,21],ndcg:8,ndcg_num:8,ndk:53,nearli:16,necessari:[8,17,24,26,36,44],need:[7,8,9,12,15,16,18,22,23,24,26,27,29,30,36,37,39,40,41,43,44,46,67,75,79],need_tran:46,neg:[2,7,8,87,88],neg_distribut:8,neg_overlap:8,neg_pos_ratio:8,ner:7,nerual:39,nest:[8,12,17,37],net:[8,9,17,35,43,88],net_conf:88,net_diagram:82,netop:[17,41],network:[4,6,7,8,10,12,13,14,15,16,17,19,22,35,38,39,40,43,44,45,46,51,56,68,72,75,79,80,81,82,87,88,89],network_config:72,neural:[8,9,10,12,14,15,17,19,39,43,45,56,58,80,87,88,89],neuralnetwork:36,never:[12,38,43,74,75],new_stat:35,newest:24,newli:79,newop:17,newremot:39,newvar:[17,43],next:[2,8,12,19,25],nfs:75,nfsdir:75,nginx:49,ngram:12,nic:[68,69,70,75],nil:23,nine:12,nlp:8,nltk:12,nms_threshold:8,nms_top_k:8,nmt:89,nmt_without_attent:46,nnz:62,no_cach:2,no_grad_set:[16,63],no_sequ:[2,8,51,86],noah:88,noavx:49,node0:75,node:[8,18,26,39,40,41,45,73,74,75,79],node_0:75,node_1:75,node_2:75,nodebook:49,nodefil:68,nois:[8,19],non:[8,19],none:[2,4,6,7,8,9,10,11,14,15,16,35,44,45,47,59,82],norm:9,norm_by_tim:8,normal:[8,9,12,74,75,82],notat:8,note:[6,8,9,11,14,15,17,22,23,27,38,41,64,88],notebook:27,noth:[5,14,43,61],notingradi:63,notori:16,novel:88,now:[8,12,18,19,40,41,43,58],np_arrai:12,nproc:60,ntst1213:89,ntst14:89,nullptr:[43,62],num:[8,9,68,70,75,87,88,89],num_channel:[8,9,81],num_chunk_typ:7,num_class:[8,9,81],num_filt:[8,9],num_flatten_dim:45,num_gradient_serv:[69,70],num_group:8,num_neg_sampl:8,num_p:39,num_parameter_serv:15,num_pass:[14,47,51,69,70,72,74,75,84,86,87,88,89],num_repeat:8,num_result:7,num_results_per_sampl:8,num_samples_process:46,num_shard:21,num_train:39,number:[7,8,9,12,17,19,21,38,40,41,44,89],numchunktyp:7,numdevices_:72,numeric_grad:16,numerical_grad:16,numlogicaldevices_:72,numofallsampl:7,numofwrongpredict:7,numpi:[6,12,14,21,38,46,47,51,63,82],numreal:22,numsampl:67,numtagtyp:7,numtimeout:23,nv_:18,nv_gpu:60,nv_librari:18,nv_test:18,nvcc:[18,49],nvidia:[46,49,60],obei:7,obj:[2,46,47,81,82,84,86],object:[2,6,8,9,10,12,14,15,22,33,41,43,45,67],observ:[8,10],obtain:26,obvious:18,occup:[75,86],occur:[12,14],occurr:17,oct:74,odd:8,off:[34,48,49,53,54,60,61,65,79],offer:[17,41,44],offici:[8,18],offlin:[19,21,79],offset:[8,22,86],often:[22,37],ograd:62,old:[16,24,26,41],omit:[46,84],on_init:2,onc:[8,19,23,31,39,40],one:[2,5,7,8,9,10,11,12,14,15,16,17,19,22,23,24,26,27,30,32,35,36,38,39,41,43,79,84,87,88],one_host_dens:86,one_hot_dens:86,onehotcrossentropyopkernel:63,ones:[30,41],onli:[7,8,9,11,14,15,16,18,22,23,24,25,26,27,29,31,32,35,36,39,40,41,44,56,58,79],onlin:[8,10,19,21,38],only_cpu:16,onto:[39,40],op_:63,op_check:63,op_class:[41,44],op_maker_class:[41,44],op_proto:44,op_test_util:63,op_typ:41,opattrcheck:63,opcreat:44,opdesc:[17,31,37,41,44,45],opdescbuild:17,open:[2,8,15,21,38,46,47,56,82,87],openbla:48,openblas_root:48,oper:[8,9,10,12,16,17,31,32,35,37,39,43,45,63,64],operartor:64,operator_grad:16,operatorbas:[17,30,41,44,63],operatorwithkernel:63,opinfo:41,opinion:88,opkernel:[63,64],opkernelkei:41,opmak:44,opproto:63,opprotoandcheckermak:63,opprotomak:[44,63],opregist:44,opregistri:44,ops:[16,17,18,31,37,41,45,63,79],ops_:17,ops_test:18,opt:[15,44,48,75],optestmeta:63,optim:[6,13,14,16,36,37,39,40,41,46,51],optimzi:46,option:[7,8,15,18,37,44,45,79],optyp:44,order:[8,9,12,14,31,38,75,79],ordereddict:14,org:[7,8,9,12,21,28,49,85],organ:[7,8],orient:44,origin:[8,12,16,43,61],other:[7,8,9,12,17,19,24,29,35,43,44,79],otherchunktyp:7,otherwis:[8,12,15,19,24,26,32,38],our:[15,18,39,40,43,45],out:[8,14,15,17,18,23,26,32,35,45,46,56,58,59,63,64,81],out_dir:75,out_left:8,out_mem:59,out_memori:9,out_right:8,out_size_i:8,out_size_x:8,outer:56,outer_mem:56,outer_rnn_st:56,outer_rnn_state_:56,outer_step:56,outlier:8,outout_lay:14,outout_layer1:14,outout_layer2:14,output:[5,6,7,9,11,12,14,15,16,17,21,26,29,30,31,32,35,38,39,40,41,43,44,45,46,47,56,59,63,64,68,72,74,75,80,81,82,84,86,87,88],output_:8,output_all_step:35,output_dir:82,output_fil:87,output_id:8,output_lay:[14,46,82],output_max_index:11,output_mem:[8,59],output_nam:16,output_num:[17,32,35],output_path:21,output_s:45,outputbuff:22,outputh:8,outputw:8,outsid:[2,8,9,43],outv:62,over:[8,9,14,15],overal:79,overlap:[7,8],overlap_threshold:[7,8],overrid:[17,19,29,63,64],overview:[23,24,25],overwrit:29,own:[24,26,39,44],pack:46,packag:[12,23,27,30,42,46],pad:9,pad_c:8,pad_h:8,pad_w:8,paddelpaddl:37,padding_attr:8,padding_i:8,padding_x:8,paddl:[2,4,5,6,7,8,9,10,11,12,14,15,17,18,19,21,27,29,30,32,33,34,35,36,39,41,44,45,47,49,51,59,60,61,62,63,65,67,68,72,74,75,79,80,81,84,86,87,88,89],paddle_begin_init_param:24,paddle_dir:63,paddle_doc:65,paddle_docs_cn:65,paddle_element_typ:24,paddle_element_type_float32:24,paddle_element_type_float64:24,paddle_element_type_int32:24,paddle_element_type_int64:24,paddle_element_type_uint32:24,paddle_element_type_uint64:24,paddle_enforc:17,paddle_enforce_eq:[63,64],paddle_error:[33,34],paddle_exampl:27,paddle_finish_init_param:24,paddle_get_param:24,paddle_gradi:24,paddle_init_param:24,paddle_job:27,paddle_matrix:[33,34],paddle_matrix_cr:34,paddle_matrix_get_shap:33,paddle_matrix_shap:33,paddle_n:[68,75],paddle_new_etcd_pserver_cli:24,paddle_new_pserver_cli:24,paddle_on_cloud:27,paddle_output:74,paddle_paramet:24,paddle_port:[68,75],paddle_ports_num:[68,75],paddle_ports_num_for_spars:68,paddle_ports_num_spars:75,paddle_process_by_paddl:75,paddle_pserver2:68,paddle_pserver_cli:24,paddle_pserver_client_releas:24,paddle_root:80,paddle_save_model:24,paddle_send_grad:24,paddle_server_num:75,paddle_source_root:80,paddle_train:[34,42,68,75],paddledev:[46,49,74,75],paddlepaddl:[8,9,12,14,18,19,21,24,25,26,27,28,29,30,31,35,36,37,38,41,43,45,49,59,60,61,67,68,76,77,79,80,87],pair:[7,8,17,31,39,41],pakcag:18,palceholder_just_ignore_the_embed:80,palmer:87,paper:[8,89],para:[22,80],paraconvert:80,paragraph:35,paragraph_data:35,paragraph_out:35,parallel:[39,40,41,67,72,74,75,89],parallel_nn:[6,69,70],parallelli:41,param:[6,8,9,12,16,17,24,36,46,64],param_attr:[8,9,22,46,47,59],param_config_proto:24,param_fil:46,paramattr:[6,8,46,47,59],paramet:[7,9,10,11,12,13,16,17,18,20,22,26,29,31,35,37,38,39,43,44,46,51,70,75,86,87,88,89],parameter_attribut:8,parameter_block_s:[69,70],parameter_block_size_for_spars:[69,70],parameter_learning_r:6,parameter_nam:[14,15],parameter_serv:15,parameter_valu:22,parameterattribut:[6,8,9,22],parameterclient2:75,parameterclient_:22,parametermap:62,parametermutex_:22,parameters_:62,parameterserver2:22,parameterset:15,parameterupdat:36,parameterupdater_:22,parametris:10,paramutil:86,paraphras:[80,89],paraphrase_data:80,paraphrase_model:80,paraphrase_modeldata:80,paraspars:62,parent:[17,37,41],parent_:[17,43],pars:[12,18,60],parse_config:4,parse_known_arg:75,parsefromstr:46,parser:75,part:[7,8,17,26,41,79,88],partial:[8,14],partial_sum:8,particular:[31,41],partit:[19,21,39,40,41],paserv:75,pass:[2,8,14,17,19,36,38,41,43,45,46,47,51,61,67,70,72,74,75,81,84,86,87,88,89],pass_id:[14,51],pass_idx:38,pass_manu:46,passtyp:62,past:15,patch:28,path:[7,12,14,19,23,24,27,38,53,54,68,70,73,74,75,87,88],pattern:[12,19,33,88],paul:87,paus:[19,26],pave:89,pdf:[8,9],peer:46,pem:[15,21],pend:[19,23],penn:87,pep425tag:46,per:[7,8,12,19,24,38],perform:[8,16,24,36,41,67,69],period:[19,26,70,87,88,89],perl:[88,89],peroid:8,persistentvolum:73,persistentvolumeclaim:[73,75],person:[7,15],perspect:41,perturb:16,pex:79,pfs:[21,29],pfsclient:21,pfspath:29,phase:79,physic:79,pid:73,piec:[8,9],pillow:[27,81],pip:[42,46,61,68,81,86],pixel:[2,8,12],pixels_float:2,pixels_str:2,place:[2,19,26,39,40,41,63,64],placehold:64,placement:40,plain:[7,8,27,34],plan:[19,41],platform:[17,53,63,64],pleas:[6,8,9,10,15,19,23,24,25,35,38,39,41,46,75],plot:[15,81],plotcurv:81,plu:[8,16],png:81,pnpairvalid:69,pod:[21,27,73,74,75],podip:75,podlist:75,point:[17,19,27,64,67,79],pointer:[17,24,41,43,64],polar:12,poli:46,poll:88,pollut:26,pool3:62,pool:[9,13],pool_attr:9,pool_bias_attr:9,pool_layer_attr:9,pool_pad:9,pool_siz:[2,8,9],pool_size_i:8,pool_strid:9,pool_typ:[8,9],pooling_lay:[9,46,84],pooling_typ:[8,46,55,84],poolingtyp:11,pop:17,popul:24,popular:18,port:[18,68,69,70,74,75],port_num:69,ports_num:[70,75],ports_num_for_spars:[22,69,70,72,75],pos:[86,88],pose:19,posit:[2,7,8,9],positive_label:7,posix:73,possibl:[15,17,23,40],possiblli:41,post:[27,28],potenti:67,pow:46,power:79,practic:8,pre:[8,9,12,15,17,24,80,88,89],pre_dictandmodel:80,pre_stat:35,precis:7,precision_evalu:7,pred:87,predetermin:8,predic:[12,87],predicate_dict:87,predicate_dict_fil:87,predicate_slot:87,predict:[2,4,7,8,10,14,39,46,68,80,81,82,84,86,87,88],predict_fil:[69,70],predict_lay:14,predict_output_dir:[69,70,84],predict_sampl:4,prediction1:14,prediction2:14,predin:81,prefer:73,prefetch:[22,62],prefix:[7,19,21],pregrad:62,premodel:80,prepar:[16,27,36,76],preprocess:[12,68,80,81,84,86,88,89],present:[15,17,61],preserv:29,prev_batch_st:[69,70],prevent:[10,15,19,23,26],preview:41,previou:[8,9,14,19,29,35,40],previous:8,price:[12,41],primari:31,principl:[15,18],print:[4,6,14,15,17,39,45,46,47,51],print_arrai:17,printallstatu:67,printer:7,printstatu:67,priorbox:8,prioriti:41,prite:7,privat:[17,34,43,44,61,64],prob:[7,14],probabilist:[8,80],probabl:[7,8,14],problem:[8,10,15,16,18,26,31,41],proc:49,proce:[12,19,38],procedur:17,proceed:[8,87],process2:56,process:[2,6,8,9,12,15,17,21,22,23,26,39,44,46,47,56,75,84,86,87],process_num:12,processdata:[81,82],processor:67,prod:[45,49,61],produc:[8,9,12,19,32,38,45],product:[8,27],productgraph:74,profil:[29,67],proflier:67,prog:75,program:[12,15,21,24,26,31,37,38,39,41,43,67,75],programm:39,progress:[19,23],proj:8,project:[8,9,27,34],promis:[8,9],prompt:29,prone:15,prop:87,propag:10,properti:37,propos:[17,40],proposit:87,protect:[44,62,63],proto:[11,41,44,45,63],proto_:44,protobuf:[14,17,27,31,37,41,44,46],protocol:[7,79],protomak:63,provid:[8,12,15,17,24,27,44,46,47,56,69,79,84,87],provis:79,prune:[8,17],ps_desir:19,pserver:[14,22,24,25,27,41,68,69,70,75],pserver_addr:24,pserver_cpu:27,pserver_id:20,pserver_mem:27,pserver_num_thread:[22,69,70],pserver_spec:14,pseudo:[15,27],psize:62,ptr:34,pull:[18,42,49],purpos:[8,19,39,40,67],push:[17,75],push_back:62,put:[18,19,22,40,45],pwd:[49,60,61],pxe:79,py_paddl:[4,46],pybind:17,pydataprovid:46,pydataprovider2:[2,4,47,75],pyramid:8,pyramid_height:8,python:[14,15,17,25,30,31,33,36,37,41,42,46,49,59,60,61,62,63,68,80,81,82,86,87,88,89],pythonpath:[46,81],pzo:88,queri:[8,89],question:[8,15,40,44],queue:40,quick:74,quick_start:[27,74,75,76,84],quick_start_data:74,quickli:41,quickstart:74,r14b:53,r_t:8,rac:8,rais:12,ramnath:88,ran:[40,67],rand:[46,67,70,72,87],random:[6,8,12,21,36,38,46,47,63],random_imag:21,random_uniform:45,randomli:26,rang:[8,12,21,38,39,44,75,84],rank:[8,15,82,84],rare:2,raspberri:54,raspberrypi:54,raspbian:54,rate:[6,7,9,10,12,24,68,75,86],rather:27,ratio:[8,70],raw:8,raw_meta:86,rdma_tcp:[69,70],reach:19,read:[2,12,14,15,19,21,38,39,40,41,46,47,79,82],read_from_realistic_imag:15,read_from_rng:15,read_lock:20,read_mnist_imag:15,read_next_from_fil:46,read_paramet:46,read_ranking_model_data:15,readabl:41,reader:[14,21,31,37,39,51],reader_cr:21,reader_creator_bool:38,reader_creator_random_imag:[12,38],reader_creator_random_image_and_label:[12,38],readi:[19,74,79],readlockguard:22,readm:[34,61,86,88],readwritebuffer_:22,real:[8,22,38],real_process:2,realist:15,realiz:[17,35],reaon:45,reason:[9,15,16,19,74],recal:7,receiv:[19,27,40],recognit:[8,82,88],recommend:[9,15,68,75,86],record:[12,23,44,45,87],recordio:[12,15,21,23,39],recov:19,recover:41,recoveri:23,rectangular:8,rectifi:8,recurr:[43,56,57,87,88],recurrent_group:[9,46,56,58,59],recurrent_lay:9,recurrentgradientmachin:34,recurrentgroup:7,recurs:[17,18,29,41],recv:[39,40],recvparametertyp:22,reduc:[10,40,41],reduce_by_kei:41,ref:86,refactor:[31,36,39,40],refactorizaiton:41,refer:[6,8,9,10,16,17,19,23,24,25,35,43,48],referenc:[8,23],reference_cblas_root:48,refine_unknown_arg:75,reflect:23,reg:44,regard:79,region:[8,43],register_gpu_profil:67,register_lay:62,register_op:[30,41,44,63],register_op_cpu_kernel:63,register_op_gpu_kernel:63,register_op_without_gradi:[41,63],register_tim:22,register_timer_info:67,registerop:44,registr:41,registri:[27,74,79],regress:8,regular:[6,46,81,84,88],rel:[9,16,26,63],relat:[19,26,27,37,43,45,79],releas:[42,53,54,73,87],reli:[16,41],reliabl:19,relu:8,rememb:8,remot:[6,18,22,41,61,68,70,72],remoteparameterupdat:[22,25],remov:[12,29,61],renam:29,reorgan:8,repeat:[17,31,37,44,45],repeatedli:31,replac:[18,23],replic:39,replicaset:27,repo:18,report:[23,39],reportdataset:23,repositori:[8,53],repres:[8,9,17,23,37,40,41,45],represent:[8,24,31,39],request:[18,19,22,26,39,42,74,79,89],requir:[7,8,15,19,24,26,27,29,35,37,39,40,41,44,45,68,79,86],res5_3_branch2c_bn:82,res5_3_branch2c_conv:82,res:87,research:[12,39,81],reserv:29,reserveoutput:62,reset:[8,19,46],reshap:[16,38,46],reshape_s:8,residu:82,resiz:[22,63,64],resnet_101:82,resnet_152:82,resnet_50:82,resolv:[18,61,74],respect:[16,35],respons:[8,22,36,74],rest:[8,17,27,28,79],restart:[19,24,74,79],restartpolici:[74,75],restor:16,restrict:43,resu:38,result:[2,5,7,8,14,16,23,31,36,39,67,84,89],result_fil:7,resum:26,retriev:[17,43],return_seq:9,reuqest:42,reus:[17,26,38,41],rev:60,reveal:15,revers:[8,9,58,59],review:[12,61,74,88],reviews_electronics_5:74,rewrit:[18,89],rgb:8,rgen:88,rho:10,right:[8,16,17,18,27,41],rkt:[27,60],rmsprop:[46,84],rnn:[8,9,17,41,43,58,59,69,88],rnn_bias_attr:59,rnn_layer_attr:59,rnn_out:59,rnn_state:56,rnn_state_:56,rnn_step:8,rnn_use_batch:[69,70],rnnlm:12,robot:81,robust:8,role:[12,15,23,24,39,87,88],roman:88,root:[10,11,49,68,74,75,80],root_dir:68,rot:8,routledg:88,row:[7,8,12,22],row_id:8,rpc:23,rpcserver:23,rpi:54,rpi_arm_neon:54,rpi_toolchain:54,rstrip:75,rtk:79,rtype:[8,12],rule:[7,31,39],run:[15,16,17,18,19,27,30,31,35,37,39,40,41,43,45,46,49,60,61,64,67,68,74,75,76,77,79,86],runinitfunct:[67,75],runnabl:40,running_on_cloud:27,runtim:[2,17,35,41,45,46],runtime_table_:17,safe:27,sai:[8,31],same:[2,7,8,9,14,15,16,23,24,26,30,35,39,41,43,56,64,84],samping_id:8,sampl:[2,7,12,44,84,86,88,89],sample_id:7,sample_num:7,santiago:88,satifi:7,satisfi:18,save:[2,8,12,14,19,21,23,24,27,31,39,45,74,86,87,88,89],save_dir:[47,70,72,74,75,81,84,86,87,88,89],save_only_on:[69,70],savetxt:46,saving_period:[69,70,75],saving_period_by_batch:[69,70,72],saw:2,sbin:49,scalabl:41,scalar:[8,17],scale:[5,39,40,44,63],scaleop:63,scaleopmak:[41,63],scalingproject:8,scan:[23,41],scatter:8,schedul:[23,27,40],scheduler_factor:6,scheme:[7,22],schmidhub:88,schwenk:89,scienc:88,scope:16,score:[7,8,86],script:[12,65],seaplane_s_000978:81,search:[8,19,43,59,89],seat:89,second:[8,15,29,35,38,43,44,63,84],section:40,sed:88,see:[8,9,15,19,39,46],seed:[46,67,70],seem:18,segment:[7,35],sel_fc:8,selcet:8,select:8,selected_indic:8,selectiv:8,selector:74,self:[16,45,62,63],selfnorm:8,semant:[12,15,42,87,88],semat:15,sen_len:87,send:[19,24,39,40,41,44,45],send_back_parameter_typ:22,sendbackparameterspars:22,sendbackparametertyp:22,sendparameterrequest:22,sendparameterrespons:22,sensit:8,sent:[15,24,39,41,44,74],sentanc:46,sentenc:[2,8,12,35,56,59,87],sentence_last_state1:56,sentence_last_state2:56,sentiment:[2,87,88],sentiment_data:88,sentiment_net:88,sentimental_provid:2,sentimental_train:2,separ:[7,24,30,86],seq:[8,12,56,86],seq_pool:[8,55],seq_silc:8,seq_text_print:7,seq_to_seq_data:[80,89],seq_typ:12,seqlastin:56,seqtext_evalu:7,seqtoseq:[8,46,80,89],seqtoseq_net:[8,80,89],sequel:2,sequenc:[2,5,7,8,9,11,12,17,31,46,51,56,58,84,86,88,89],sequence_conv_pool:84,sequence_group:8,sequence_layer_group:56,sequence_nest_group:8,sequence_nest_layer_group:56,sequencegen:56,sequencestartposit:8,sequencetextprint:7,sequencetyp:[2,8,51],sequenti:[8,17,87],seri:[9,56,88],serial:[14,17,23,36,41,45],serv:[39,41],server:[15,18,22,25,26,39,41,46,68,70,73,75,79],serverless:19,servic:79,set:[2,6,7,8,9,12,14,15,19,27,32,35,41,43,46,47,56,60,64,67,68,74,80,81,82,84,87,88,89],set_active_typ:62,set_default_parameter_nam:6,set_drop_r:62,set_float_el:16,set_input:[8,17],set_output:[17,32],set_siz:62,set_typ:62,setdefault:63,setq:60,settotalbyteslimit:46,setup:[42,62,63,79],sever:[7,8,16,22,35,36],sexstant:79,sgd:[10,14,15,19,27,36,39,40,47,51,68,69,88],shall:18,shape:[8,12,14,16,17,35,41,51],shard:[19,20,21,22,23,24,26,39,40],share:[8,18,34,36,41,87],shared_bia:9,shared_bias:8,shared_librari:18,shared_ptr:[33,34,43,64],ship:81,shold:88,shorten:8,should:[6,7,8,12,14,15,16,17,24,27,30,32,35,36,37,38,39,41,44,58,63],should_be_fals:15,should_be_tru:15,should_shuffl:[2,56,87],show:[7,10,17,19,29,35,60,87,88,89],show_check_sparse_distribution_log:[69,70],show_layer_stat:[69,70],show_parameter_stats_period:[69,70,72,74,87,88,89],shown:[8,15,39],shuf:[46,86],shuffl:[12,46],side:[8,14,36],sigint:68,sigmod:44,sigmod_op:44,sigmod_output:44,sigmoid:[8,17,44],sigmoidactiv:[9,56],sign:28,similar:[8,38,40,41,79],similarli:[8,12],simpl:[5,7,8,9,12,14,31,40,43,44,75],simple_attent:59,simple_gru:[59,84],simple_lstm:[8,46,84],simple_rnn:[8,59],simpler:36,simpli:[8,15,24,35],simplifi:[15,44],sinc:[8,19,23,25,26,38,39,45,79],singl:[7,9,12,19,39,40,41,43],sinlg:14,site:18,size:[2,7,8,9,10,12,14,19,21,22,24,38,39,44,46,47,51,56,59,62,63,64,81,82,84,88,89],size_a:8,size_b:8,size_t:[22,62],sizeof:17,skip:[38,46,47,61,82],sleep:75,sliceproject:8,slide:[8,10,12,19],slope:8,slopeinterceptlay:8,slot:[86,87],small:[12,16],small_messag:[69,70],small_vgg:81,smaller:[8,16,19],smart:43,smartli:8,smith:88,smooth:8,snap:74,snapshot:[20,26],snippet:30,sock:27,sock_recv_buf_s:[69,70],sock_send_buf_s:[69,70],socket:75,softmax:[8,9,15,17,32,39,40,46,59,62,84],softmax_param:46,softmax_param_attr:9,softmax_selfnorm_alpha:8,softmaxactiv:[46,56,84],softwar:79,solid:41,solut:[37,79],solv:[15,39,41],some:[6,8,12,14,15,17,18,22,23,24,26,27,30,31,35,37,39,40,41,43,45,79],some_c_api_funct:34,some_inst:34,some_op:35,some_other_valu:17,some_python_class:33,some_valu:17,somecppclass:33,somedata:14,somegotyp:33,someth:[8,22],sometim:[8,10,38],somewhat:24,somewher:43,soon:19,sort:[8,12,75,88],sourc:[8,12,16,18,29,34,38,89],source_dict_dim:59,source_language_word:59,space:[7,8,40],space_seperated_tokens_from_dictionary_according_to_seq:7,space_seperated_tokens_from_dictionary_according_to_sub_seq:7,spars:[6,8,10,12,22,46,62,68,70,72,75],sparse_binary_vector:[2,12,46,51],sparse_binary_vector_sequ:12,sparse_float_vector:[2,51],sparse_non_value_slot:12,sparse_remot:22,sparse_upd:[6,22,46],sparse_value_slot:12,sparse_vector:[12,46],sparse_vector_sequ:12,sparseparam:62,sparseprefetchrowcpumatrix:62,spatial:8,spec:[74,75],special:[8,24,32,39],specif:[14,18,19,29,41,43],specifi:[7,8,15,16,22,23,24,27,29,39,43,44],speech:8,speed:[9,79],sphinx:[33,49,65],split:[2,8,26,41,56,68,86,87],split_count:75,sqrt:8,squar:[8,10,11],square_error_cost:[47,51],squarerootnpool:8,squash:61,srand:70,src:[18,75,89],src_backward:59,src_dict:46,src_dict_path:46,src_embed:59,src_forward:59,src_root:4,src_word_id:59,srl:[12,87],ssd:8,ssh:[49,68],sshd:49,sstabl:15,stabil:16,stabl:[42,73],stack:41,stacked_lstm_net:88,stacked_num:88,stackexchang:8,stage:[18,25,45,61],stake:89,stale:19,standalon:53,standard:[6,41],stanford:[12,16,74],star:18,start:[8,9,14,18,19,22,23,24,26,27,36,39,46,61,70,74,75],start_paddl:75,start_pass:[69,70],start_po:8,start_pserv:[69,70],startpaddl:75,startup:[19,27],stat:[67,70,87,88,89],state:[8,9,19,35,43,46,58,74],state_act:[8,9,56],statement:31,statfulset:75,static_cast:64,staticinput:[8,58,59],statist:8,statset:67,statu:[27,61,67,74,75],status:74,std:[14,18,22,33,34,41,43,44,62,63,64,70],stderr:68,stdout:68,step:[8,9,11,16,17,19,24,31,36,39,40,41,44,56,58,59,79],step_net:17,stepnet:[17,35,41,43],stepout:56,still:[23,26,39],stochast:[10,19,23,26],stock:88,stop:8,storag:[28,73],store:[7,8,12,14,16,17,18,22,36,39,41,43],str:[14,27,72,75],strategi:[11,19,40,70,87],street:[8,87],strict:38,stride:[8,9],stride_i:8,stride_x:8,string:[2,7,8,14,17,23,29,41,43,44,45,62,63,70],strip:[46,56,87],struct:[23,24,28,34,44,46],structur:[17,23,37,41],stuff:[17,61],stun:2,style:[8,41,44],sub:[7,8,12,15,26,35,36,39],sub_nest_seq:8,sub_sequ:[2,8,51],subcommand:29,subgradi:10,subgraph:40,submiss:39,subnet:15,subobjectpath:74,subseq:[55,58],subsequ:8,subsequenceinput:[8,56],succeed:[23,74],success:[8,24,74],successfulcr:74,sudo:[60,81],suffer:16,suffic:38,suffix:27,suggest:[8,18],suit:79,sum:[8,10,17,20],sum_:8,sumpool:[8,46],support:[6,7,8,11,12,16,17,19,26,27,30,36,38,39,40,41,56,79],suppos:[18,30],suppress:[8,29],sure:[41,88],surpass:8,svs:44,swagger:28,swap_channel:82,swig:[25,33,34],swig_paddl:4,switchop:17,symbol:[8,17,34],symbols_ready_:17,symbolt:[17,41],symlink:61,sync:19,syncflag:62,synchron:[10,19,23],syntax:38,sys:82,sysroot:53,system:[17,18,19,24,26,28,30,39,40,46,88],t2b:80,tab:84,tabl:[7,8,17,45],tableproject:8,tag:[7,12,46,49],tagtyp:7,tainer_id:75,take:[2,7,8,9,15,17,18,19,26,31,32,37,41,45],talk:24,tanh:[8,9,39,59,62],tanhactiv:[9,56],tar:14,target:[8,12,14,17,18,39,41,45,89],target_dict_dim:59,target_dictionary_dim:8,target_language_embed:8,target_language_word:59,target_link_librari:18,targetinlink:[8,56],task:[7,8,39,44,87],task_queu:23,taskentri:23,taskqueu:23,tbd:[25,56],tconf:88,tcp:70,tear:67,technic:19,tee:[74,81,87,88,89],tell:[19,23,24,44],tellig:88,templat:[30,44,63,64,74,75,79],tempor:8,temporari:27,tensor:[16,17,18,35,40,45,63],tensor_desc:45,tensor_s:16,tensor_test:18,tensor_to_check:16,tensorflow:[17,39,40],term:[8,9,19],termin:74,tesh:87,test100:12,test10:12,test1:21,test:[1,2,8,12,14,15,16,18,34,38,42,61,62,63,64,67,68,70,72,81,82,84,86,87,88,89],test_:63,test_all_data_in_one_period:[74,81,86,87,88],test_compar:46,test_comparespars:46,test_comparetwonet:46,test_comparetwoopt:46,test_config_pars:46,test_cpu_gpu_compar:63,test_data:[4,89],test_fcgrad:62,test_gpuprofil:67,test_ignore_i:63,test_ignore_x:63,test_layergrad:62,test_list:[2,46,47,81],test_mul_op:63,test_networkcompar:46,test_norm:63,test_part_000:88,test_pass:[69,70,72,89],test_period:[69,70,72],test_predict:46,test_pydataprovid:46,test_pydataprovider2:46,test_pydataproviderwrapp:46,test_ratio:86,test_recurrent_machine_gener:46,test_recurrentgradientmachin:[46,56],test_swig_api:46,test_train:46,test_traineronepass:46,test_wait:[69,70],testa:15,testb:15,testbilinearfwdbwd:67,testcas:63,testconfig:62,tester:[86,89],testfcgrad:62,testfclay:62,testlayergrad:62,testmulgradop:63,testmulop:63,testq:15,testresult:14,testutil:62,text1:29,text:[2,7,9,12,15,35,80,88,89],text_conv:84,text_conv_pool:86,text_fil:[12,88],tflop:67,tftp:79,tgz:12,than:[6,7,8,9,19,27,30,41,46,79],thei:[8,15,18,19,24,26,29,31,40,41,44,67],them:[7,8,9,15,16,18,19,22,27,30,38,40,41,43,44,45,67],themselv:18,therein:[8,17],thi:[2,6,7,8,9,10,12,14,15,16,17,18,19,22,23,24,25,26,27,30,31,35,36,37,38,39,40,41,44,45,67,79,88],thing:2,think:[15,18],third:[8,19],third_parti:[8,53,54],those:[17,18,19,30,31,87],though:79,thought:18,thread:67,thread_local_rand_use_global_se:[69,70],threadid:72,threadloc:67,three:[7,8,16,19,31,36,38,82],threshold:[6,7,8,19,23,70],through:[8,18,19,23,25,32],throughput:67,thrust:41,thu:[8,26],tier:74,time:[8,9,11,12,15,16,18,19,23,26,30,35,38,39,40,41,44,45,56,67,70,74,75,79,88],timelin:[8,41],timeout:[19,23],timeout_sec:12,timer:67,timestamp:[8,20],timestep:[8,43],titl:[12,75,86],tls:28,tmp:2,to_no_sequ:[8,55],to_proto:45,to_sequ:[8,55,56],to_your_paddle_clone_path:65,todo:[7,10,12,17,19,23,26,44],toend:8,togeth:[8,9,12,14],token:[7,8,15,59,88],toler:[14,16],too:[12,16,39],took:79,tool:[53,54,65,75],toolchain:[53,54],top:[7,14,35,82],top_k:7,top_level_rnn:35,toplevel:60,topolog:[15,19,36,39],topolopi:14,torch:17,toronto:[12,81],tostr:46,total:[14,19,38,40,67,74,79,89],total_pass:38,touch:88,tourist:89,trace:[17,37,45],track:[19,23,61],tractabl:8,tradit:[8,17],traffic:39,trail:12,train100:12,train10:12,train:[1,2,6,7,8,10,12,17,21,23,24,26,31,36,37,41,45,46,47,51,59,68,70,72,74,75,76,77,80,81,82,84,86,87,88,89],train_arg:75,train_args_dict:75,train_args_list:75,train_conf:[80,89],train_config_dir:75,train_data:89,train_i:51,train_list:[2,46,47,81,82],train_part_000:88,train_read:51,train_x:51,trainabl:8,trainer:[2,4,15,20,21,22,23,25,36,39,40,41,47,51,62,70,72,75,87,88,89],trainer_config:[1,2,4,47,68,74,75,84,86,88],trainer_config_help:[2,46,47,62,81],trainer_count:[46,69,70,72,74,75,86,87,88,89],trainer_cpu:27,trainer_cr:27,trainer_gpu:27,trainer_id:[70,75],trainer_intern:22,trainer_mem:27,trainer_packag:27,trainerconfighelp:46,trainerid:[26,75],trainerintern:[84,86,89],trainonebatch:22,tran:62,tranform:41,transact:[19,23,88],transform:[8,41],transform_param_attr:9,translat:[8,9,46,80,89],transpos:8,transposedfullmatrixproject:8,travel:2,travers:31,travi:61,treat:[8,24],treatment:24,tree:[8,17,75],trg:89,trg_embed:59,tricki:33,trigger:[26,36],trim:8,truck:81,true_block:[17,32],true_imag:38,true_label:38,true_read:38,truth:[7,8,89],tune:[6,8,69],tupl:[8,9,12,14,38],ture:8,turn:[8,38,58],tutori:[74,75,76,77,80,89],tweet:88,twice:40,twine:42,twitter:88,two:[7,8,9,15,17,24,25,26,27,29,31,38,39,43,44,45,63,64,67],txt:[2,18,27,29,62,68,73,84,86,88],type:[2,7,8,9,11,12,15,17,19,22,23,26,27,28,29,33,34,35,37,38,41,44,45,47,51,56,59,62,63,64,72,74,82,84,86,87],type_nam:44,typedef:[24,33,34],typeid:44,typenam:[30,44,63,64],typic:[7,39],ubuntu:42,ubyt:38,uci:12,ufldl:8,uid:74,uint32:28,uint64:33,uint64_t:33,unawar:24,unclear:26,unconstrain:88,under:[18,23,40],understand:79,understand_senti:59,undeterminist:67,unidirect:8,uniform:[6,8,12,21,38],uninstal:46,uniqu:[15,17,19,26,27,43],unique_ptr:[43,62],unit:[8,9,18],unittest:[34,46,63],univ:89,unix:68,unk:[45,80,89],unk_idx:87,unknown:8,unlik:8,unordered_map:43,unrol:35,unseg:8,unsign:24,unsup:88,unsupbow:88,unsupport:63,until:[19,24,40,43,75],untrack:61,unzip:53,updat:[6,8,10,17,19,23,24,28,35,36,39,43,61,68,72],update_equ:[14,51],update_hook:6,updatecallback:62,upgrad:46,upload:[19,27,28,42],upon:19,upper:8,upstream:[46,61],url:[12,88],urls_neg:88,urls_po:88,urls_unsup:88,usag:[7,8,9,14,36,75],use:[6,7,8,9,11,12,14,15,16,17,18,19,25,35,36,39,40,43,44,45,61,63,67,75,82,87,88,89],use_etcd:14,use_global_stat:8,use_gpu:[4,46,51,69,70,72,74,75,81,82,84,86,87,88,89],use_jpeg:81,use_old_updat:[22,69,70],use_seq:[47,86],use_sparse_remote_updat:22,used:[2,7,8,9,10,11,12,14,15,16,17,18,19,25,26,35,36,38,39,41,43,44,45,67,88],useful:[8,9,16,43],usegpu:62,user:[6,8,9,12,14,15,16,17,18,21,23,26,27,29,30,31,38,39,40,41,43,44,73,79,86],user_id:[75,86],user_info:12,user_nam:21,usercert:21,userinfo:12,userkei:21,usernam:[21,61],uses:[19,26,36,39],using:[6,8,9,12,14,15,17,18,19,23,24,26,27,29,30,35,38,39,43,44,63,87],usr:[46,48,49,68,70,75],usrdict:80,usrmodel:80,usual:[8,14,27,67],utc:85,util:[67,75,79,81,86],uuid:[20,26],v28:8,v7a:53,valid:[8,38,41,43],valu:[2,4,6,7,8,10,11,12,14,16,17,19,32,35,36,37,39,41,43,44,45,46,62,72,75,82,84,87],value1:70,value2:70,value_evalu:7,value_rang:12,valueerror:46,vanilla:59,var1:45,var2:45,varabl:40,varaibl:17,vardesc:[17,31,37,41],vardescbuild:17,variabl:[8,12,15,16,17,31,32,35,39,40,44,74],variablenamemap:63,varianc:8,variant:8,variou:17,varproto:44,vars_:[17,43],vec:46,vector:[8,9,12,15,17,22,24,32,35,41,84,86],vendor:18,verb:12,verbos:29,veri:[8,11,18,23,30,36,40,43,81],verifi:17,version:[8,9,18,27,29,42,49,67,69,70],versu:15,vertic:8,vgg:[9,81],vgg_16_cifar:81,via:[19,35,38,61,79],view:8,virtualenv:60,visibl:[26,43],vision:81,visipedia:81,visit:14,visual:8,vlog:22,vocab:88,volum:[49,73,74,75],volumemount:[74,75],vutbr:12,wai:[7,8,9,15,24,26,39,89],wait:[10,19,24,75],wall:87,wangkuiyi:18,want:[2,8,15,27,38,43],warn:[29,46,75],warp:8,warpctc:8,watch:19,wbia:82,wei:[87,88],weight:[7,8,9,10,62,82],weight_act:9,weightlist:62,weights_:62,weights_t:62,welcom:18,well:[27,30,39,41],were:[7,18],wether:8,wget:53,what:[6,8,10,17,18,40,44,61,79],when:[2,6,7,8,12,14,16,17,18,19,22,23,24,27,29,36,37,39,40,41,45,67,79],where:[8,9,10,15,17,19,26,31,32,35,37,41],wherea:[17,23,30],whether:[7,8,14,16,17,38,88],which:[6,7,8,9,10,12,14,15,16,17,18,19,21,23,24,26,27,30,35,36,37,38,39,40,41,43,44,45,79,86],whileop:[17,37],who:[30,31],whoever:24,whole:[2,7,12,33,34,35,44,79],whose:[8,12,16,19,26,32,35,41,44],why:[9,16,34,67],wide:18,width:[7,8,12,22,33,38,46,62,63,89],wiki:[8,18],wikipedia:[8,12],wilder:2,window:[8,11,12,60,73],wise:[8,40,41],with_avx:[48,49,53,61],with_bia:44,with_c_api:[53,54],with_doc:48,with_doubl:[48,62],with_dso:48,with_gpu:[48,49,53,60,61],with_profil:67,with_python:[48,53],with_rdma:[48,53],with_style_check:[48,61],with_swig_pi:[48,53],with_test:[48,49,61,63],with_tim:[48,67],within:[8,23],without:[7,8,19,24,38,40,41],wmt14:[59,89],wmt14_data:89,wmt14_model:89,wmt_shrinked_data:12,woboq:49,won:56,wonder:2,word2vec:[27,46],word:[2,7,8,12,31,35,40,41,44,46,56,58,84,87],word_dict:[56,84,87],word_dim:[46,56,84],word_id:[2,46],word_idx:12,word_slot:87,word_vector_dim:[8,59,80],words_freq_sort:12,work:[12,15,17,18,19,38,49,56,61,74,75,79],worker:[40,45],workflow:41,workspac:68,would:[14,17,18,19,26,30,31,36,38,40,79,87],wouldn:31,wrap:[30,79],wrapper:[9,18,30,39,67],write:[12,15,19,26,30,38,39,40,41,46,87],write_lock:20,writelin:47,writer:15,written:[17,39,41],wrong:38,wrote:40,wsj:87,wuyi:73,www:[8,12,81,89],x64:54,x86:[53,54],x86_64:53,x_1:8,x_n:8,x_neg:16,x_po:16,xarg:[7,46,49,62],xgbe0:70,xgbe1:70,xiaojun:88,xmap_read:12,xrang:[16,38,45,47,51,62],xxbow:88,xxx:[15,82,89],xxxx:20,y_neg:16,y_po:16,y_predict:51,yaml:[18,74,75,79],yancey1989:27,yann:12,yapf:61,yeild:14,yet:79,yield:[2,12,15,21,38,46,47,51,56,87],yin:8,you:[2,6,8,9,14,16,27,43,79,82,88],your:[8,14,15,18,22,27,29,41,46,53,54,79],your_param_nam:46,your_repo:75,your_source_root:34,yourself:8,yuyang18:[10,12],zachari:88,zeng:88,zero:[6,8,9,10,12,16,19,36,37,70],zhou:[87,88],zip:[12,53,75,85],zoo:80},titles:["API","DataProvider\u7684\u4ecb\u7ecd","PyDataProvider2\u7684\u4f7f\u7528","API\u4e2d\u6587\u624b\u518c","\u57fa\u4e8ePython\u7684\u9884\u6d4b","Activation","Parameter Attribute","Evaluators","Layers","Networks","Optimizer","Pooling","Data Reader Interface and DataSets","Model Configuration","Training and Inference","PaddlePaddle Design Doc","Auto Gradient Checker Design","Design Doc: Block and Scope","Required CMake Function","Design Doc: Distributed Training","\u6a21\u578b\u53c2\u6570\u68c0\u67e5\u70b9\uff08Checkpointing\uff09","\u8bad\u7ec3\u6570\u636e\u7684\u5b58\u50a8\u548c\u5206\u53d1","Alalysis of large model distributed training in Paddle","Design Doc: Master Server","Design Doc: The Client Library of Parameter Server","Design Doc: Remote Parameter Updater for Cluster Train","Design Doc: Save Model","Submit a Distributed Training Job","FileManager\u8bbe\u8ba1\u6587\u6863","PFSClient","Design Doc: Functions, Operators, and Layers","Design Doc: Computations as a Graph","&lt;no title&gt;","Paddle\u591a\u8bed\u8a00\u63a5\u53e3\u5b9e\u73b0","C-API \u6a21\u578b\u63a8\u65ad\u5b9e\u73b0\u6587\u6863","RNNOp design","Design Doc: The C++ Class <code class=\"docutils literal\"><span class=\"pre\">Parameters</span></code>","Design Doc: ProgramDesc","Python Data Reader Design Doc","Design Doc: Distributed Training Architecture","Design Doc: Operation Graph Based Parameter Server","Design Doc: Refactorization Overview","PaddlePaddle\u53d1\u884c\u89c4\u8303","Design of Scope in Paddle","Interaction between C++ and Python","Background","FAQ","\u7ecf\u5178\u7684\u7ebf\u6027\u56de\u5f52\u4efb\u52a1","PaddlePaddle\u7684\u7f16\u8bd1\u9009\u9879","PaddlePaddle\u7684Docker\u5bb9\u5668\u4f7f\u7528\u65b9\u5f0f","\u5b89\u88c5\u4e0e\u7f16\u8bd1","\u57fa\u672c\u4f7f\u7528\u6982\u5ff5","\u65b0\u624b\u5165\u95e8","\u6784\u5efaAndroid\u5e73\u53f0\u4e0a\u7684PaddlePaddle\u5e93","\u6784\u5efaRaspberry Pi\u5e73\u53f0\u4e0a\u7684PaddlePaddle\u5e93","\u652f\u6301\u53cc\u5c42\u5e8f\u5217\u4f5c\u4e3a\u8f93\u5165\u7684Layer","\u5355\u53cc\u5c42RNN API\u5bf9\u6bd4\u4ecb\u7ecd","RNN\u76f8\u5173\u6a21\u578b","Recurrent Group\u6559\u7a0b","RNN\u914d\u7f6e","\u7f16\u8bd1PaddlePaddle\u548c\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5","\u5982\u4f55\u8d21\u732e\u4ee3\u7801","\u5b9e\u73b0\u65b0\u7684\u7f51\u7edc\u5c42","\u5982\u4f55\u5199\u65b0\u7684Operator","\u5728Paddle\u4e2d\u5982\u4f55\u4f7f\u7528Eigen","\u5982\u4f55\u8d21\u732e/\u4fee\u6539\u6587\u6863","\u8fdb\u9636\u6307\u5357","GPU\u6027\u80fd\u5206\u6790\u4e0e\u8c03\u4f18","\u8fd0\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3","\u53c2\u6570\u6982\u8ff0","\u7ec6\u8282\u63cf\u8ff0","\u8bbe\u7f6e\u547d\u4ee4\u884c\u53c2\u6570","\u4f7f\u7528\u6848\u4f8b","Kubernetes \u7b80\u4ecb","Kubernetes\u5355\u673a\u8bad\u7ec3","Kubernetes\u5206\u5e03\u5f0f\u8bad\u7ec3","&lt;no title&gt;","&lt;no title&gt;","PaddlePaddle \u6587\u6863","Cluster bootstrapping tool survey","\u4e2d\u6587\u8bcd\u5411\u91cf\u6a21\u578b\u7684\u4f7f\u7528","\u56fe\u50cf\u5206\u7c7b\u6559\u7a0b","Model Zoo - ImageNet","\u5b8c\u6574\u6559\u7a0b","\u5feb\u901f\u5165\u95e8\u6559\u7a0b","MovieLens\u6570\u636e\u96c6","MovieLens\u6570\u636e\u96c6\u8bc4\u5206\u56de\u5f52\u6a21\u578b","\u8bed\u4e49\u89d2\u8272\u6807\u6ce8\u6559\u7a0b","\u60c5\u611f\u5206\u6790\u6559\u7a0b","\u6587\u672c\u751f\u6210\u6559\u7a0b"],titleterms:{"\u4e00\u4e9b\u7ec6\u8282\u7684\u8865\u5145":75,"\u4e0a\u4f20\u8bad\u7ec3\u6587\u4ef6":21,"\u4e0b\u8f7d\u4e0e\u89e3\u538b\u7f29":89,"\u4e0b\u8f7d\u548c\u6570\u636e\u62bd\u53d6":80,"\u4e0b\u8f7d\u5e76\u89e3\u538b\u6570\u636e\u96c6":86,"\u4e0b\u8f7d\u6570\u636e":74,"\u4e0d\u4f7f\u7528":33,"\u4e0d\u4f7f\u7528swig\u8fd9\u79cd\u4ee3\u7801\u751f\u6210\u5668":33,"\u4e0d\u5bfc\u51fapaddle\u5185\u90e8\u7684\u7ed3\u6784\u4f53":33,"\u4e0d\u5f15\u7528\u5176\u4ed6\u52a8\u6001\u5e93":33,"\u4e24\u79cd\u4f7f\u7528":46,"\u4e2d\u4e0d\u540c\u7684":46,"\u4e2d\u6587\u5b57\u5178":80,"\u4e2d\u6587\u77ed\u8bed\u6539\u5199\u7684\u4f8b\u5b50":80,"\u4e2d\u6587\u8bcd\u5411\u91cf\u6a21\u578b\u7684\u4f7f\u7528":80,"\u4e2d\u6587\u8bcd\u5411\u91cf\u7684\u9884\u8bad\u7ec3\u6a21\u578b":80,"\u4e3a\u4ec0\u4e48\u8981":60,"\u4e3a\u4ec0\u4e48\u9700\u8981\u6027\u80fd\u5206\u6790":67,"\u4ec0\u4e48\u662f\u6027\u80fd\u5206\u6790":67,"\u4ec5\u4ec5\u4f7f\u7528void":33,"\u4ecb\u7ecd":[80,82],"\u4ece\u5feb\u7167\u6062\u590d":20,"\u4ee3\u7801\u8981\u6c42":61,"\u4efb\u52a1\u7b80\u4ecb":47,"\u4f18\u5316\u7b97\u6cd5":84,"\u4f7f\u7528":[61,74],"\u4f7f\u7528\u52a8\u6001\u5e93\u6765\u5206\u53d1paddl":33,"\u4f7f\u7528\u6848\u4f8b":72,"\u4f7f\u7528\u6982\u8ff0":84,"\u4f7f\u7528\u6a21\u578b\u521d\u59cb\u5316\u7f51\u7edc":72,"\u4f7f\u7528\u73af\u5883\u53d8\u91cf":75,"\u4f7f\u7528\u7528\u6237\u6307\u5b9a\u7684\u8bcd\u5411\u91cf\u5b57\u5178":80,"\u4f7f\u7528\u8bf4\u660e":66,"\u4f7f\u7528\u8f6c\u6362\u5e93":21,"\u4f7f\u7528docker\u6784\u5efa":65,"\u4f7f\u7528paddlepaddle\u751f\u6210\u6a21\u578b":89,"\u4f7f\u7528paddlepaddle\u8bad\u7ec3\u6a21\u578b":89,"\u4fdd\u6301\u672c\u5730\u4ed3\u5e93\u6700\u65b0":61,"\u4fee\u6539\u542f\u52a8\u811a\u672c":74,"\u4fee\u6539\u6587\u6863":65,"\u514b\u9686":61,"\u5177\u4f53\u67d0\u79cd\u7c7b\u578b\u7684\u5934\u6587\u4ef6":34,"\u5177\u4f53\u67d0\u79cd\u7c7b\u578b\u7684\u5b9e\u73b0\u6587\u4ef6":34,"\u5185\u5b58\u4e0d\u591f\u7528\u7684\u60c5\u51b5":2,"\u5185\u7f6e\u5b9a\u65f6\u5668":67,"\u5199\u68af\u5ea6\u68c0\u67e5\u5355\u5143\u6d4b\u8bd5":62,"\u51c6\u5907\u4ea4\u53c9\u7f16\u8bd1\u73af\u5883":[53,54],"\u51c6\u5907\u5de5\u4f5c\u7a7a\u95f4":68,"\u51c6\u5907\u6570\u636e":[47,86],"\u51c6\u5907\u8bad\u7ec3\u6570\u636e":75,"\u51c6\u5907\u96c6\u7fa4\u4f5c\u4e1a\u914d\u7f6e":68,"\u51cf\u5c11\u6570\u636e\u8f7d\u5165\u7684\u8017\u65f6":46,"\u51cf\u5c11dataprovider\u7f13\u51b2\u6c60\u5185\u5b58":46,"\u51fa\u73b0":46,"\u5206\u5272\u8bad\u7ec3":86,"\u5206\u5757\u6587\u4ef6\u4f20\u8f93":28,"\u5206\u652f\u89c4\u8303":42,"\u521b\u5efa\u672c\u5730\u5206\u652f":61,"\u521b\u5efajob":75,"\u521b\u5efapaddl":74,"\u5220\u9664\u672c\u5730\u5206\u652f":61,"\u5220\u9664\u8fdc\u7a0b\u5206\u652f":61,"\u5229\u7528\u66f4\u591a\u7684\u8ba1\u7b97\u8d44\u6e90":46,"\u5230\u8fdc\u7a0b\u4ed3\u5e93":61,"\u5236\u4f5c\u955c\u50cf":75,"\u5236\u4f5cdocker\u955c\u50cf":74,"\u524d\u5411operator\u5355\u5143\u6d4b\u8bd5":63,"\u524d\u63d0\u6761\u4ef6":68,"\u52a0\u8f7dpaddlepaddl":51,"\u52a0\u901f\u6267\u884c":20,"\u52a0\u901f\u8bad\u7ec3\u901f\u5ea6":46,"\u52a8\u6001\u5e93\u4e2d\u4e0d\u5d4c\u5165\u4efb\u4f55\u5176\u4ed6\u8bed\u8a00\u7684\u89e3\u91ca\u5668":33,"\u52a8\u6001\u6269\u5bb9":20,"\u5355\u5143\u6d4b\u8bd5":70,"\u5355\u53cc\u5c42rnn":56,"\u5377\u79ef\u6a21\u578b":84,"\u5377\u79ef\u795e\u7ecf\u7f51\u7edc":81,"\u539f\u56e0":33,"\u539f\u56e0\u5217\u8868":33,"\u53c2\u6570\u4fe1\u606f":82,"\u53c2\u6570\u5185\u5b58":46,"\u53c2\u6570\u670d\u52a1\u5668\u548c\u5206\u5e03\u5f0f\u901a\u4fe1":70,"\u53c2\u6570\u6982\u8ff0":69,"\u53c2\u6570\u8bfb\u53d6":82,"\u53c2\u8003":2,"\u53c2\u8003\u6587\u6863":[28,88],"\u53c2\u8003\u8d44\u6599":67,"\u53cc\u5411lstm":88,"\u53cc\u5c42rnn":56,"\u53cc\u5c42rnn\u4ecb\u7ecd":58,"\u53cc\u5c42rnn\u7684\u4f7f\u7528":58,"\u53cd\u5411operator\u5355\u5143\u6d4b\u8bd5":63,"\u53ef\u80fd\u7684\u5185\u5b58\u6cc4\u9732\u95ee\u9898":2,"\u53ef\u80fd\u78b0\u5230\u7684\u95ee\u9898":60,"\u53ef\u9009\u529f\u80fd":80,"\u540d\u8bcd\u89e3\u91ca":28,"\u5411\u7cfb\u7edf\u4f20\u9001\u6570\u636e":84,"\u5411\u91cf":70,"\u542f\u52a8\u4efb\u52a1":75,"\u542f\u52a8\u96c6\u7fa4\u4f5c\u4e1a":68,"\u5440":60,"\u547d\u4ee4\u884c\u53c2\u6570":84,"\u548c":55,"\u56fe\u50cf\u5206\u7c7b\u6559\u7a0b":81,"\u5728\u4e0d\u540c\u8bbe\u5907\u4e0a\u6307\u5b9a\u5c42":72,"\u5728paddle\u4e2d\u5982\u4f55\u4f7f\u7528eigen":64,"\u5728paddlepaddle\u5e73\u53f0\u8bad\u7ec3\u6a21\u578b":80,"\u57fa\u4e8epython\u7684\u9884\u6d4b":4,"\u57fa\u672c\u4f7f\u7528\u6982\u5ff5":51,"\u57fa\u672c\u539f\u7406":58,"\u57fa\u672c\u8981\u6c42":33,"\u5982\u4f55\u4e66\u5199\u6587\u6863":65,"\u5982\u4f55\u4f7f\u7528":46,"\u5982\u4f55\u5171\u4eab\u53c2\u6570":46,"\u5982\u4f55\u5199\u65b0\u7684oper":63,"\u5982\u4f55\u51cf\u5c11\u5185\u5b58\u5360\u7528":46,"\u5982\u4f55\u521d\u59cb\u5316\u53c2\u6570":46,"\u5982\u4f55\u52a0\u8f7d\u9884\u8bad\u7ec3\u53c2\u6570":46,"\u5982\u4f55\u52a0\u901fpaddlepaddle\u7684\u8bad\u7ec3\u901f\u5ea6":46,"\u5982\u4f55\u548c\u660e\u6587\u8fdb\u884c\u76f8\u4e92\u8f6c\u5316":46,"\u5982\u4f55\u6307\u5b9agpu\u8bbe\u5907":46,"\u5982\u4f55\u66f4\u65b0\u6587\u6863\u4e3b\u9898":65,"\u5982\u4f55\u66f4\u65b0doc":65,"\u5982\u4f55\u6784\u5efa\u6587\u6863":65,"\u5982\u4f55\u8bbe\u7f6e\u5b66\u4e60\u7387\u9000\u706b":46,"\u5982\u4f55\u8c03\u7528":46,"\u5982\u4f55\u8d21\u732e":65,"\u5982\u4f55\u8d21\u732e\u4ee3\u7801":61,"\u5982\u4f55\u8fdb\u884c\u6027\u80fd\u5206\u6790":67,"\u5982\u4f55\u9009\u62e9sgd\u7b97\u6cd5\u7684\u5b66\u4e60\u7387":46,"\u5b50\u5e8f\u5217\u95f4\u65e0memori":56,"\u5b50\u5e8f\u5217\u95f4\u6709memori":56,"\u5b57\u6bb5\u914d\u7f6e\u6587\u4ef6":86,"\u5b89\u88c5":84,"\u5b89\u88c5\u4e0e\u7f16\u8bd1":50,"\u5b89\u88c5\u6d41\u7a0b":50,"\u5b89\u88c5kubectl":73,"\u5b8c\u6574\u6559\u7a0b":83,"\u5b9a\u4e49operator\u7c7b":63,"\u5b9a\u4e49opkernel\u7c7b":63,"\u5b9a\u4e49protomaker\u7c7b":63,"\u5b9e\u73b0":33,"\u5b9e\u73b0\u5355\u5143\u6d4b\u8bd5":63,"\u5b9e\u73b0\u65b0\u7684\u7f51\u7edc\u5c42":62,"\u5b9e\u73b0\u65b9\u5f0f":34,"\u5b9e\u73b0\u8ba1\u7b97":64,"\u5b9e\u73b0c":[62,63],"\u5b9e\u73b0python\u5c01\u88c5":62,"\u5bfc\u51fac":33,"\u5c06\u547d\u4ee4\u53c2\u6570\u4f20\u7ed9\u7f51\u7edc\u914d\u7f6e":72,"\u5c0f\u7ed3":2,"\u5de5\u5177":67,"\u5e38\u7528\u6a21\u578b":83,"\u5e76\u5b8c\u6210":61,"\u5efa\u7acb":61,"\u5f00\u53d1\u6807\u51c6":66,"\u5f00\u59cb\u5f00\u53d1":61,"\u5f02\u6b65\u968f\u673a\u68af\u5ea6\u4e0b\u964d":70,"\u5f15\u7528":87,"\u5feb\u7167\u4fdd\u5b58\u7684\u8bbe\u8ba1\u5982\u4e0b":20,"\u5feb\u901f\u5165\u95e8\u6559\u7a0b":84,"\u6027\u80fd\u4f18\u5316":66,"\u6027\u80fd\u5206\u6790\u5c0f\u6280\u5de7":67,"\u6027\u80fd\u5206\u6790\u5de5\u5177\u4ecb\u7ecd":67,"\u6027\u80fd\u8c03\u4f18":70,"\u603b\u4f53\u6548\u679c\u603b\u7ed3":84,"\u603b\u4f53\u6d41\u7a0b":60,"\u60c5\u611f\u5206\u6790\u6559\u7a0b":88,"\u6216\u8005\u662f":46,"\u627e\u5230\u7684pythonlibs\u548cpythoninterp\u7248\u672c\u4e0d\u4e00\u81f4":46,"\u62a5importerror":46,"\u6307\u9488\u4f5c\u4e3a\u7c7b\u578b\u7684\u53e5\u67c4":33,"\u63a5\u53e3":82,"\u63a5\u53e3\u8f93\u51fa\u591a\u4e2alayer\u7684\u9884\u6d4b\u7ed3\u679c":46,"\u63a8\u5bfc\u65b9\u7a0b":62,"\u63a8\u6d4b\u6267\u884c":20,"\u63d0\u4ea4":61,"\u63d0\u4ea4\u4ee3\u7801\u7684\u4e00\u4e9b\u7ea6\u5b9a":61,"\u63d0\u4ea4\u955c\u50cf":74,"\u63d0\u53d6\u7535\u5f71\u6216\u7528\u6237\u7684\u7279\u5f81\u5e76\u751f\u6210python\u5bf9\u8c61":86,"\u642d\u5efa\u795e\u7ecf\u7f51\u7edc":51,"\u652f\u6301\u53cc\u5c42\u5e8f\u5217\u4f5c\u4e3a\u8f93\u5165\u7684layer":55,"\u652f\u6301\u7528\u6237\u81ea\u5b9a\u4e49\u7684\u6570\u636e\u9884\u5904\u7406job":21,"\u6570\u636e\u51c6\u5907":[81,86,89],"\u6570\u636e\u63cf\u8ff0":87,"\u6570\u636e\u63d0\u4f9b":87,"\u6570\u636e\u63d0\u4f9b\u811a\u672c":86,"\u6570\u636e\u652f\u6301":70,"\u6570\u636e\u683c\u5f0f\u51c6\u5907":84,"\u6570\u636e\u7684\u51c6\u5907\u548c\u9884\u5904\u7406":80,"\u6570\u636e\u96c6\u7279\u5f81":85,"\u6570\u636e\u9884\u5904\u7406":89,"\u6570\u6910\u4ecb\u7ecd":88,"\u6570\u6910\u51c6\u5907":88,"\u6574\u4f53\u65b9\u6848":75,"\u6587\u4ef6\u4f20\u8f93\u4f18\u5316":28,"\u6587\u4ef6\u8bbf\u95ee\u65b9\u5f0f":21,"\u6587\u4ef6\u8bbf\u95ee\u7684\u6743\u9650":21,"\u6587\u4ef6\u9884\u5904\u7406":21,"\u6587\u672c\u751f\u6210":89,"\u6587\u672c\u751f\u6210\u6559\u7a0b":89,"\u6587\u6863":[49,78],"\u65b0\u624b\u5165\u95e8":52,"\u65e5\u5fd7\u4e2d\u4fdd\u5b58\u5747\u4e3a\u7f51\u7edc\u901a\u4fe1\u7c7b\u9519\u8bef":46,"\u65f6\u5e8f\u6a21\u578b":84,"\u65f6\u5e8f\u6a21\u578b\u7684\u4f7f\u7528\u573a\u666f":2,"\u65f6\u95f4\u5e8f\u5217":56,"\u65f6\u95f4\u6b65":56,"\u66b4\u9732\u63a5\u53e3\u539f\u5219":34,"\u672c\u5730\u6d4b\u8bd5":72,"\u672c\u5730\u8bad\u7ec3":72,"\u672f\u8bed":20,"\u6784\u5efa\u548c\u6d4b\u8bd5":61,"\u6784\u5efaandroid\u5e73\u53f0\u4e0a\u7684paddlepaddle\u5e93":53,"\u6784\u5efaraspberri":54,"\u67b6\u6784\u56fe":28,"\u67e5\u770b\u8bad\u7ec3\u7ed3\u679c":74,"\u67e5\u770b\u8f93\u51fa":75,"\u6837\u4f8b\u6570\u636e":2,"\u6846\u67b6\u751f\u6210":28,"\u6848\u4f8b\u4e00":72,"\u6848\u4f8b\u4e8c":72,"\u68c0\u67e5\u6a21\u578b\u8f93\u51fa":68,"\u68c0\u67e5\u96c6\u7fa4\u8bad\u7ec3\u7ed3\u679c":68,"\u6982\u5ff5\u7b80\u4ecb":63,"\u6982\u5ff5\u89e3\u91ca":21,"\u6982\u8ff0":[55,58],"\u6a21\u5757":28,"\u6a21\u578b":82,"\u6a21\u578b\u4e0b\u8f7d":82,"\u6a21\u578b\u53c2\u6570\u68c0\u67e5\u70b9":20,"\u6a21\u578b\u63a8\u65ad\u5b9e\u73b0\u6587\u6863":34,"\u6a21\u578b\u68c0\u9a8c":47,"\u6a21\u578b\u7f51\u7edc\u7ed3\u6784":84,"\u6a21\u578b\u8bad\u7ec3":[81,89],"\u6a21\u578b\u8bc4\u4f30\u548c\u9884\u6d4b":86,"\u6a21\u578b\u914d\u7f6e":[56,66],"\u6a21\u578b\u914d\u7f6e\u7684\u6a21\u578b\u914d\u7f6e":56,"\u6ce8\u518coper":63,"\u6ce8\u610f\u4e8b\u9879":[2,63],"\u6d41\u7a0b\u4ecb\u7ecd":21,"\u6d4b\u8bd5":[70,87],"\u6d4b\u8bd5\u6587\u4ef6":86,"\u6d4b\u8bd5\u6a21\u578b":88,"\u7279\u5f81":87,"\u7279\u5f81\u63d0\u53d6":82,"\u751f\u6210\u5e8f\u5217":59,"\u751f\u6210\u6a21\u578b\u7684\u547d\u4ee4\u4e0e\u7ed3\u679c":89,"\u751f\u6210\u6d41\u7a0b\u7684\u4f7f\u7528\u65b9\u6cd5":58,"\u751f\u6210sparse\u6587\u4ef6":28,"\u7528\u6237\u4f7f\u7528\u6d41\u7a0b":28,"\u7528\u6237\u6587\u4ef6\u63cf\u8ff0":85,"\u7528\u6237\u81ea\u5b9a\u4e49\u6570\u636e\u96c6":89,"\u7528\u6237\u81ea\u5b9a\u4e49\u6570\u6910\u9884\u5904\u7406":88,"\u7535\u5f71\u6587\u4ef6\u63cf\u8ff0":85,"\u7684\u533a\u522b":46,"\u7684\u53c2\u6570":46,"\u7684\u65b9\u6cd5\u6709\u4f55\u533a\u522b":46,"\u76ee\u5f55\u7ed3\u6784":34,"\u76ee\u6807":28,"\u76f4\u63a5\u6784\u5efa":65,"\u76f8\u5173\u6982\u5ff5":58,"\u77e9\u9635":70,"\u793a\u4f8b1":56,"\u793a\u4f8b2":56,"\u793a\u4f8b3":56,"\u793a\u4f8b4":56,"\u793a\u4f8b\u7a0b\u5e8f":21,"\u795e\u7ecf\u5143\u6fc0\u6d3b\u5185\u5b58":46,"\u795e\u7ecf\u7f51\u7edc\u7ed3\u6784\u914d\u7f6e":86,"\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e":87,"\u7a00\u758f\u8bad\u7ec3":72,"\u7b26\u53f7":33,"\u7b80\u4ecb":[73,89],"\u7b80\u5355\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":59,"\u7c7b":[33,62,63],"\u7ebf\u6027\u56de\u5f52\u5b8c\u6574\u793a\u4f8b":51,"\u7ec3\u4e60":81,"\u7ec6\u8282\u63a2\u7a76":81,"\u7ec6\u8282\u63cf\u8ff0":70,"\u7ec8\u6b62\u96c6\u7fa4\u4f5c\u4e1a":68,"\u7ecf\u5178\u7684\u7ebf\u6027\u56de\u5f52\u4efb\u52a1":47,"\u7ed1\u5b9apython":63,"\u7f16\u5199yaml\u6587\u4ef6":74,"\u7f16\u8bd1":63,"\u7f16\u8bd1\u548c\u5b89\u88c5":[53,54],"\u7f16\u8bd1\u548c\u6267\u884c\u5355\u5143\u6d4b\u8bd5":63,"\u7f16\u8bd1\u5b89\u88c5\u540e\u6267\u884c":46,"\u7f16\u8bd1\u6d41\u7a0b":50,"\u7f16\u8bd1\u9009\u9879":34,"\u7f16\u8bd1\u9009\u9879\u7684\u8bbe\u7f6e":48,"\u7f16\u8bd1paddlepaddle\u548c\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5":60,"\u7f29\u5bb9":20,"\u7f51\u7edc\u53ef\u89c6\u5316":82,"\u7f51\u7edc\u914d\u7f6e\u4e2d\u7684\u8c03\u7528":2,"\u800c\u662f\u624b\u5199\u591a\u8bed\u8a00\u7ed1\u5b9a":33,"\u80cc\u666f":33,"\u81ea\u7136\u8bed\u8a00\u5904\u7406":70,"\u8986\u76d6\u4e0d\u4e00\u81f4\u7684\u90e8\u5206":28,"\u89c2\u6d4b\u8bcd\u5411\u91cf":80,"\u8bad\u7ec3":[70,86,87],"\u8bad\u7ec3\u5668\u914d\u7f6e\u6587\u4ef6":86,"\u8bad\u7ec3\u56e0\u6b64\u9000\u51fa\u600e\u4e48\u529e":46,"\u8bad\u7ec3\u6570\u636e\u5b58\u50a8":21,"\u8bad\u7ec3\u6570\u636e\u7684\u5b58\u50a8\u548c\u5206\u53d1":21,"\u8bad\u7ec3\u6a21\u578b":[47,51,84,88],"\u8bad\u7ec3\u6a21\u578b\u7684\u547d\u4ee4\u4e0e\u7ed3\u679c":89,"\u8bad\u7ec3\u6d41\u7a0b\u7684\u4f7f\u7528\u65b9\u6cd5":58,"\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u51fa\u73b0":46,"\u8bbe\u7f6e\u547d\u4ee4\u884c\u53c2\u6570":71,"\u8bc4\u5206\u6587\u4ef6\u63cf\u8ff0":85,"\u8bcd\u5411\u91cf\u6a21\u578b":84,"\u8bcd\u5411\u91cf\u6a21\u578b\u7684\u4fee\u6b63":80,"\u8bcd\u6c47\u8868":56,"\u8be6\u7ec6\u6559\u7a0b":67,"\u8bed\u4e49\u89d2\u8272\u6807\u6ce8\u6559\u7a0b":87,"\u8bfb\u53d6\u53cc\u5c42\u5e8f\u5217\u6570\u636e":56,"\u8f6c\u6362\u5e93":21,"\u8f93\u5165":58,"\u8f93\u5165\u4e0d\u7b49\u957f":56,"\u8f93\u5165\u793a\u4f8b":58,"\u8f93\u51fa":58,"\u8f93\u51fa\u65e5\u5fd7":84,"\u8fd0\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3":68,"\u8fd0\u884c\u5bb9\u5668":74,"\u8fd0\u884cdocker":46,"\u8fd0\u884cpaddlepaddl":49,"\u8fd9\u4e2a\u52a8\u6001\u5e93\u4f7f\u7528c99\u6807\u51c6\u7684\u5934\u6587\u4ef6\u5bfc\u51fa\u4e00\u4e9b\u51fd\u6570":33,"\u8fdb\u884c\u8bad\u7ec3":[21,74],"\u8fdb\u9636\u6307\u5357":66,"\u9009\u62e9\u5b58\u50a8\u65b9\u6848":73,"\u901a\u7528":70,"\u901a\u8fc7docker\u5bb9\u5668\u5f00\u53d1paddlepaddl":49,"\u903b\u8f91\u56de\u5f52\u6a21\u578b":84,"\u9047\u5230":46,"\u90e8\u7f72kubernetes\u96c6\u7fa4":73,"\u914d\u7f6e\u4e2d\u7684\u6570\u636e\u52a0\u8f7d\u5b9a\u4e49":84,"\u914d\u7f6e\u4ea4\u53c9\u7f16\u8bd1\u53c2\u6570":[53,54],"\u914d\u7f6e\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u67b6\u6784":59,"\u914d\u7f6e\u7f51\u7edc":51,"\u914d\u7f6ekubectl":73,"\u914d\u7f6ekubectl\u8bbf\u95ee\u4f60\u7684kubernetes\u96c6\u7fa4":73,"\u94a9\u5b50":61,"\u9519\u8bef\u600e\u4e48\u529e":46,"\u9644\u5f55":84,"\u968f\u673a\u6570":70,"\u96c6\u7fa4\u591a\u8282\u70b9\u8bad\u7ec3":46,"\u96c6\u7fa4\u8bad\u7ec3":72,"\u9700\u8981\u7684\u8f6f\u786c\u4ef6":60,"\u975e\u6cd5\u6307\u4ee4":46,"\u9884\u5904\u7406":81,"\u9884\u5904\u7406\u547d\u4ee4\u548c\u7ed3\u679c":89,"\u9884\u5904\u7406\u5de5\u4f5c\u6d41\u7a0b":89,"\u9884\u6d4b":[81,82,84,87,88],"\u9884\u6d4b\u6d41\u7a0b":4,"\u9884\u6d4bdemo":4,"\u9884\u8bad\u7ec3\u7684\u6a21\u578b":89,"abstract":[39,40,79],"api\u4e2d\u6587\u624b\u518c":3,"api\u5bf9\u6bd4\u4ecb\u7ecd":56,"beam_search\u7684\u751f\u6210":56,"blas\u8def\u5f84\u76f8\u5173\u7684\u7f16\u8bd1\u9009\u9879":48,"bleu\u8bc4\u4f30":89,"book\u4e2d\u6240\u6709\u7ae0\u8282":42,"bool\u578b\u7684\u7f16\u8bd1\u9009\u9879":48,"class":[36,45],"cmake\u6e90\u7801\u7f16\u8bd1":46,"cudnn\u7684\u7f16\u8bd1\u9009\u9879":48,"dataprovider\u7684\u4ecb\u7ecd":1,"dataprovider\u7684\u4f7f\u7528":2,"docker\u4f7f\u7528\u5165\u95e8":49,"filemanager\u8bbe\u8ba1\u6587\u6863":28,"float":46,"function":[18,30],"gpu\u548ccpu\u6df7\u5408\u4f7f\u7528":72,"gpu\u6027\u80fd\u5206\u6790\u4e0e\u8c03\u4f18":67,"gpu\u955c\u50cf\u51fa\u73b0":46,"group\u6559\u7a0b":58,"import":46,"kubernetes\u5206\u5e03\u5f0f\u8bad\u7ec3":75,"kubernetes\u5355\u673a\u8bad\u7ec3":74,"meta\u6587\u4ef6":86,"meta\u914d\u7f6e\u6587\u4ef6":86,"mnist\u7684\u4f7f\u7528\u573a\u666f":2,"movielens\u6570\u636e\u96c6":85,"movielens\u6570\u636e\u96c6\u8bc4\u5206\u56de\u5f52\u6a21\u578b":86,"paddle\u52a8\u6001\u5e93\u4e2d":33,"paddle\u591a\u8bed\u8a00\u63a5\u53e3\u5b9e\u73b0":33,"paddle\u7248\u672c\u53f7\u4e3a0":46,"paddlepaddle\u53d1\u5e03\u7684docker\u955c\u50cf\u4f7f\u7528\u8bf4\u660e":49,"paddlepaddle\u53d1\u884c\u89c4\u8303":42,"paddlepaddle\u56de\u5f52\u6d4b\u8bd5\u5217\u8868":42,"paddlepaddle\u5b58\u50a8\u7684\u53c2\u6570\u683c\u5f0f\u662f\u4ec0\u4e48":46,"paddlepaddle\u7684\u7f16\u8bd1\u9009\u9879":48,"paddlepaddle\u7684docker\u5bb9\u5668\u4f7f\u7528\u65b9\u5f0f":49,"pi\u5e73\u53f0\u4e0a\u7684paddlepaddle\u5e93":54,"pod\u95f4\u901a\u4fe1":75,"pydataprovider2\u7684\u4f7f\u7528":2,"python\u63a5\u53e3":82,"python\u76f8\u5173\u7684\u5355\u5143\u6d4b\u8bd5\u90fd\u8fc7\u4e0d\u4e86":46,"python\u811a\u672c\u8bfb\u53d6\u6570\u636e":84,"return":38,"rnn\u76f8\u5173\u6a21\u578b":57,"rnn\u914d\u7f6e":59,"tensor\u4f7f\u7528\u6837\u4f8b":64,"tensor\u5230eigentensor\u7684\u8f6c\u6362":64,"tensor\u6a21\u5757":64,Abs:5,For:18,The:[17,24,31,36,41],USE:41,Use:17,Using:[18,24],With:27,activ:[5,8],adadelta:10,adagrad:10,adam:10,adamax:10,addto:8,aggreg:8,aggregatelevel:8,alalysi:22,algorithm:[16,19,35,39],all:43,analysi:39,anneal:46,api:[0,3,34,44],appendix:79,applic:3,architectur:39,argument:[29,38],arrai:16,associ:43,assumpt:79,async:70,attent:59,attribut:6,auc:7,auto:16,avg:11,backgraound:16,background:[40,45],backward:[31,41],base:[27,40],basepool:11,basic:79,batch:38,batch_norm:8,batch_siz:38,beam_search:8,becaus:46,benefit:[40,41],between:[15,44],bidirectional_gru:9,bidirectional_lstm:9,big:46,bilinear_interp:8,binari:17,bla:48,block:[17,31,41],block_expand:8,book:49,bootstrap:79,brelu:5,bring:79,build:41,cach:2,can:43,capi:34,capi_priv:34,challeng:40,check:[8,16],checker:16,checkpoint:[19,20,26],choos:18,chunk:7,cifar:12,classif:7,classification_error:7,classification_error_print:7,client:24,clip:8,clone:61,close:16,cluster:[25,79],cmake:18,code:27,column_sum:7,commit:61,compar:79,compil:[17,41],compos:38,comput:[17,31,41],con:79,concat:8,concept:41,conclus:[26,79],config:3,configur:13,conll05:12,connect:8,construct:31,content:[2,34,46,55,67],context_project:8,control:41,conv:8,conv_oper:8,conv_project:8,conv_shift:8,convert:[26,39,40],core:16,cos_sim:8,cost:8,creat:[38,41,43],creation:23,creator:38,crf:8,crf_decod:8,cross_channel_norm:8,cross_entropy_cost:8,cross_entropy_with_selfnorm_cost:8,ctc:8,ctc_error:7,cuda:[46,48],cudnn:48,cudnnavg:11,cudnnmax:11,custom:38,dat:85,data:[8,12,19,38,39],datafeed:12,dataprovid:[3,70],dataset:[12,19,23],datatyp:12,decayedadagrad:10,decor:38,deep:17,definit:45,dens:26,deploi:27,descript:[29,41],design:[15,16,17,19,23,24,25,26,30,31,35,36,37,38,39,40,41,43],destroi:43,detail:22,detect:[7,8],detection_map:7,detection_output:8,develop:41,dictionari:38,differ:41,discuss:40,dispatch:[19,23],distribut:[15,19,22,27,39],doc:[15,17,19,23,24,25,26,30,31,36,37,38,39,40,41],docker:[27,60],dotmul_oper:8,dotmul_project:8,driver:46,drop_out:46,dropout:8,duplic:46,dylib:34,dynam:19,eigen:64,elect:26,els:17,embed:8,enough:16,entri:38,environ:27,eos:8,eval:39,evalu:7,event:[14,15],exampl:[15,18,34],except:46,execut:[17,41],exp:5,expand:[8,55],expandlevel:8,explain:16,faq:46,fault:19,file:17,first_seq:[8,55],fork:61,format:[17,19],forward:31,frame:17,framework:[16,64],from:[15,26,44],full_matrix_project:8,fulli:8,gate:59,gated_unit:8,gener:79,get_output:8,gpu:70,gradient:[16,24],gradient_print:7,graident:16,graph:[31,40,41],group:8,gru:[9,70],gru_group:9,gru_step:8,gru_unit:9,grumemori:8,handler:[15,33],happen:26,have:45,hierarchi:17,high:44,how:[16,22,38,41],hsigmoid:8,huber_classification_cost:8,huber_regression_cost:8,ident:5,identity_project:8,ifelseop:17,illeg:46,imag:[8,9,27],imagenet:82,imdb:[12,88],img_cmrnorm:8,img_conv:8,img_conv_bn_pool:9,img_conv_group:9,img_pool:8,imikolov:12,implement:[16,18,22,35,38],infer:[14,46],ingredi:15,ingress:28,init_hook:2,initi:24,input_typ:2,insid:43,instal:79,instruct:46,insuffici:46,interact:44,interfac:[12,16,19,24,25,38,43],intermedi:41,interpol:8,isn:38,issu:61,job:[19,27,74],join:8,kernel:41,kmax_sequence_scor:8,kubernet:[27,73,74],lambda_cost:8,languag:17,larg:22,last_seq:[8,55],layer:[8,15,30,46],learn:[17,46],learnabl:8,leval:44,level:44,libpaddle_capi_shar:34,libpaddle_capi_whol:34,librari:[24,41],limit:39,linear:5,linear_comb:8,list:[20,38],local:[39,43],lodtensor:35,lodtensordesc:45,log:5,logic:23,low:44,lstm:[9,70,87,88],lstm_step:8,lstmemori:8,lstmemory_group:9,lstmemory_unit:9,macro:41,manag:18,map:[38,41],master:[19,23,27],math:8,mathemat:16,max:11,maxframe_print:7,maxid:8,maxid_print:7,maxout:8,memori:[8,35,46,56,58],messag:[44,46],migrat:41,mileston:41,mini:38,minibatch:12,misc:8,mix:8,mnist:12,model:[3,13,15,22,24,26,59,82],modul:[41,46],momentum:10,movi:85,movielen:12,multi_binary_label_cross_entropy_cost:8,multibox_loss:8,multipl:38,multiplex:8,name:[43,46],nce:8,necessari:41,need:38,nest:35,network:[9,41,59],neural:59,nlp:[9,70],norm:8,note:16,numer:16,numpi:16,nvprof:67,nvvp:67,object:19,onli:[38,43],op_mak:41,oper:[30,40,41],opinfomap:41,opkernel:41,opproto:44,optim:[10,19,24,31],option:29,opwithkernel:41,order:29,org:65,origin:41,orthogon:43,output:8,overview:[26,41,43],packag:18,pad:8,paddl:[22,38,43,46,64],paddlejob:27,paddlepaddl:[15,17,39,42,46,65,78],parallel_nn:72,paramet:[6,8,14,15,19,24,25,27,36,40],parent:43,part:31,partit:24,path:[26,29],perform:70,persist:23,pfsclient:[28,29],pfsserver:28,placement:39,platform:46,pnpair:7,point:46,pool:[8,11,55],power:8,pre:61,precision_recal:7,prefetch:38,prelu:8,print:7,pro:79,procedur:79,process:[19,24,27,41],program:17,programdesc:37,project:18,protocol:46,provid:[2,38],pserver:26,pull:61,push:61,python:[16,27,35,38,39,44,45],qualiti:41,queue:[19,23],rank:7,rank_cost:8,rate:[46,85],reader:[12,15,38],realiz:41,recoveri:19,recurr:[8,9,46,58,59],recurrent_group:8,ref:16,refactor:41,refer:[2,39,40],regist:[41,44],registri:41,regular:24,reject:46,relat:41,relu:5,remot:25,repeat:8,represent:[17,41],request:61,requir:18,reshap:8,resnet:82,retri:23,rmsprop:10,rnn:[35,56,70],rnnop:[17,35,41],rotat:8,row_conv:8,row_l2_norm:8,runtim:[27,39],sampl:8,sampling_id:8,save:26,scale:[8,19],scale_shift:8,scaling_project:8,scope:[17,35,41,43],select:24,selective_fc:8,sentiment:12,separ:41,seq_concat:8,seq_reshap:8,seq_slic:8,seqtext_print:7,sequenc:59,sequence_conv_pool:9,sequencesoftmax:5,server:[19,23,24,27,40],session:39,sextant:79,sgd:70,share:[15,43],should:[43,45],shuffl:38,sigmoid:5,simple_attent:9,simple_gru2:9,simple_gru:9,simple_img_conv_pool:9,simple_lstm:9,singl:38,slice:8,slice_project:8,slope_intercept:8,small_vgg:9,smooth_l1_cost:8,softmax:5,softrelu:5,spars:[24,25,26],spp:8,squar:5,square_error_cost:8,squarerootn:11,stack:[17,88],stanh:5,start:15,step:35,store:19,sub_nested_seq:8,subcommond:29,submit:27,suffici:38,suitabl:18,sum:[7,11],sum_cost:8,sum_to_one_norm:8,summar:15,support:46,survei:79,synopsi:29,tabl:34,table_project:8,take:38,tanh:5,task:[19,23],tecton:79,tensor:[8,41,64],text_conv_pool:9,theori:16,thi:[43,46],timelin:26,todo:[20,21],togeth:43,toler:19,too:46,tool:[18,79],train:[14,15,19,22,25,27,38,39],trainer:[14,19,24,26,27],tran:8,trans_full_matrix_project:8,tune:70,two:16,uci_h:12,updat:[15,25,26],usag:[35,38],use:[22,38],user:[19,85],util:7,value_print:7,vardesc:45,variabl:[41,43,45],version:46,vgg_16_network:9,warp_ctc:8,what:[22,26,45],wheel:46,when:[26,43],whl:46,why:[38,41],wmt14:12,zoo:82}})
\ No newline at end of file
+Search.setIndex({docnames:["api/index_cn","api/v1/data_provider/dataprovider_cn","api/v1/data_provider/pydataprovider2_cn","api/v1/index_cn","api/v1/predict/swig_py_paddle_cn","api/v2/config/activation","api/v2/config/attr","api/v2/config/evaluators","api/v2/config/layer","api/v2/config/networks","api/v2/config/optimizer","api/v2/config/pooling","api/v2/data","api/v2/model_configs","api/v2/run_logic","design/api","design/auto_gradient_check","design/block","design/build_system/README","design/cluster_train/README","design/cluster_train/checkpointing","design/cluster_train/data_dispatch","design/cluster_train/large_model_dist_train","design/cluster_train/master_server","design/cluster_train/pserver_client","design/cluster_train/remote_parameter_updater","design/cluster_train/save_model","design/cluster_train/submit-job","design/file_manager/README","design/file_manager/pfs/pfsclient","design/functions_operators_layers","design/graph","design/if_else_op","design/multi_language_interface/00.why_plain_c","design/multi_language_interface/01.inference_implementation","design/ops/rnn","design/parameters_in_cpp","design/program","design/reader/README","design/refactor/distributed_architecture","design/refactor/parameter_server","design/refactorization","design/releasing_process","design/scope","design/simple_op_design","design/var_desc","faq/build_and_install/index_cn","faq/cluster/index_cn","faq/index_cn","faq/local/index_cn","faq/model/index_cn","faq/parameter/index_cn","getstarted/basic_usage/index_cn","getstarted/build_and_install/cmake/build_from_source_cn","getstarted/build_and_install/docker_install_cn","getstarted/build_and_install/index_cn","getstarted/concepts/use_concepts_cn","getstarted/index_cn","howto/cross_compiling/cross_compiling_for_android_cn","howto/cross_compiling/cross_compiling_for_raspberry_cn","howto/deep_model/rnn/hierarchical_layer_cn","howto/deep_model/rnn/hrnn_rnn_api_compare_cn","howto/deep_model/rnn/index_cn","howto/deep_model/rnn/recurrent_group_cn","howto/deep_model/rnn/rnn_config_cn","howto/dev/build_cn","howto/dev/contribute_to_paddle_cn","howto/dev/new_layer_cn","howto/dev/new_op_cn","howto/dev/use_eigen_cn","howto/dev/write_docs_cn","howto/index_cn","howto/optimization/gpu_profiling_cn","howto/usage/cluster/cluster_train_cn","howto/usage/cmd_parameter/arguments_cn","howto/usage/cmd_parameter/detail_introduction_cn","howto/usage/cmd_parameter/index_cn","howto/usage/cmd_parameter/use_case_cn","howto/usage/k8s/k8s_basis_cn","howto/usage/k8s/k8s_cn","howto/usage/k8s/k8s_distributed_cn","howto/usage/k8s/src/k8s_data/README","howto/usage/k8s/src/k8s_train/README","index_cn","survey/cluster_bootstrapping_tools","tutorials/embedding_model/index_cn","tutorials/image_classification/index_cn","tutorials/imagenet_model/resnet_model_cn","tutorials/index_cn","tutorials/quick_start/index_cn","tutorials/rec/ml_dataset_cn","tutorials/rec/ml_regression_cn","tutorials/semantic_role_labeling/index_cn","tutorials/sentiment_analysis/index_cn","tutorials/text_generation/index_cn"],envversion:50,filenames:["api/index_cn.rst","api/v1/data_provider/dataprovider_cn.rst","api/v1/data_provider/pydataprovider2_cn.rst","api/v1/index_cn.rst","api/v1/predict/swig_py_paddle_cn.rst","api/v2/config/activation.rst","api/v2/config/attr.rst","api/v2/config/evaluators.rst","api/v2/config/layer.rst","api/v2/config/networks.rst","api/v2/config/optimizer.rst","api/v2/config/pooling.rst","api/v2/data.rst","api/v2/model_configs.rst","api/v2/run_logic.rst","design/api.md","design/auto_gradient_check.md","design/block.md","design/build_system/README.md","design/cluster_train/README.md","design/cluster_train/checkpointing.md","design/cluster_train/data_dispatch.md","design/cluster_train/large_model_dist_train.md","design/cluster_train/master_server.md","design/cluster_train/pserver_client.md","design/cluster_train/remote_parameter_updater.md","design/cluster_train/save_model.md","design/cluster_train/submit-job.md","design/file_manager/README.md","design/file_manager/pfs/pfsclient.md","design/functions_operators_layers.md","design/graph.md","design/if_else_op.md","design/multi_language_interface/00.why_plain_c.md","design/multi_language_interface/01.inference_implementation.md","design/ops/rnn.md","design/parameters_in_cpp.md","design/program.md","design/reader/README.md","design/refactor/distributed_architecture.md","design/refactor/parameter_server.md","design/refactorization.md","design/releasing_process.md","design/scope.md","design/simple_op_design.md","design/var_desc.md","faq/build_and_install/index_cn.rst","faq/cluster/index_cn.rst","faq/index_cn.rst","faq/local/index_cn.rst","faq/model/index_cn.rst","faq/parameter/index_cn.rst","getstarted/basic_usage/index_cn.rst","getstarted/build_and_install/cmake/build_from_source_cn.rst","getstarted/build_and_install/docker_install_cn.rst","getstarted/build_and_install/index_cn.rst","getstarted/concepts/use_concepts_cn.rst","getstarted/index_cn.rst","howto/cross_compiling/cross_compiling_for_android_cn.md","howto/cross_compiling/cross_compiling_for_raspberry_cn.md","howto/deep_model/rnn/hierarchical_layer_cn.rst","howto/deep_model/rnn/hrnn_rnn_api_compare_cn.rst","howto/deep_model/rnn/index_cn.rst","howto/deep_model/rnn/recurrent_group_cn.md","howto/deep_model/rnn/rnn_config_cn.rst","howto/dev/build_cn.md","howto/dev/contribute_to_paddle_cn.md","howto/dev/new_layer_cn.rst","howto/dev/new_op_cn.md","howto/dev/use_eigen_cn.md","howto/dev/write_docs_cn.rst","howto/index_cn.rst","howto/optimization/gpu_profiling_cn.rst","howto/usage/cluster/cluster_train_cn.md","howto/usage/cmd_parameter/arguments_cn.md","howto/usage/cmd_parameter/detail_introduction_cn.md","howto/usage/cmd_parameter/index_cn.rst","howto/usage/cmd_parameter/use_case_cn.md","howto/usage/k8s/k8s_basis_cn.md","howto/usage/k8s/k8s_cn.md","howto/usage/k8s/k8s_distributed_cn.md","howto/usage/k8s/src/k8s_data/README.md","howto/usage/k8s/src/k8s_train/README.md","index_cn.rst","survey/cluster_bootstrapping_tools.md","tutorials/embedding_model/index_cn.md","tutorials/image_classification/index_cn.md","tutorials/imagenet_model/resnet_model_cn.md","tutorials/index_cn.md","tutorials/quick_start/index_cn.rst","tutorials/rec/ml_dataset_cn.md","tutorials/rec/ml_regression_cn.rst","tutorials/semantic_role_labeling/index_cn.md","tutorials/sentiment_analysis/index_cn.md","tutorials/text_generation/index_cn.md"],objects:{},objnames:{},objtypes:{},terms:{"00012\u7684\u6a21\u578b\u6709\u7740\u6700\u9ad8\u7684bleu\u503c27":94,"0005\u4e58\u4ee5batch":86,"000\u4e2a\u5df2\u6807\u6ce8\u8fc7\u7684\u9ad8\u6781\u6027\u7535\u5f71\u8bc4\u8bba\u7528\u4e8e\u8bad\u7ec3":93,"000\u4e2a\u7528\u4e8e\u6d4b\u8bd5":93,"000\u4e2atxt\u6587\u4ef6":93,"000\u4f4d\u7528\u6237\u5bf94":90,"000\u5e45\u56fe\u50cf\u4e0a\u6d4b\u8bd5\u4e86\u6a21\u578b\u7684\u5206\u7c7b\u9519\u8bef\u7387":87,"000\u5f20\u7070\u5ea6\u56fe\u7247\u7684\u6570\u5b57\u5206\u7c7b\u6570\u636e\u96c6":2,"000\u6761\u8bc4\u4ef7":90,"000\u90e8\u7535\u5f71\u76841":90,"00186201e":4,"00m":72,"01852v1":8,"03m":72,"0424m":72,"0473v3":9,"05d":86,"0630u":72,"06u":72,"0810u":72,"08823112e":4,"0957m":72,"0\u4e4b\u540e\u7684\u7248\u672c":54,"0\u53f7\u8bad\u7ec3\u8282\u70b9\u662f\u4e3b\u8bad\u7ec3\u8282\u70b9":75,"0\u5c42\u5e8f\u5217":60,"0\u8868\u793a\u7b2c\u4e00\u6b21\u7ecf\u8fc7\u8bad\u7ec3\u96c6":93,"0ab":8,"0rc1":42,"0rc2":42,"10000\u5f20\u4f5c\u4e3a\u6d4b\u8bd5\u96c6":86,"10007_10":93,"10014_7":93,"100m":49,"101\u5c42\u548c152\u5c42\u7684\u7f51\u7edc\u7ed3\u6784\u4e2d":87,"101\u5c42\u548c152\u5c42\u7684\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6\u53ef\u53c2\u7167":87,"101\u5c42\u7f51\u7edc\u6a21\u578b":87,"10\u4e2d\u7684\u8bad\u7ec3\u96c6\u548c\u6d4b\u8bd5\u96c6":86,"10\u6570\u636e\u96c6":86,"10\u6570\u636e\u96c6\u5305\u542b60000\u5f2032x32\u7684\u5f69\u8272\u56fe\u7247":86,"10\u6570\u636e\u96c6\u7684\u5b98\u65b9\u7f51\u5740":86,"10\u6570\u636e\u96c6\u8bad\u7ec3\u4e00\u4e2a\u5377\u79ef\u795e\u7ecf\u7f51\u7edc":86,"10g":27,"1150u":72,"11\u5b9e\u73b0\u4e86c":34,"11e6":79,"12194102e":4,"124n":72,"128\u7ef4\u548c256\u7ef4":85,"12\u64cd\u4f5c\u7cfb\u7edf":46,"13m":79,"1490u":72,"14\u4ee5\u4e0a":59,"14\u6570\u636e\u96c6":94,"14\u6570\u636e\u96c6\u4e0a\u5f97\u5230\u826f\u597d\u8868\u73b0\u7684\u8bad\u7ec3\u8fc7\u7a0b":94,"14\u8fd9\u79cd\u5199\u6cd5\u5c06\u4f1a\u6d4b\u8bd5\u6a21\u578b":77,"152\u5c42\u7f51\u7edc\u6a21\u578b":87,"15501715e":4,"1550u":72,"15\u884c":61,"1636k":94,"16\u5b57\u8282\u8868\u793a\u4fdd\u5b58\u7684\u53c2\u6570\u603b\u4e2a\u6570":51,"16u":72,"173m":87,"173n":72,"1770u":72,"18\u5c81\u4ee5\u4e0b":90,"18e457ce3d362ff5f3febf8e7f85ffec852f70f3b629add10aed84f930a68750":79,"197u":72,"1\u7684\u5c42\u4e4b\u5916":77,"1\u7a00\u758f\u6570\u636e":67,"1\u8f6e\u5b58\u50a8\u7684\u6240\u6709\u6a21\u578b":77,"1\u9664\u4ee5batch":86,"1m\u6570\u636e\u96c6\u4e2d":91,"1m\u7684\u5b57\u6bb5\u914d\u7f6e\u6587\u4ef6\u5728\u76ee\u5f55":91,"200\u6570\u636e\u96c6\u4e0a\u4f7f\u7528vgg\u6a21\u578b\u8bad\u7ec3\u4e00\u4e2a\u9e1f\u7c7b\u56fe\u7247\u5206\u7c7b\u6a21\u578b":86,"210u":72,"211839e770f7b538e2d8":9,"215n":72,"228u":72,"234m":87,"24\u5c81":90,"2520u":72,"25639710e":4,"25k":89,"2680u":72,"26\u884c":61,"27787406e":4,"279n":72,"27m":72,"285m":72,"2863m":72,"28\u7684\u56fe\u7247\u50cf\u7d20\u7070\u5ea6\u503c":2,"28\u7ef4\u7684\u7a20\u5bc6\u6d6e\u70b9\u6570\u5411\u91cf\u548c\u4e00\u4e2a":2,"28m":72,"2977m":72,"29997\u4e2a\u6700\u9ad8\u9891\u5355\u8bcd\u548c3\u4e2a\u7279\u6b8a\u7b26\u53f7":94,"2\u4e09\u7c7b\u7684\u6bd4\u4f8b\u4e3a":51,"2\u4e2a\u6d6e\u70b9\u6570":52,"2\u5206\u522b\u4ee3\u88683\u4e2a\u8282\u70b9\u7684trainer":80,"2\u610f\u5473\u77400\u53f7\u548c1\u53f7gpu\u5c06\u4f1a\u4f7f\u7528\u6570\u636e\u5e76\u884c\u6765\u8ba1\u7b97fc1\u548cfc2\u5c42":77,"2\u8fd9\u51e0\u4e2a\u76ee\u5f55\u8868\u793apaddlepaddle\u8282\u70b9\u4e0etrain":80,"2nd":8,"302n":72,"30u":72,"3206326\u4e2a\u8bcd\u548c4\u4e2a\u7279\u6b8a\u6807\u8bb0":85,"32777140e":4,"328n":72,"32\u7ef4":85,"32u":72,"32x32":12,"331n":72,"3320u":72,"34\u5c81":90,"35\u65f6":94,"36540484e":4,"36u":72,"3710m":72,"3768m":72,"387u":72,"38u":72,"3920u":72,"39u":72,"3\u4e2a\u7279\u6b8a\u7b26\u53f7":94,"3\u53f7gpu":49,"4035m":72,"4090u":72,"4096mb":75,"4279m":72,"43630644e":4,"43u":72,"448a5b355b84":79,"44\u5c81":90,"4560u":72,"4563m":72,"45u":72,"4650u":72,"4726m":72,"473m":79,"48565123e":4,"48684503e":4,"49316648e":4,"49\u5c81":90,"4\u5b57\u8282\u8868\u793apaddlepaddle\u7248\u672c\u4fe1\u606f":51,"4gb":75,"4yf":8,"500\u4e2atxt\u6587\u4ef6":93,"500m":49,"50\u5c42":87,"50\u5c42\u7f51\u7edc\u6a21\u578b":87,"51111044e":4,"514u":72,"525n":72,"526u":72,"53018653e":4,"536u":72,"5460u":72,"5470u":72,"54u":72,"55\u5c81":90,"55g":94,"5690m":72,"573u":72,"578n":72,"5798m":72,"586u":72,"58s":79,"5969m":72,"5\u4e2a\u6d4b\u8bd5\u6837\u4f8b\u548c2\u4e2a\u751f\u6210\u5f0f\u6837\u4f8b":85,"5delta":8,"6080u":72,"6082v4":8,"6140u":72,"6305m":72,"639u":72,"64\u5e73\u53f0\u4e3a\u4f8b":[58,59],"64\u73af\u5883":59,"64\u7ef4":85,"655u":72,"6780u":72,"6810u":72,"682u":72,"6970u":72,"6\u4e07\u4ebf\u6b21\u6d6e\u70b9\u8fd0\u7b97\u6bcf\u79d2":72,"6\u4e2a\u8282\u70b9":73,"6\u5143\u4e0a\u4e0b\u6587\u4f5c\u4e3a\u8f93\u5165\u5c42":85,"704u":72,"70634608e":4,"7090u":72,"72296313e":4,"72u":72,"73u":72,"75u":72,"760u":72,"767u":72,"783n":72,"784u":72,"78m":72,"7\u548cpip":46,"7\u7248\u672c\u5f00\u59cb":58,"7eamaa":12,"7kb":79,"8250u":72,"8300u":72,"830n":72,"849m":72,"85625684e":4,"861u":72,"864k":94,"8661m":72,"877\u4e2a\u88ab\u5411\u91cf\u5316\u7684\u8bcd":85,"877\u884c":85,"892m":72,"8\u5b57\u8282\u8868\u793a\u6bcf\u4e2a\u53c2\u6570\u5360\u7528\u7684\u5b57\u8282\u6570":51,"901n":72,"90u":72,"918u":72,"9247m":72,"924n":72,"9261m":72,"93137714e":4,"9330m":72,"94u":72,"9530m":72,"96644767e":4,"983m":72,"988u":72,"997u":72,"99982715e":4,"99m":87,"99u":72,"9\u4e2d\u7684\u4e00\u4e2a\u6570\u5b57":2,"9f18":79,"\u0233":52,"\u03b5":52,"\u4e00":61,"\u4e00\u4e2a0\u5c42\u5e8f\u5217":60,"\u4e00\u4e2a0\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u6269\u5c55\u6210\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":60,"\u4e00\u4e2a\u4e0d\u5171\u4eab\u7684\u4f8b\u5b50\u662f":68,"\u4e00\u4e2a\u4f8b\u5b50\u662f\u623f\u4ea7\u4f30\u503c":52,"\u4e00\u4e2a\u5178\u578b\u7684\u795e\u7ecf\u7f51\u7edc\u5982\u4e0b\u56fe\u6240\u793a":86,"\u4e00\u4e2a\u5178\u578b\u7684chunk\u5982\u4e0b\u6240\u793a":28,"\u4e00\u4e2a\u5206\u5e03\u5f0f\u7684\u5b58\u50a8\u7cfb\u7edf":78,"\u4e00\u4e2a\u5206\u5e03\u5f0fpaddle\u8bad\u7ec3\u4efb\u52a1\u4e2d\u7684\u6bcf\u4e2a\u8fdb\u7a0b\u90fd\u53ef\u4ee5\u4ececeph\u8bfb\u53d6\u6570\u636e":79,"\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":60,"\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217\u6216\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":60,"\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u6269\u5c55\u6210\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":60,"\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217\u8fdb\u5165":63,"\u4e00\u4e2a\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u5305\u542b\u5982\u4e0b\u5c42":86,"\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":60,"\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217\u6216\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":60,"\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u53d8\u6210\u4e00\u4e2a0\u5c42\u5e8f\u5217":60,"\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u53d8\u6210\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":60,"\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217\u8fdb\u5165":63,"\u4e00\u4e2a\u53cc\u5c42rnn\u7531\u591a\u4e2a\u5355\u5c42rnn\u7ec4\u6210":63,"\u4e00\u4e2a\u53ef\u8c03\u7528\u7684\u51fd\u6570":63,"\u4e00\u4e2a\u57fa\u672c\u7684\u5e94\u7528\u573a\u666f\u662f\u533a\u5206\u7ed9\u5b9a\u6587\u672c\u7684\u8912\u8d2c\u4e24\u6781\u6027":93,"\u4e00\u4e2a\u5bb9\u5668\u5c31\u662f\u4e00\u4e2a\u64cd\u4f5c\u7cfb\u7edf\u7684\u8fdb\u7a0b":54,"\u4e00\u4e2a\u6216\u591a\u4e2a":78,"\u4e00\u4e2a\u6570\u636e\u96c6\u5927\u90e8\u5206\u5e8f\u5217\u957f\u5ea6\u662f100":49,"\u4e00\u4e2a\u6587\u4ef6":2,"\u4e00\u4e2a\u662f\u6d6e\u70b9\u8ba1\u7b97\u91cf":72,"\u4e00\u4e2a\u662f\u76f4\u63a5\u628a\u4e00\u4e2a\u5bb9\u5668\u8f6c\u6362\u6210\u955c\u50cf":54,"\u4e00\u4e2a\u72ec\u7acb\u7684\u5143\u7d20":60,"\u4e00\u4e2a\u72ec\u7acb\u7684\u8bcd\u8bed":60,"\u4e00\u4e2a\u795e\u7ecf\u7f51\u7edc\u7684\u6a21\u578b\u7531\u5927\u91cf\u7684\u53c2\u6570\u7ec4\u6210":20,"\u4e00\u4e2a\u7b80\u5355\u7684\u4f8b\u5b50\u5982":93,"\u4e00\u4e2a\u7ec8\u7aef\u8fd0\u884cvi":54,"\u4e00\u4e2a\u7f51\u7edc\u5c42\u7684\u524d\u5411\u4f20\u64ad\u90e8\u5206\u628a\u8f93\u5165\u8f6c\u5316\u4e3a\u76f8\u5e94\u7684\u8f93\u51fa":67,"\u4e00\u4e2a\u7f51\u7edc\u5c42\u7684\u53c2\u6570\u662f\u5728":67,"\u4e00\u4e2a\u7f51\u7edc\u5c42\u7684c":67,"\u4e00\u4e2a\u8f93\u51fa\u7ec4\u6210":68,"\u4e00\u4e2a\u91cd\u8981\u7684\u95ee\u9898\u662f\u9009\u62e9\u6b63\u786e\u7684learning_r":51,"\u4e00\u4e2achunk\u7531\u6240\u5728\u7684\u6587\u4ef6\u504f\u79fb":28,"\u4e00\u4e2adocker\u955c\u50cf\u662f\u4e00\u4e2a\u6253\u5305\u597d\u7684\u8f6f\u4ef6":54,"\u4e00\u4e2agpu\u8bbe\u5907\u4e0a\u4e0d\u5141\u8bb8\u914d\u7f6e\u591a\u4e2a\u6a21\u578b":75,"\u4e00\u4e2alabel":61,"\u4e00\u4e2alogging\u5bf9\u8c61":2,"\u4e00\u4e2amemory\u5305\u542b":64,"\u4e00\u4e2apass\u610f\u5473\u7740paddlepaddle\u8bad\u7ec3\u6570\u636e\u96c6\u4e2d\u7684\u6240\u6709\u6837\u672c\u88ab\u904d\u5386\u4e00\u6b21":92,"\u4e00\u4e2apass\u8868\u793a\u8fc7\u4e00\u904d\u6240\u6709\u8bad\u7ec3\u6837\u672c":89,"\u4e00\u4e2apod\u4e2d\u7684\u6240\u6709\u5bb9\u5668\u4f1a\u88ab\u8c03\u5ea6\u5230\u540c\u4e00\u4e2anode\u4e0a":78,"\u4e00\u4e2aposix\u517c\u5bb9\u7684\u6587\u4ef6\u7cfb\u7edf":28,"\u4e00\u4eba":61,"\u4e00\u53e5\u8bdd\u662f\u7531\u8bcd\u8bed\u6784\u6210\u7684\u5e8f\u5217":63,"\u4e00\u53f0\u673a\u5668\u4e0a\u9762\u7684\u7ebf\u7a0b\u6570\u91cf":91,"\u4e00\u53f0\u7535\u8111":65,"\u4e00\u65e9":61,"\u4e00\u662fbatch":49,"\u4e00\u6761\u6837\u672c":2,"\u4e00\u6837\u7684\u65b9\u5f0f":65,"\u4e00\u6837\u8bbe\u7f6e":73,"\u4e00\u6b21\u4f5c\u4e1a\u79f0\u4e3a\u4e00\u4e2ajob":78,"\u4e00\u6b21\u6027\u676f\u5b50":61,"\u4e00\u6b21yield\u8c03\u7528":2,"\u4e00\u79cd\u5e38\u7528\u7684\u505a\u6cd5\u662f\u7528\u5b66\u4e60\u7684\u6a21\u578b\u5bf9\u53e6\u5916\u4e00\u7ec4\u6d4b\u8bd5\u6570\u636e\u8fdb\u884c\u9884\u6d4b":52,"\u4e00\u79cd\u5e38\u7528\u7684cmake\u914d\u7f6e\u5982\u4e0b":58,"\u4e00\u7bc7\u8bba\u6587":94,"\u4e00\u7ea7\u76ee\u5f55":[93,94],"\u4e00\u81f4":[60,61],"\u4e00\u822c\u4e0d\u5141\u8bb8\u518d\u4ece":42,"\u4e00\u822c\u4ece":66,"\u4e00\u822c\u5728paddlepaddle\u4e2d":61,"\u4e00\u822c\u60c5\u51b5\u4e0b":[1,52],"\u4e00\u822c\u63a8\u8350\u8bbe\u7f6e\u6210true":2,"\u4e00\u822c\u662f\u7531\u4e8e\u76f4\u63a5\u4f20\u9012\u5927\u5b57\u5178\u5bfc\u81f4\u7684":51,"\u4e00\u822c\u6765\u8bf4":64,"\u4e00\u822c\u800c\u8a00":94,"\u4e00\u822c\u8868\u793a":61,"\u4e00\u822c\u8bbe\u7f6e":51,"\u4e00\u884c\u4e3a\u4e00\u4e2a\u6837\u672c":89,"\u4e09\u79cd\u5e8f\u5217\u6a21\u5f0f":[2,56],"\u4e09\u7ea7\u76ee\u5f55":[93,94],"\u4e0a":66,"\u4e0a\u4e0b\u6587\u5927\u5c0f\u8bbe\u7f6e\u4e3a1\u7684\u4e00\u4e2a\u6837\u672c\u7684\u7279\u5f81\u5982\u4e0b":92,"\u4e0a\u4f20\u5230cloud\u6216\u8005\u4e0b\u8f7d\u5230\u672c\u5730\u7684\u65f6\u95f4\u53ef\u80fd\u6bd4\u8f83\u957f":28,"\u4e0a\u4f20\u65b9\u6cd5":42,"\u4e0a\u53d1\u8868\u7684\u8bc4\u8bba\u5206\u6210\u6b63\u9762\u8bc4\u8bba\u548c\u8d1f\u9762\u8bc4\u8bba\u4e24\u7c7b":93,"\u4e0a\u56fe\u4e2d\u865a\u7ebf\u7684\u8fde\u63a5":61,"\u4e0a\u56fe\u63cf\u8ff0\u4e86\u4e00\u4e2a3\u8282\u70b9\u7684\u5206\u5e03\u5f0f\u8bad\u7ec3\u573a\u666f":80,"\u4e0a\u6ce8\u518c\u4e00\u4e0b":28,"\u4e0a\u7f16\u8bd1\u5f88\u6162":65,"\u4e0a\u7f16\u8bd1android\u5e73\u53f0\u4e0a\u9002\u7528\u7684paddlepaddle\u5e93":58,"\u4e0a\u7f51":61,"\u4e0a\u8fd0\u884c":54,"\u4e0a\u8ff0\u4ee3\u7801\u5c06bias\u5168\u90e8\u521d\u59cb\u5316\u4e3a1":51,"\u4e0a\u8ff0\u547d\u4ee4\u7f16\u8bd1\u51fa\u4e00\u4e2a":65,"\u4e0a\u8ff0\u7684":50,"\u4e0a\u8ff0\u7684\u4ee3\u7801\u7247\u6bb5\u5305\u542b\u4e86\u4e24\u79cd\u65b9\u6cd5":72,"\u4e0a\u8ff0\u811a\u672c\u4f7f\u7528":73,"\u4e0a\u9762\u7684\u4ee3\u7801\u5728":68,"\u4e0a\u9762\u7684\u4ee3\u7801\u9996\u5148\u5bfc\u5165\u4f9d\u8d56\u7684\u5305":68,"\u4e0b":[68,86],"\u4e0b\u540c":51,"\u4e0b\u56fe\u4e2d\u5c31\u5c55\u793a\u4e86\u4e00\u4e9b\u5173\u4e8e\u5185\u5b58\u6570\u636e\u8fc1\u5f99\u548c\u8ba1\u7b97\u8d44\u6e90\u5229\u7528\u7387\u7684\u5efa\u8bae":72,"\u4e0b\u56fe\u5c55\u793a\u4e86\u6240\u6709\u7684\u56fe\u7247\u7c7b\u522b":86,"\u4e0b\u56fe\u5c55\u793a\u4e86\u65f6\u95f4\u6269\u5c55\u76842\u5c42":92,"\u4e0b\u56fe\u5c55\u793a\u7684\u662f\u57fa\u4e8e\u6b8b\u5dee\u7684\u8fde\u63a5\u65b9\u5f0f":87,"\u4e0b\u56fe\u662f\u4e00\u4e2a\u5168\u8fde\u63a5\u5c42\u7684\u793a\u610f\u56fe":67,"\u4e0b\u5b58\u653e\u516c\u5171\u6570\u636e\u96c6\u5408":21,"\u4e0b\u6587\u4ee5nlp\u4efb\u52a1\u4e3a\u4f8b":63,"\u4e0b\u6587\u4f7f\u7528":80,"\u4e0b\u6587\u5c31\u662f\u7528job\u7c7b\u578b\u7684\u8d44\u6e90\u6765\u8fdb\u884c\u8bad\u7ec3":79,"\u4e0b\u6b21":61,"\u4e0b\u7684":80,"\u4e0b\u7684\u4f5c\u4e3a\u7f16\u8bd1\u5de5\u5177":59,"\u4e0b\u8868\u5c55\u793a\u4e86batch":87,"\u4e0b\u8f7d":28,"\u4e0b\u8f7d\u5230\u672c\u5730":28,"\u4e0b\u8f7d\u5b8c\u6570\u636e\u540e":79,"\u4e0b\u8f7d\u6570\u636e\u96c6":86,"\u4e0b\u8f7draspberri":59,"\u4e0b\u8f7dwmt":94,"\u4e0b\u9762\u4e3e\u4e2a\u7b80\u5355\u7684\u4f8b\u5b50":72,"\u4e0b\u9762\u4ecb\u7ecd\u4ecb\u7ecd":68,"\u4e0b\u9762\u4ecb\u7ecd\u9884\u5904\u7406\u8fc7\u7a0b\u5177\u4f53\u7684\u6b65\u9aa4":91,"\u4e0b\u9762\u4ee5\u77e9\u9635\u4e58\u64cd\u4f5c":68,"\u4e0b\u9762\u4ee5addop\u4e3a\u4f8b\u8bf4\u660etensor\u7684\u4f7f\u7528\u8fc7\u7a0b":69,"\u4e0b\u9762\u5148\u7b80\u8981\u4ecb\u7ecd\u4e00\u4e0b\u672c\u6587\u7528\u5230\u7684\u51e0\u4e2akubernetes\u6982\u5ff5":78,"\u4e0b\u9762\u5206\u522b\u4ecb\u7ecd\u67d0\u4e00\u7c7b\u6587\u4ef6\u7684\u5b9e\u73b0\u65b9\u5f0f":34,"\u4e0b\u9762\u5217\u51fa\u4e86":64,"\u4e0b\u9762\u5217\u51fa\u4e86\u5168\u8fde\u63a5\u5c42\u7684\u68af\u5ea6\u68c0\u67e5\u5355\u5143\u6d4b\u8bd5":67,"\u4e0b\u9762\u5c31\u6839\u636e\u8fd9\u51e0\u4e2a\u6b65\u9aa4\u5206\u522b\u4ecb\u7ecd":80,"\u4e0b\u9762\u6211\u4eec\u7ed9\u51fa\u4e86\u4e00\u4e2a\u914d\u7f6e\u793a\u4f8b":86,"\u4e0b\u9762\u662f":68,"\u4e0b\u9762\u662f\u4e00\u4e2a\u8bef\u5dee\u66f2\u7ebf\u56fe\u7684\u793a\u4f8b":86,"\u4e0b\u9762\u662f\u5bf9":68,"\u4e0b\u9762\u662fcifar":86,"\u4e0b\u9762\u7684\u4ee3\u7801\u5c06\u968f\u673a\u751f\u6210\u7684\u77e9\u9635\u8f6c\u5316\u4e3a\u53ef\u4ee5\u88abpaddlepaddle\u52a0\u8f7d\u7684\u6a21\u578b\u53c2\u6570":51,"\u4e0b\u9762\u7684\u4ee3\u7801\u7247\u6bb5\u5b9e\u73b0\u4e86":67,"\u4e0b\u9762\u7684\u4f8b\u5b50\u4f7f\u7528\u4e86":87,"\u4e0b\u9762\u7684\u4f8b\u5b50\u540c\u6837\u4f7f\u7528\u4e86":87,"\u4e0b\u9762\u7684\u547d\u4ee4\u628a\u5f53\u524d\u76ee\u5f55\u6302\u8f7d\u5230\u4e86\u5bb9\u5668\u4e2d\u7684":54,"\u4e0b\u9762\u7684\u70b9\u5b9e\u73b0\u4e86mulop\u7684\u5b9a\u4e49":68,"\u4e0b\u9762\u7ed9\u51fa\u4e86\u4e00\u4e2a\u4f8b\u5b50":67,"\u4e0b\u9762\u7ed9\u51fa\u5728\u4e09\u7ef4\u7a7a\u95f4\u4e2d\u4f7f\u7528\u7ebf\u6027\u56de\u5f52\u62df\u5408\u4e00\u6761\u76f4\u7ebf\u7684\u4f8b\u5b50":56,"\u4e0b\u9762\u811a\u672c\u7b26\u5408paddlepaddle\u671f\u5f85\u7684\u8bfb\u53d6\u6570\u636e\u7684python\u7a0b\u5e8f\u7684\u6a21\u5f0f":52,"\u4e0b\u9762\u89e3\u91ca\u4ee3\u7801\u4e2d\u4e00\u4e9b\u5173\u952e\u7684\u5730\u65b9":68,"\u4e0b\u9762\u8fd9\u4e9blayer\u80fd\u591f\u63a5\u53d7\u53cc\u5c42\u5e8f\u5217\u4f5c\u4e3a\u8f93\u5165":60,"\u4e0d":61,"\u4e0d\u4e00\u5b9a\u548c\u65f6\u95f4\u6709\u5173\u7cfb":2,"\u4e0d\u4e00\u81f4\u7684\u7531pfsclient\u4e0b\u8f7d\u6216\u8005\u4f20\u8f93chunk\u5b8c\u6210":28,"\u4e0d\u4f1a\u4fdd\u7559\u5728\u78c1\u76d8\u4e0a":65,"\u4e0d\u4f1a\u518d\u4ece":49,"\u4e0d\u4f1a\u865a\u62df\u4efb\u4f55\u786c\u4ef6":65,"\u4e0d\u4f7f\u7528\u9759\u6001\u5e93":33,"\u4e0d\u4f7f\u7528\u989d\u5916\u7a7a\u95f4":67,"\u4e0d\u4f7f\u7528c":33,"\u4e0d\u4f7f\u7528swig":33,"\u4e0d\u5141\u8bb8\u4e00\u4e2a\u6587\u4ef6\u4e2d\u5305\u542b\u591a\u4e2aop":68,"\u4e0d\u5171\u4eab\u5219\u4e0d\u52a0":68,"\u4e0d\u5171\u4eab\u7684\u4f8b\u5b50\u53ef\u4ee5\u53c2\u8003":68,"\u4e0d\u5305\u542b\u5728\u5b57\u5178\u4e2d\u7684\u5355\u8bcd":94,"\u4e0d\u540c":92,"\u4e0d\u540c\u4e3b\u673a":78,"\u4e0d\u540c\u4e8e\u4e0a\u8ff0\u4ecb\u7ecd\u7684recurr":50,"\u4e0d\u540c\u4e8eop\u7684\u7f16\u8bd1\u6d4b\u8bd5":68,"\u4e0d\u540c\u4ea7\u54c1":93,"\u4e0d\u540c\u53c2\u6570\u670d\u52a1\u5668\u4e0a\u6570\u636e\u5927\u5c0f\u7684\u6700\u5927\u503c\u4e0e\u6700\u5c0f\u503c\u7684\u6bd4\u7387":75,"\u4e0d\u540c\u5c42\u7684\u7279\u5f81\u7531\u5206\u53f7":87,"\u4e0d\u540c\u65f6\u95f4\u6b65\u7684\u8f93\u5165\u662f\u4e0d\u540c\u7684":64,"\u4e0d\u540c\u7248\u672c\u7684\u7f16\u8bd1\u5668\u4e4b\u95f4":33,"\u4e0d\u540c\u7684\u4f18\u5316\u7b97\u6cd5\u9700\u8981\u4f7f\u7528\u4e0d\u540c\u5927\u5c0f\u7684\u5185\u5b58":49,"\u4e0d\u540c\u7684\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf":80,"\u4e0d\u540c\u7684\u6570\u636e\u7c7b\u578b\u548c\u5e8f\u5217\u6a21\u5f0f\u8fd4\u56de\u7684\u683c\u5f0f\u4e0d\u540c":[2,56],"\u4e0d\u540c\u7a7a\u95f4\u7684\u8d44\u6e90\u540d\u53ef\u4ee5\u91cd\u590d":78,"\u4e0d\u540c\u8bbe\u5907":68,"\u4e0d\u540c\u8bed\u8a00\u7684\u63a5\u53e3\u9002\u5e94\u4e0d\u540c\u8bed\u8a00\u7684\u7279\u6027":33,"\u4e0d\u540c\u8f93\u5165\u542b\u6709\u7684\u5b50\u53e5":63,"\u4e0d\u540c\u8f93\u5165\u5e8f\u5217\u542b\u6709\u7684\u8bcd\u8bed\u6570\u5fc5\u987b\u4e25\u683c\u76f8\u7b49":63,"\u4e0d\u540cdataprovider\u5bf9\u6bd4\u5982\u4e0b":61,"\u4e0d\u540cpod\u4e4b\u95f4\u53ef\u4ee5\u901a\u8fc7ip\u5730\u5740\u8bbf\u95ee":78,"\u4e0d\u540crank\u7684tensor\u662f\u4e0d\u540c\u7c7b\u578b":69,"\u4e0d\u5728":34,"\u4e0d\u5bb9\u6613\u51fa\u9519":28,"\u4e0d\u5c11":61,"\u4e0d\u5d4c\u5165\u5176\u4ed6\u8bed\u8a00\u89e3\u91ca\u5668":33,"\u4e0d\u5d4c\u5165python\u89e3\u91ca\u5668":33,"\u4e0d\u5e94\u8be5\u88ab\u62c6\u89e3":63,"\u4e0d\u5fc5\u518d\u5c06\u4efb\u610f\u957f\u5ea6\u6e90\u8bed\u53e5\u4e2d\u7684\u6240\u6709\u4fe1\u606f\u538b\u7f29\u81f3\u4e00\u4e2a\u5b9a\u957f\u7684\u5411\u91cf\u4e2d":94,"\u4e0d\u6307\u5b9a\u65f6":63,"\u4e0d\u63d0\u4f9b\u5206\u5e03\u5f0f\u5b58\u50a8":78,"\u4e0d\u652f\u6301":92,"\u4e0d\u662f\u4e00\u6761\u5e8f\u5217":[2,56],"\u4e0d\u662f\u771f\u6b63\u7684layer":50,"\u4e0d\u662f\u901a\u8fc7\u4e00\u822c\u7684\u65b9\u5f0f\u6765\u5b9e\u73b0\u5bf9\u8f93\u51fa\u7684\u6fc0\u6d3b":50,"\u4e0d\u663e\u793a\u7684\u5199\u6bcf\u4e2a\u7c7b\u5177\u4f53\u5305\u542b\u4ec0\u4e48":33,"\u4e0d\u6ee1\u8db3\u94a9\u5b50\u7684":66,"\u4e0d\u7528mount\u7684\u65b9\u5f0f\u6765\u8bbf\u95ee\u6570\u636e":21,"\u4e0d\u7f13\u5b58\u4efb\u4f55\u6570\u636e":2,"\u4e0d\u80fd\u4fee\u6539op\u7684\u6210\u5458\u53d8\u91cf":68,"\u4e0d\u80fd\u592a\u968f\u610f":66,"\u4e0d\u80fd\u88ab\u63d0\u4ea4\u5230":66,"\u4e0d\u8fc7":61,"\u4e0d\u8fc7\u5b9e\u9645\u4e0a\u662f\u8fd0\u884c\u5728\u4e00\u4e2a":65,"\u4e0d\u8fdc":61,"\u4e0d\u9519":61,"\u4e0d\u9700\u8981\u4f9d\u8d56\u5176\u4ed6\u4efb\u4f55\u8f6f\u4ef6\u4e86":65,"\u4e0e":[68,80,85,94],"\u4e0e\u4e4b\u76f8\u5bf9\u7684\u662flocal":28,"\u4e0e\u529f\u80fd\u5206\u652f\u4e0d\u540c\u7684\u662f":42,"\u4e0e\u5355\u5c42rnn\u7684\u914d\u7f6e\u7c7b\u4f3c":61,"\u4e0e\u53ef\u80fd\u6709\u7684":42,"\u4e0e\u5728":92,"\u4e0e\u5f53\u524d\u7684\u8870\u51cf\u56e0\u5b50\u7684\u4e58\u79ef":51,"\u4e0e\u672c\u5730\u8bad\u7ec3\u76f8\u540c":73,"\u4e0e\u6b64\u4e0d\u540c\u7684\u662f":80,"\u4e0e\u7ffb\u8bd1":94,"\u4e0e\u8bad\u7ec3\u4e0d\u540c":93,"\u4e0e\u8bad\u7ec3\u6a21\u578b\u4e0d\u540c\u7684\u662f":94,"\u4e0e\u8fd9\u4e2a\u8bad\u7ec3\u6570\u636e\u4ea4\u4e92\u7684layer":49,"\u4e0eimdb\u7f51\u7ad9\u63d0\u4f9b\u7684\u4e00\u81f4":90,"\u4e0ejob":80,"\u4e0eoperator\u524d\u5411\u8ba1\u7b97\u7684\u8f93\u51fa\u8fdb\u884c\u5bf9\u6bd4":68,"\u4e0eoperator\u6ce8\u518c\u65f6\u6ce8\u518c\u7684\u7c7b\u578b\u4e00\u81f4":68,"\u4e0etime":90,"\u4e14":61,"\u4e14\u4e0d\u6392\u9664commit\u4e4b\u95f4\u7684\u4fee\u6539\u5b58\u5728\u76f8\u4e92\u8986\u76d6\u7684\u60c5\u51b5":66,"\u4e14\u589e\u52a0\u4e00\u4e2a\u7b2c\u4e09\u65b9\u8bed\u8a00":33,"\u4e14\u5e8f\u5217\u7684\u6bcf\u4e00\u4e2a\u5143\u7d20\u8fd8\u662f\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217":[2,56],"\u4e14\u652f\u6301\u90e8\u7f72\u5230":78,"\u4e14\u6bcf\u4e2a\u53e5\u5b50\u8868\u793a\u4e3a\u5bf9\u5e94\u7684\u8bcd\u8868\u7d22\u5f15\u6570\u7ec4":61,"\u4e14\u8c03\u7528\u65f6\u4e0d\u80fd\u629b\u51fa\u5f02\u5e38\u6216\u51fa\u73b0\u8fd0\u884c\u65f6\u9519\u8bef":34,"\u4e14\u9ed8\u8ba4\u5728\u8bad\u7ec3\u96c6\u4e0a\u6784\u5efa\u5b57\u5178":93,"\u4e14c99\u652f\u6301bool\u7c7b\u578b\u548c\u5b9a\u957f\u6574\u6570":33,"\u4e14c99\u76f8\u5bf9\u4e8ec11\u4f7f\u7528\u66f4\u52a0\u5e7f\u6cdb":33,"\u4e24":61,"\u4e24\u4e2a\u5217\u8868\u6587\u4ef6":73,"\u4e24\u4e2a\u5b50\u76ee\u5f55\u4e0b":70,"\u4e24\u4e2a\u5d4c\u5957\u7684":63,"\u4e24\u4e2a\u64cd\u4f5c":72,"\u4e24\u4e2a\u6587\u4ef6\u5939":86,"\u4e24\u4e2a\u8f93\u5165\u7279\u5f81\u5728\u8fd9\u4e2a\u6d41\u7a0b\u4e2d\u8d77\u7740\u81f3\u5173\u91cd\u8981\u7684\u4f5c\u7528":92,"\u4e24\u4e2a\u8f93\u5165\u7684\u5b50\u5e8f\u5217\u957f\u5ea6\u4e5f\u5e76\u4e0d\u76f8\u540c":61,"\u4e24\u4e2a\u90e8\u5206":70,"\u4e24\u79cd\u65b9\u6cd5\u7684\u533a\u522b":49,"\u4e24\u79cd\u7c7b\u522b":89,"\u4e24\u8005\u5747\u4e3a\u7eaf\u6587\u672c\u6587\u4ef6":1,"\u4e24\u8005\u90fd\u662f\u5bf9\u68af\u5ea6\u7684\u622a\u65ad":49,"\u4e25\u683c\u7684\u547d\u540d\u89c4\u8303pep":42,"\u4e2a":89,"\u4e2a\u5185\u5b58\u6c60\u5b9e\u9645\u4e0a\u51b3\u5b9a\u4e86shuffle\u7684\u7c92\u5ea6":49,"\u4e2a\u5355\u8bcd":94,"\u4e2a\u6027\u5316\u63a8\u8350":[42,88],"\u4e2a\u6279\u6b21\u540e\u6253\u5370\u4e00\u4e2a":91,"\u4e2a\u6279\u6b21\u7684\u53c2\u6570\u5e73\u5747\u503c\u8fdb\u884c\u6d4b\u8bd5":75,"\u4e2a\u6a21\u578b\u6d4b\u8bd5\u6570\u636e":75,"\u4e2d":[33,34,49,67,68,69,80,86,89,91,92,93],"\u4e2d\u4e0d\u8981\u6dfb\u52a0\u5927\u6587\u4ef6\u7b49":66,"\u4e2d\u4ecb\u7ecd\u7684\u65b9\u6cd5":85,"\u4e2d\u4efb\u610f\u7b2ci\u884c\u7684\u53e5\u5b50\u4e4b\u95f4\u90fd\u5fc5\u987b\u6709\u7740\u4e00\u4e00\u5bf9\u5e94\u7684\u5173\u7cfb":94,"\u4e2d\u4efb\u610f\u7b2ci\u884c\u7684\u53e5\u5b50\u4e4b\u95f4\u90fd\u6709\u7740\u4e00\u4e00\u5bf9\u5e94\u7684\u5173\u7cfb":94,"\u4e2d\u5143\u7d20\u7684\u4e2a\u6570\u7b49\u4e8e\u7f51\u7edc\u4e2d\u8f93\u51fa\u5c42\u7684\u4e2a\u6570":49,"\u4e2d\u5173\u4e8e\u65f6\u95f4\u9012\u5f52\u795e\u7ecf\u7f51\u7edc\u7684\u4ecb\u7ecd":61,"\u4e2d\u5199\u5165json\u5185\u5bb9":20,"\u4e2d\u5305\u542b\u4e00\u4e2aandroid\u7248\u672c\u7684\u5e93":58,"\u4e2d\u5305\u542b\u4e00\u4e2araspberri":59,"\u4e2d\u5305\u542b\u4e86\u8bad\u7ec3\u6a21\u578b\u7684\u57fa\u672c\u547d\u4ee4":89,"\u4e2d\u5305\u542b\u5982\u4e0b\u8868\u6240\u793a\u76843\u4e2a\u6587\u4ef6\u5939":94,"\u4e2d\u5305\u542bc":[58,59],"\u4e2d\u5355\u5143\u6d4b\u8bd5\u7684\u4e00\u90e8\u5206":66,"\u4e2d\u5355\u5143\u6d4b\u8bd5\u80fd\u987a\u5229\u901a\u8fc7":66,"\u4e2d\u5b89\u88c5":73,"\u4e2d\u5b8c\u5168\u4e00\u81f4":33,"\u4e2d\u5b8c\u6210":93,"\u4e2d\u5b9a\u4e49":64,"\u4e2d\u5b9a\u4e49\u4f7f\u7528\u54ea\u79cddataprovid":1,"\u4e2d\u5b9a\u4e49\u548c\u4f7f\u7528":63,"\u4e2d\u5b9e\u73b0\u7684\u7ed3\u6784\u4f53":34,"\u4e2d\u5bfc\u51fa\u9884\u5b9a\u4e49\u7684\u7f51\u7edc":93,"\u4e2d\u5c55\u793a\u4e86\u5982\u4f55\u4f7f\u7528python\u6765\u63d0\u53d6\u7279\u5f81":87,"\u4e2d\u6307\u5b9a":75,"\u4e2d\u6307\u5b9a\u7684\u540d\u5b57":77,"\u4e2d\u6307\u5b9a\u7684\u5c42\u987a\u5e8f\u4e00\u81f4":87,"\u4e2d\u63d0\u4f9b\u4e86\u4e00\u4e9b\u5168\u5c40\u51fd\u6570\u7528\u6765\u5b9e\u73b0paddl":69,"\u4e2d\u63d0\u51fa\u7684resnet\u7f51\u7edc\u7ed3\u6784\u57282015\u5e74imagenet\u5927\u89c4\u6a21\u89c6\u89c9\u8bc6\u522b\u7ade\u8d5b":87,"\u4e2d\u641c\u7d22\u8fd9\u51e0\u4e2a\u5e93":53,"\u4e2d\u6587\u6587\u6863":70,"\u4e2d\u6587\u6587\u6863\u76ee\u5f55":70,"\u4e2d\u6587\u7ef4\u57fa\u767e\u79d1\u9875\u9762":61,"\u4e2d\u6709\u8bb8\u591a\u7684\u7279\u5f81":90,"\u4e2d\u6bcf\u4e2a\u5143\u7d20\u662f\u4e00\u4e2alayer\u7684\u8f93\u51fa\u7ed3\u679c\u77e9\u9635":49,"\u4e2d\u6bcf\u4e2apod\u7684ip\u5730\u5740":80,"\u4e2d\u6bcf\u5c42\u7684\u6570\u503c\u7edf\u8ba1":75,"\u4e2d\u6dfb\u52a0\u4e24\u4e2a\u8f93\u5165":68,"\u4e2d\u7684":[69,87],"\u4e2d\u7684\u4e00\u884c":[2,66],"\u4e2d\u7684\u5185\u5bb9":92,"\u4e2d\u7684\u5bf9\u5e94\u5206\u652f\u5373\u53ef":66,"\u4e2d\u7684\u63a5\u53e3":91,"\u4e2d\u7684\u6570\u636e":87,"\u4e2d\u7684\u6570\u636e\u662f\u5426\u4e3a\u5e8f\u5217\u6a21\u5f0f":91,"\u4e2d\u7684\u6570\u636e\u8fdb\u884c\u9884\u6d4b":87,"\u4e2d\u7684\u6570\u6910\u96c6\u7684\u7ed3\u6784\u5982\u4e0b":93,"\u4e2d\u7684\u6bcf\u4e00\u884c\u547d\u4ee4":91,"\u4e2d\u7684\u751f\u6210\u7ed3\u679c\u5982\u4e0b\u6240\u793a":94,"\u4e2d\u7684\u7528\u6237\u8bc1\u4e66":78,"\u4e2d\u7684\u7b2ci\u884c":94,"\u4e2d\u7684\u8bf4\u660e":2,"\u4e2d\u7684\u8fd9\u4e9b\u6570\u636e\u6587\u4ef6":90,"\u4e2d\u770b\u5230\u4e0b\u9762\u7684\u6587\u4ef6":93,"\u4e2d\u83b7\u53d6":80,"\u4e2d\u8ba4\u771f\u8bbe\u7f6e":73,"\u4e2d\u8bbe\u7f6e":73,"\u4e2d\u8bbe\u7f6e\u7684\u6240\u6709\u8282\u70b9":73,"\u4e2d\u8be6\u7ec6\u4ecb\u7ecd":67,"\u4e2d\u8bfb\u53d6":2,"\u4e2d\u8c03\u7528":68,"\u4e2d\u8fd0\u884c\u4efb\u52a1\u7684\u89d2\u5ea6":21,"\u4e2d\u914d\u7f6e\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":64,"\u4e2d\u914d\u7f6e\u7684\u6548\u679c\u4e00\u81f4":2,"\u4e34\u65f6\u53d8\u91cf\u7b49\u7b49":49,"\u4e3a":[2,58,59,64,68],"\u4e3a0":2,"\u4e3a\u4e86\u4f7f":68,"\u4e3a\u4e86\u4f7f\u7528\u63d0\u524d\u7f16\u5199\u7684\u811a\u672c":93,"\u4e3a\u4e86\u4f7f\u8bc4\u5ba1\u4eba\u5728\u8bc4\u5ba1\u4ee3\u7801\u65f6\u66f4\u597d\u5730\u4e13\u6ce8\u4e8e\u4ee3\u7801\u672c\u8eab":66,"\u4e3a\u4e86\u4fdd\u8bc1\u6548\u7387":67,"\u4e3a\u4e86\u4fdd\u8bc1gpu\u9a71\u52a8\u80fd\u591f\u5728\u955c\u50cf\u91cc\u9762\u6b63\u5e38\u8fd0\u884c":54,"\u4e3a\u4e86\u5145\u5206\u7684\u968f\u673a\u6253\u4e71\u8bad\u7ec3\u96c6":93,"\u4e3a\u4e86\u5b8c\u6210\u5206\u5e03\u5f0f\u673a\u5668\u5b66\u4e60\u8bad\u7ec3\u4efb\u52a1":78,"\u4e3a\u4e86\u5c01\u88c5\u80fd\u591f\u6b63\u786e\u5de5\u4f5c":67,"\u4e3a\u4e86\u5e94\u5bf9\u4ee5\u4e0a\u7684\u95ee\u9898":28,"\u4e3a\u4e86\u5f00\u53d1paddlepaddl":65,"\u4e3a\u4e86\u63cf\u8ff0\u65b9\u4fbf":63,"\u4e3a\u4e86\u65b9\u4fbf\u5728\u56fd\u5185\u7684\u5f00\u53d1\u8005\u4e0b\u8f7ddocker\u955c\u50cf":54,"\u4e3a\u4e86\u65b9\u4fbf\u5927\u5bb6":66,"\u4e3a\u4e86\u65b9\u4fbf\u8d77\u89c1":73,"\u4e3a\u4e86\u66b4\u9732\u7684\u63a5\u53e3\u5c3d\u91cf\u7b80\u5355":34,"\u4e3a\u4e86\u6ee1\u8db3\u8bad\u7ec3":73,"\u4e3a\u4e86\u7b80\u5316cmake\u914d\u7f6e":[58,59],"\u4e3a\u4e86\u8fbe\u5230\u6027\u80fd\u6700\u4f18":72,"\u4e3a\u4e86\u8fd8\u539f":52,"\u4e3a\u4ec0\u4e48\u7528":65,"\u4e3a\u4f8b":[50,68,89],"\u4e3a\u4f8b\u521b\u5efa\u5206\u5e03\u5f0f\u7684\u5355\u8fdb\u7a0b\u8bad\u7ec3":73,"\u4e3a\u4f8b\u6765\u4ecb\u7ecd\u5982\u4f55\u5199\u5e26kernel\u7684oper":68,"\u4e3a\u4f8b\u8fdb\u884c\u9884\u6d4b":89,"\u4e3a\u53c2\u6570\u77e9\u9635\u7684\u5bbd\u5ea6":51,"\u4e3a\u60a8\u505a\u6027\u80fd\u8c03\u4f18\u63d0\u4f9b\u4e86\u65b9\u5411":72,"\u4e3a\u60f3\u4fee\u6b63\u8bcd\u5411\u91cf\u6a21\u578b\u7684\u7528\u6237\u63d0\u4f9b\u4e86\u5c06\u6587\u672c\u8bcd\u5411\u91cf\u6a21\u578b\u8f6c\u6362\u4e3a\u4e8c\u8fdb\u5236\u6a21\u578b\u7684\u547d\u4ee4":85,"\u4e3a\u65b9\u4fbf\u4f5c\u4e1a\u542f\u52a8\u63d0\u4f9b\u4e86\u4e24\u4e2a\u72ec\u7279\u7684\u547d\u4ee4\u9009\u9879":73,"\u4e3a\u6b64":79,"\u4e3a\u6bcf\u4e2aop\u521b\u5efa\u5355\u72ec\u7684":68,"\u4e3a\u8f93\u51fa\u5206\u914d\u5185\u5b58":67,"\u4e3a\u96c6\u7fa4\u4f5c\u4e1a\u8bbe\u7f6e\u989d\u5916\u7684":73,"\u4e3aconst\u51fd\u6570":68,"\u4e3ajson\u6216yaml\u683c\u5f0f":91,"\u4e3aoutput_\u7533\u8bf7\u5185\u5b58":67,"\u4e3b\u8981\u4e3a\u5f00\u53d1\u8005\u4f7f\u7528":75,"\u4e3b\u8981\u529f\u80fd\u5305\u62ec":28,"\u4e3b\u8981\u5305\u62ec\u4ee5\u4e0b\u4e94\u4e2a\u6b65\u9aa4":4,"\u4e3b\u8981\u5305\u62ec\u56db\u79cd\u7c7b\u578b":56,"\u4e3b\u8981\u539f\u56e0":61,"\u4e3b\u8981\u539f\u56e0\u5305\u62ec\u4e24\u4e2a\u65b9\u9762":49,"\u4e3b\u8981\u539f\u56e0\u662f\u589e\u52a0\u4e86\u521d\u59cb\u5316\u673a\u5236":2,"\u4e3b\u8981\u6765\u81ea\u5317\u7f8e\u6d32":86,"\u4e3b\u8981\u7528\u4e8epython":68,"\u4e3b\u8981\u804c\u8d23\u5728\u4e8e\u5c06\u8bad\u7ec3\u6570\u636e\u4f20\u5165\u5185\u5b58\u6216\u8005\u663e\u5b58":89,"\u4e3e\u4e00\u4e2a\u4f8b\u5b50":51,"\u4e3e\u4f8b":49,"\u4e3e\u4f8b\u8bf4\u660e":61,"\u4e4b\u524d":66,"\u4e4b\u524d\u914d\u7f6e\u6587\u4ef6\u4e2d":89,"\u4e4b\u540e":[52,56,67],"\u4e4b\u540e\u4f60\u4f1a\u5f97\u5230\u8bad\u7ec3":73,"\u4e4b\u540e\u4f7f\u7528":67,"\u4e4b\u540e\u4f7f\u7528\u77e9\u9635\u8fd0\u7b97\u51fd\u6570\u6765\u8ba1\u7b97":67,"\u4e4b\u540e\u521d\u59cb\u5316\u6240\u6709\u7684\u6743\u91cd\u77e9\u9635":67,"\u4e4b\u540e\u5b9a\u4e49\u7684":86,"\u4e4b\u5916\u7684\u6240\u6709\u5934\u6587\u4ef6":34,"\u4e4b\u7c7b\u7684\u7a0b\u5e8f\u6765\u7f16\u8bd1\u6e90\u7801":65,"\u4e4b\u95f4\u7684\u8ddd\u79bb":52,"\u4e4b\u95f4\u7684\u8fd0\u7b97\u662f\u72ec\u7acb\u7684":63,"\u4e58\u4e0a\u8f93\u51fa\u7684\u68af\u5ea6":67,"\u4e58\u9664\u7b49\u65f6\u5019":49,"\u4e5d\u4e2a":92,"\u4e5f":61,"\u4e5f\u4e0d\u4f7f\u7528\u5176\u4ed6\u52a8\u6001\u5e93":33,"\u4e5f\u4e0d\u5b58\u5728\u4e00\u4e2asubseq\u76f4\u63a5\u751f\u6210\u4e0b\u4e00\u4e2asubseq\u7684\u60c5\u51b5":63,"\u4e5f\u4e0d\u5e94\u8be5\u62a5\u9519":34,"\u4e5f\u4e0d\u751f\u6210":34,"\u4e5f\u4e0d\u80fd\u63a5\u6536\u5e8f\u5217\u6570\u636e\u4f5c\u4e3a\u8f93\u5165":50,"\u4e5f\u4f1a\u5360\u7528\u78c1\u76d8":65,"\u4e5f\u53ef\u4ee5\u4f7f\u7528":66,"\u4e5f\u53ef\u4ee5\u53bb\u6389\u8fd9\u4e9b\u8bc1\u4e66\u7684\u914d\u7f6e":78,"\u4e5f\u53ef\u4ee5\u5728\u5f00\u53d1\u955c\u50cf\u4e2d\u542f\u52a8\u4e00\u4e2asshd\u670d\u52a1":54,"\u4e5f\u53ef\u4ee5\u662f\u4e00\u4e2a\u8bcd\u8bed":63,"\u4e5f\u53ef\u4ee5\u8bf4\u662f\u67d0\u4e9b\u7279\u5b9a\u6307\u4ee4\u7684\u4f7f\u7528\u60c5\u51b5":72,"\u4e5f\u53ef\u4ee5\u901a\u8fc7\u4fee\u6539":80,"\u4e5f\u53ef\u4ee5\u901a\u8fc7saving_period_by_batches\u8bbe\u7f6e\u6bcf\u9694\u591a\u5c11batch\u4fdd\u5b58\u4e00\u6b21\u6a21\u578b":89,"\u4e5f\u53ef\u4ee5\u914d\u7f6e\u4e0d\u540c\u7684\u91cd\u8bd5\u673a\u5236":78,"\u4e5f\u53ef\u5199\u6210":68,"\u4e5f\u53ef\u901a\u8fc7\u4ee5\u4e0b\u547d\u4ee4\u83b7\u53d6":[58,59],"\u4e5f\u5c31\u662f":66,"\u4e5f\u5c31\u662f\u5c06\u8bcd\u5411\u91cf\u6a21\u578b\u8fdb\u4e00\u6b65\u6f14\u5316\u4e3a\u4e09\u4e2a\u65b0\u6b65\u9aa4":89,"\u4e5f\u5c31\u662f\u672c\u5730\u7684\u6e90\u7801\u6811\u6839\u76ee\u5f55\u91cc\u7684":65,"\u4e5f\u5c31\u662f\u81ea\u5df1\u7528\u6237\u540d\u4e0b\u7684":66,"\u4e5f\u5c31\u662f\u8bf4":[75,77,85],"\u4e5f\u5f97\u5230\u4e00\u4e2a\u7528\u6237\u7279\u5f81":91,"\u4e5f\u63cf\u8ff0\u4e86\u5bb9\u5668\u9700\u8981\u4f7f\u7528\u7684\u5b58\u50a8\u5377\u6302\u8f7d\u7684\u60c5\u51b5":80,"\u4e5f\u652f\u6301cpu\u7684\u6027\u80fd\u5206\u6790":72,"\u4e5f\u662f\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217":61,"\u4e5f\u662f\u5bb9\u5668\u4e0enode\u4e4b\u95f4\u5171\u4eab\u6587\u4ef6\u7684\u65b9\u5f0f":78,"\u4e5f\u662fdecoder\u5faa\u73af\u5c55\u5f00\u7684\u4f9d\u636e":63,"\u4e5f\u662fpaddlepaddle\u6240\u80fd\u591f\u4fdd\u8bc1\u7684shuffle\u7c92\u5ea6":2,"\u4e5f\u6ca1\u7528":46,"\u4e5f\u79f0\u4e3arnn\u6a21\u578b":89,"\u4e5f\u8bb8\u662f\u56e0\u4e3a\u9700\u8981\u5b89\u88c5":86,"\u4e5f\u9700\u8981\u4e24\u6b21\u968f\u673a\u9009\u62e9\u5230\u76f8\u540cgenerator\u7684\u65f6\u5019":2,"\u4e66\u5199":33,"\u4e7e":61,"\u4e86":[61,65],"\u4e86\u89e3\u60a8\u7684\u786c\u4ef6":72,"\u4e86\u89e3\u66f4\u591a\u7ec6\u8282":64,"\u4e86\u89e3\u66f4\u591a\u8be6\u7ec6\u4fe1\u606f":64,"\u4e8c\u7ea7\u76ee\u5f55":[93,94],"\u4e8c\u7ef4\u77e9\u9635":87,"\u4e8c\u8005\u8bed\u610f\u4e0a\u5b8c\u5168\u4e00\u81f4":61,"\u4e8c\u8fdb\u5236":85,"\u4e92\u76f8\u901a\u4fe1":78,"\u4e92\u8054\u7f51\u7535\u5f71\u6570\u636e\u5e93":93,"\u4e94\u661f\u7ea7":61,"\u4e9a\u9a6c\u900a":93,"\u4ea4\u53c9\u7f16\u8bd1\u5de5\u5177\u94fe":59,"\u4ea4\u53c9\u7f16\u8bd1\u5de5\u5177\u94fe\u4e3a":58,"\u4ea4\u53c9\u7f16\u8bd1android\u7248\u672c\u7684paddlepaddle\u5e93\u65f6":58,"\u4ea4\u53c9\u7f16\u8bd1raspberri":59,"\u4ea4\u7531cmake\u7cfb\u7edf\u672c\u8eab\u6765\u5904\u7406":58,"\u4ea4\u901a":61,"\u4ea4\u901a\u4fbf\u5229":61,"\u4ea6\u53ef\u4ee5\u901a\u8fc7\u624b\u52a8\u8bbe\u7f6e":[58,59],"\u4eab\u53d7\u60a8\u7684\u65c5\u7a0b":54,"\u4eba\u8138\u8bc6\u522b":21,"\u4ec0\u4e48":91,"\u4ec0\u4e48\u662f":65,"\u4ec5\u4ec5\u4f7f\u7528":33,"\u4ec5\u4ec5\u662f\u4e00\u4e9b\u5173\u952e\u8bcd":93,"\u4ec5\u4ec5\u662f\u4e24\u4e2a\u5168\u8fde\u63a5\u5c42":91,"\u4ec5\u4ec5\u662f\u7b80\u5355\u7684\u5d4c\u5165":91,"\u4ec5\u5305\u542b\u8bad\u7ec3\u548c\u6d4b\u8bd5\u6570\u6910\u96c6":93,"\u4ec5\u5728\u8fdc\u7a0b\u7a00\u758f\u8bad\u7ec3\u65f6\u6709\u6548":67,"\u4ec5\u5bf9\u7a00\u758f\u6570\u636e\u6709\u6548":67,"\u4ec5\u9700\u8981\u77e5\u9053\u5982\u4f55\u4ece":2,"\u4ecb\u7ecd\u4e86\u4e00\u79cd\u901a\u8fc7ssh\u8fdc\u7a0b\u5206\u53d1\u4efb\u52a1":80,"\u4ecb\u7ecd\u4ea4\u53c9\u7f16\u8bd1android\u5e73\u53f0\u4e0a\u9002\u7528\u7684paddlepaddle\u5e93\u7684\u65b9\u6cd5\u548c\u6b65\u9aa4":58,"\u4ecb\u7ecd\u4ea4\u53c9\u7f16\u8bd1raspberri":59,"\u4ecb\u7ecd\u5206\u5e03\u5f0f\u8bad\u7ec3\u4e4b\u524d":78,"\u4ecb\u7ecdpaddlepaddle\u7684\u57fa\u672c\u4f7f\u7528\u65b9\u6cd5":89,"\u4ece":[42,47,72,92],"\u4ece0\u5230num":75,"\u4ece\u4e00\u4e2aword\u751f\u6210\u4e0b\u4e00\u4e2aword":63,"\u4ece\u5185\u6838\u51fd\u6570\u7684\u89d2\u5ea6":72,"\u4ece\u56fe\u4e2d\u53ef\u4ee5\u770b\u5230":52,"\u4ece\u5916\u90e8\u7f51\u7ad9\u4e0a\u4e0b\u8f7d\u7684\u539f\u59cb\u6570\u6910\u96c6":93,"\u4ece\u5927\u5230\u5c0f":94,"\u4ece\u6570\u636e\u63d0\u4f9b\u7a0b\u5e8f\u52a0\u8f7d\u5b9e\u4f8b":92,"\u4ece\u6a21\u578b\u6587\u4ef6\u5c06\u9884\u8bad\u7ec3\u53c2\u6570\u8f7d\u5165":51,"\u4ece\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u6765\u770b":61,"\u4ece\u6bcf\u4e2a\u5355\u8bcd\u5de6\u53f3\u4e24\u7aef\u5206\u522b\u83b7\u53d6k\u4e2a\u76f8\u90bb\u7684\u5355\u8bcd":89,"\u4ece\u6e90\u7801\u4ea4\u53c9\u7f16\u8bd1paddlepaddl":[58,59],"\u4ece\u78c1\u76d8\u6587\u4ef6\u4e2d\u52a0\u8f7duuid\u6587\u4ef6\u540d\u7684\u68c0\u67e5\u70b9\u5feb\u7167\u6587\u4ef6":20,"\u4ece\u7b2c0\u4e2a\u8bc4\u4f30\u5230\u5f53\u524d\u8bc4\u4f30\u4e2d":94,"\u4ece\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u7684\u5e73\u5747\u635f\u5931":93,"\u4ece\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u7684\u5e73\u5747cost":94,"\u4ece\u800c\u53ef\u4ee5\u505a\u4e00\u4e9b\u4e0e\u8ba1\u7b97\u91cd\u53e0\u7684\u5de5\u4f5c":67,"\u4ece\u800c\u53ef\u4ee5\u62df\u5408\u4efb\u610f\u7684\u51fd\u6570\u6765\u5b66\u4e60\u590d\u6742\u7684\u6570\u636e\u5173\u7cfb":52,"\u4ece\u800c\u5f15\u53d1\u5176\u4ed6\u8282\u70b9\u65e0\u6cd5\u8fde\u63a5\u5bfc\u81f4":47,"\u4ece\u800c\u751f\u6210\u591a\u4e2agener":2,"\u4ece\u800c\u80fd\u591f\u88abpaddlepaddl":89,"\u4ece\u800c\u9632\u6b62\u8fc7\u62df\u5408":1,"\u4ece\u8be5\u94fe\u63a5":94,"\u4ece\u8bed\u4e49\u4e0a\u770b":63,"\u4ece\u8f93\u5165\u6570\u636e\u4e0a\u770b":61,"\u4ece\u8f93\u51fa\u65e5\u5fd7\u53ef\u4ee5\u770b\u5230":52,"\u4ece\u9884\u8bad\u7ec3\u6a21\u578b\u4e2d":85,"\u4ececmake":58,"\u4eceetcd\u4e2d\u8bfb\u53d6\u8282\u70b9":20,"\u4ecestart":75,"\u4ecetest":94,"\u4ed3\u5e93\u7684\u8fdc\u7a0b\u4e3b\u673a":66,"\u4ed4\u7ec6\u89c2\u5bdf":87,"\u4ed6\u4e3b\u8981\u5305\u542b\u4e86\u5b9e\u9645\u66b4\u9732\u7684\u7c7b\u578b\u7ed3\u6784":34,"\u4ed6\u4eec\u5206\u522b\u662f":61,"\u4ed6\u4eec\u5728\u81ea\u5df1\u7684":65,"\u4ed6\u4eec\u5728paddle\u7684\u6587\u6863\u548capi\u4e2d\u662f\u4e00\u4e2a\u6982\u5ff5":61,"\u4ed6\u4eec\u63d0\u51fa\u6b8b\u5dee\u5b66\u4e60\u7684\u6846\u67b6\u6765\u7b80\u5316\u7f51\u7edc\u7684\u8bad\u7ec3":87,"\u4ed6\u662f\u5c06":34,"\u4ed6\u7684\u76ee\u6807\u662f\u4f7f\u7528c":33,"\u4ee3\u66ff":80,"\u4ee3\u7801":91,"\u4ee3\u7801\u4e2d9":61,"\u4ee3\u7801\u5982\u4e0b":[49,50,51,64],"\u4ee3\u7801\u6ce8\u91ca\u8bf7\u9075\u5b88":66,"\u4ee3\u7801\u751f\u6210\u7684\u7b26\u53f7\u53ef\u80fd\u4e0d\u4e00\u81f4":33,"\u4ee3\u7801\u793a\u4f8b\u5982\u4e0b":68,"\u4ee3\u8868\u5bbf\u4e3b\u673a\u76ee\u5f55":80,"\u4ee3\u8868\u7f16\u53f7":91,"\u4ee3\u8868\u8fd9\u4e2ashard\u7684\u6700\u5927index":21,"\u4ee3\u8868shard\u7684index":21,"\u4ee5":50,"\u4ee5\u4e0a":66,"\u4ee5\u4e0a\u4ee3\u7801\u7684reader\u8f93\u51fa\u7684data":21,"\u4ee5\u4e0a\u547d\u4ee4\u4f1a\u5728\u5f53\u524d\u76ee\u5f55\u4e0b\u751f\u6210100\u4e2a\u6587\u4ef6":21,"\u4ee5\u4e0a\u547d\u4ee4\u4f1a\u6309\u7167\u751f\u6210\u7684dockerfile\u628a\u751f\u6210\u7684\u7a0b\u5e8f\u62f7\u8d1d\u5230\u751f\u4ea7\u955c\u50cf\u4e2d\u5e76\u505a\u76f8\u5e94\u7684\u914d\u7f6e":54,"\u4ee5\u4e0a\u547d\u4ee4\u4f1a\u7f16\u8bd1paddlepaddl":54,"\u4ee5\u4e0a\u65b9\u6cd5\u5728gpu\u955c\u50cf\u91cc\u4e5f\u80fd\u7528":54,"\u4ee5\u4e0b":[21,91],"\u4ee5\u4e0b\u4ee3\u7801\u7247\u6bb5\u5b9a\u4e49":64,"\u4ee5\u4e0b\u547d\u4ee4\u542f\u52a8\u4e00\u4e2a":65,"\u4ee5\u4e0b\u6211\u4eec\u7ffb\u8bd1\u6570\u636e\u96c6\u7f51\u7ad9\u4e2dreadme\u6587\u4ef6\u7684\u63cf\u8ff0":90,"\u4ee5\u4e0b\u6307\u4ee4\u80fd\u68c0\u67e5linux\u7535\u8111\u662f\u5426\u652f\u6301avx":54,"\u4ee5\u4e0b\u6559\u7a0b\u5c06\u6307\u5bfc\u60a8\u63d0\u4ea4\u4ee3\u7801":66,"\u4ee5\u4e0b\u662f\u5bf9\u4e0a\u8ff0\u6570\u636e\u52a0\u8f7d\u7684\u89e3\u91ca":89,"\u4ee5\u4e0b\u6b65\u9aa4\u57fa\u4e8e":73,"\u4ee5\u4e0b\u793a\u8303\u5982\u4f55\u4f7f\u7528\u9884\u8bad\u7ec3\u7684\u4e2d\u6587\u5b57\u5178\u548c\u8bcd\u5411\u91cf\u8fdb\u884c\u77ed\u8bed\u6539\u5199":85,"\u4ee5\u4e0b\u9009\u9879\u5fc5\u987b\u5728":73,"\u4ee5\u4ea4\u4e92\u5bb9\u5668\u65b9\u5f0f\u8fd0\u884c\u5f00\u53d1\u955c\u50cf":54,"\u4ee5\u4f7f\u7528adam\u7b97\u6cd5\u4e3a\u4f8b":51,"\u4ee5\u4fbf\u6211\u4eec\u53ef\u4ee5\u628a\u66f4\u591a\u7684\u7cbe\u529b\u653e\u5230\u903b\u8f91\u672c\u8eab\u4e0a":28,"\u4ee5\u4fbf\u7528\u6237":73,"\u4ee5\u4fdd\u8bc1\u68af\u5ea6\u7684\u6b63\u786e\u8ba1\u7b97":67,"\u4ee5\u4fdd\u8bc1\u68af\u5ea6\u8ba1\u7b97\u7684\u6b63\u786e\u6027":67,"\u4ee5\u4fdd\u8bc1\u7f16\u8bd1\u9ad8\u6548":65,"\u4ee5\u5206\u7c7b\u6765\u81ea":93,"\u4ee5\u53ca":[65,67],"\u4ee5\u53ca\u4f7f\u7528\u5b50\u5e8f\u5217\u6765\u5b9a\u4e49\u5206\u7ea7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u67b6\u6784":64,"\u4ee5\u53ca\u5207\u6362\u673a\u5668\u65f6\u9700\u8981\u65b0\u5b89\u88c5\u7684\u8f9b\u82e6":65,"\u4ee5\u53ca\u53cc\u5c42\u5e8f\u5217":60,"\u4ee5\u53ca\u5728wmt":94,"\u4ee5\u53ca\u5982\u4f55\u5728\u5c42\u4e4b\u95f4\u8fdb\u884c\u8fde\u63a5":86,"\u4ee5\u53ca\u751f\u6210\u521b\u5efa\u751f\u4ea7\u955c\u50cf\u7684dockerfil":54,"\u4ee5\u53ca\u76ee\u6807\u673a\u7248openblas\u5e93":59,"\u4ee5\u53ca\u76f8\u5173\u7684\u5c5e\u6027\u53c2\u6570":68,"\u4ee5\u53ca\u8ba1\u7b97\u903b\u8f91\u5728\u5e8f\u5217\u4e0a\u7684\u5faa\u73af\u5c55\u5f00":63,"\u4ee5\u53ca\u8f93\u5165\u7684\u68af\u5ea6":67,"\u4ee5\u53caandroid":58,"\u4ee5\u53caandroid\u6240\u9700":58,"\u4ee5\u53canumpi":21,"\u4ee5\u53capaddle\u5982\u4f55\u5904\u7406\u591a\u79cd\u7c7b\u578b\u7684\u8f93\u5165":91,"\u4ee5\u53carelu":67,"\u4ee5\u63d0\u4f9b\u4e00\u4e9b\u9ed8\u8ba4\u7684\u7f16\u8bd1\u5668\u548c\u7f16\u8bd1\u53c2\u6570\u76f8\u5173\u914d\u7f6e":[58,59],"\u4ee5\u76f8\u5bf9\u8def\u5f84\u5f15\u7528":1,"\u4ee5\u786e\u4fdd\u6240\u6709\u7684\u7b2c\u4e09\u65b9\u4f9d\u8d56\u5e93\u548cpaddlepaddle\u4ee3\u7801\u90fd\u662f\u9488\u5bf9\u65b0\u7684cmake\u914d\u7f6e\u91cd\u65b0\u7f16\u8bd1\u7684":[58,59],"\u4ee5\u83b7\u5f97\u66f4\u597d\u7684\u7f51\u7edc\u6027\u80fd":73,"\u4ee5\u8f93\u51fa":49,"\u4ee5\u9017\u53f7":85,"\u4ee5\u9017\u53f7\u95f4\u9694":75,"\u4ee5eigentensor\u4e3a\u4f8b":69,"\u4ee5embedding\u5c42\u4e3a\u4f8b":51,"\u4ee5lstm\u4e3a\u4f8b":50,"\u4ef7\u683c":61,"\u4efb\u52a1":91,"\u4efb\u52a1\u6765\u7ec8\u6b62\u96c6\u7fa4\u4f5c\u4e1a":73,"\u4efb\u610f\u5c06\u4e00\u4e9b\u6570\u636e\u7ec4\u5408\u6210\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217":61,"\u4efb\u610f\u65f6\u523b\u53ea\u53ef\u80fd\u540c\u65f6\u6709\u4e00\u53f0\u670d\u52a1\u5668\u6545\u969c":20,"\u4f18\u5316":93,"\u4f18\u5316\u5668\u5219\u7528\u94fe\u5f0f\u6cd5\u5219\u6765\u5bf9\u6bcf\u4e2a\u53c2\u6570\u8ba1\u7b97\u635f\u5931\u51fd\u6570\u7684\u68af\u5ea6":67,"\u4f1a\u5148\u8fdb\u884c\u53c2\u6570\u7684\u521d\u59cb\u5316\u4e0e\u89e3\u6790":80,"\u4f1a\u5171\u4eab\u53c2\u6570":51,"\u4f1a\u5173\u8054\u53c2\u6570":50,"\u4f1a\u52a0\u8f7d\u4e0a\u4e00\u8f6e\u7684\u53c2\u6570":75,"\u4f1a\u53d8\u6210\u8bcd\u8868\u4e2d\u7684\u4f4d\u7f6e":61,"\u4f1a\u542f\u52a8pserver\u4e0etrainer\u8fdb\u7a0b":80,"\u4f1a\u5728\u5f53\u524d\u76ee\u5f55\u751f\u6210\u4e24\u4e2a\u5b50\u76ee\u5f55":70,"\u4f1a\u5bf9\u6bcf\u4e00\u4e2a\u6fc0\u6d3b\u6682\u5b58\u4e00\u4e9b\u6570\u636e":49,"\u4f1a\u5bf9\u8fd9\u7c7b\u8f93\u5165\u8fdb\u884c\u62c6\u89e3":63,"\u4f1a\u5bfc\u81f4\u4e0d\u540c\u7248\u672cpython\u5728\u4e00\u4e2a\u8fdb\u7a0b\u91cc\u7684bug":33,"\u4f1a\u5c06\u6bcf\u4e2a\u65f6\u95f4\u6b65\u7684\u8f93\u51fa\u62fc\u63a5":63,"\u4f1a\u5c06\u7b2c\u4e00\u4e2a":49,"\u4f1a\u6210\u4e3astep\u51fd\u6570\u7684\u8f93\u5165":63,"\u4f1a\u6254\u5230\u8fd9\u6761\u6570\u636e":2,"\u4f1a\u62a5\u5982\u4e0b\u7684\u9519\u8bef":49,"\u4f1a\u62a5\u9519":63,"\u4f1a\u6839\u636e\u547d\u4ee4\u884c\u53c2\u6570\u6307\u5b9a\u7684\u6d4b\u8bd5\u65b9\u5f0f":1,"\u4f1a\u6839\u636einput_types\u68c0\u67e5\u6570\u636e\u7684\u5408\u6cd5\u6027":2,"\u4f1a\u76f4\u63a5\u62a5\u9519\u9000\u51fa":33,"\u4f1a\u76f8\u5e94\u5730\u6539\u53d8\u8f93\u51fa\u7684\u5c3a\u5bf8":67,"\u4f1a\u81ea\u52a8\u5173\u95ed\u5bf9\u5e94\u7684issu":66,"\u4f1a\u81ea\u9002\u5e94\u5730\u4ece\u8fd9\u4e9b\u5411\u91cf\u4e2d\u9009\u62e9\u4e00\u4e2a\u5b50\u96c6\u51fa\u6765":94,"\u4f1a\u83b7\u53d6\u5f53\u524dnamespace\u4e0b\u7684\u6240\u6709pod":80,"\u4f1a\u88ab\u62c6\u89e3\u4e3a\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":63,"\u4f1a\u88ab\u62c6\u89e3\u4e3a\u975e\u5e8f\u5217":63,"\u4f1a\u88abpickle\u5e8f\u5217\u5316\u6210\u5b57\u7b26\u4e32":21,"\u4f1a\u9020\u6210\u90ae\u4ef6\u707e\u96be":66,"\u4f20\u5165":[2,21],"\u4f20\u5165\u4e0a\u4e00\u6b65\u89e3\u6790\u51fa\u6765\u7684\u6a21\u578b\u914d\u7f6e\u5c31\u53ef\u4ee5\u521b\u5efa\u4e00\u4e2a":4,"\u4f20\u5165\u9884\u6d4b\u6570\u636e":4,"\u4f20\u7ed9dataprovider\u7684\u67d0\u4e00\u4e2aargs\u8fc7\u5927":51,"\u4f20\u9012\u7ed9\u914d\u7f6e\u6587\u4ef6\u7684\u53c2\u6570":75,"\u4f46":34,"\u4f46\u4e0d\u66b4\u9732":34,"\u4f46\u4e0d\u7528\u4e8e\u8ba1\u7b97\u68af\u5ea6":67,"\u4f46\u4e0d\u9700\u8981\u63d0\u524d\u521b\u5efa":75,"\u4f46\u4e8e\u53cc\u5c42\u5e8f\u5217\u7684lstm\u6765\u8bf4":61,"\u4f46\u548c\u5355\u5c42rnn\u4e0d\u540c":61,"\u4f46\u5728\u8d77\u521d\u7684\u51e0\u8f6e\u8bad\u7ec3\u4e2d\u5b83\u4eec\u90fd\u5728\u5feb\u901f\u903c\u8fd1\u771f\u5b9e\u503c":52,"\u4f46\u5b50\u53e5\u542b\u6709\u7684\u8bcd\u8bed\u6570\u53ef\u4ee5\u4e0d\u76f8\u7b49":63,"\u4f46\u5c3d\u91cf\u8bf7\u4fdd\u6301\u7f16\u8bd1\u548c\u8fd0\u884c\u4f7f\u7528\u7684cudnn\u662f\u540c\u4e00\u4e2a\u7248\u672c":53,"\u4f46\u5e76\u6ca1\u6709\u7ecf\u8fc7\u56de\u5f52\u6d4b\u8bd5":42,"\u4f46\u5e8f\u5217\u8f93\u51fa\u65f6":61,"\u4f46\u5f53\u8c03\u7528\u8fc7\u4e00\u6b21\u540e":2,"\u4f46\u622a\u65ad\u65f6\u673a\u4e0d\u540c":49,"\u4f46\u6240\u6709fork\u7684\u7248\u672c\u5e93\u7684\u6240\u6709\u5206\u652f\u90fd\u76f8\u5f53\u4e8e\u7279\u6027\u5206\u652f":42,"\u4f46\u662f":[49,61],"\u4f46\u662f2008\u5e74\u4e4b\u524d\u751f\u4ea7\u7684\u65e7\u7535\u8111\u4e0d\u652f\u6301avx":54,"\u4f46\u662f\u4e5f\u6ca1\u6709\u5fc5\u8981\u5220\u9664\u65e0\u7528\u7684\u6587\u4ef6":73,"\u4f46\u662f\u53c8\u8fc7\u4e8e\u7410\u788e":34,"\u4f46\u662f\u5927\u90e8\u5206\u53c2\u6570\u662f\u4e3a\u5f00\u53d1\u8005\u63d0\u4f9b\u7684":74,"\u4f46\u662f\u5b50\u5e8f\u5217\u7684\u6570\u76ee\u5fc5\u987b\u4e00\u6837":61,"\u4f46\u662f\u6211\u4eec\u5e76\u4e0d\u63a8\u8350\u8fd9\u79cd\u65b9\u6cd5":54,"\u4f46\u662f\u652f\u6301avx\u6307\u4ee4\u96c6":66,"\u4f46\u662f\u65b9\u4fbf\u8c03\u8bd5\u548c\u6d4bbenchmark":53,"\u4f46\u662f\u662f\u8fd0\u884c\u5728\u72ec\u7acb\u7684\u8fdb\u7a0b\u7a7a\u95f4":54,"\u4f46\u662f\u6bcf\u4e2a\u6837\u672c\u4ec5\u5305\u542b\u51e0\u4e2a\u8bcd":77,"\u4f46\u662f\u7a81\u7136\u6709\u4e00\u4e2a10000\u957f\u7684\u5e8f\u5217":49,"\u4f46\u662f\u865a\u62df\u7684\u4e0d\u4ec5\u4ec5\u662f":65,"\u4f46\u662f\u89e3\u91ca\u6027\u8bed\u8a00":33,"\u4f46\u662f\u8fd9\u79cd\u65b9\u6cd5\u5728\u6bcf\u5c42\u53ea\u4fdd\u5b58\u9884\u8bbe\u6570\u91cf\u7684\u6700\u4f18\u72b6\u6001":94,"\u4f46\u662f\u9690\u85cf\u5c42\u4e2d\u7684\u6bcf\u4e2a\u666e\u901a\u8282\u70b9\u88ab\u4e00\u4e2a\u8bb0\u5fc6\u5355\u5143\u66ff\u6362":93,"\u4f46\u662fbatch":49,"\u4f46\u6709\u503c\u7684\u5730\u65b9\u5fc5\u987b\u4e3a1":[2,56],"\u4f46\u6709\u503c\u7684\u90e8\u5206\u53ef\u4ee5\u662f\u4efb\u4f55\u6d6e\u70b9\u6570":[2,56],"\u4f46\u8fd9\u4e2a\u5173\u7cfb\u53ef\u80fd\u4e0d\u6b63\u786e":2,"\u4f46\u9700\u6ce8\u610f\u53cd\u5411op\u6ca1\u6709":68,"\u4f46eigen":69,"\u4f4d\u7f6e":61,"\u4f4f":61,"\u4f53\u88c1\u5b57\u5178":91,"\u4f53\u88c1\u5b57\u6bb5":91,"\u4f59\u5f26\u76f8\u4f3c\u5ea6\u56de\u5f52":91,"\u4f59\u5f26\u76f8\u4f3c\u5ea6\u5c42":91,"\u4f5c\u4e3a\u4e0b\u4e00\u4e2a\u5b50\u53e5memory\u7684\u521d\u59cb\u72b6\u6001":61,"\u4f5c\u4e3a\u4f8b\u5b50\u6f14\u793a\u5982\u4f55\u914d\u7f6e\u590d\u6742\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u6a21\u578b":64,"\u4f5c\u4e3a\u53c2\u6570\u5c5e\u6027":68,"\u4f5c\u4e3a\u53c2\u6570\u7684id":51,"\u4f5c\u4e3a\u5b58\u50a8\u7cfb\u7edf":21,"\u4f5c\u4e3a\u5f53\u524d\u65f6\u523b\u8f93\u5165":63,"\u4f5c\u4e3a\u793a\u4f8b\u6570\u636e":90,"\u4f5c\u4e3a\u7c7b\u53e5\u67c4":33,"\u4f5c\u4e3a\u7ebf\u6027\u56de\u5f52\u7684\u8f93\u5165":52,"\u4f5c\u4e3a\u8f93\u51fa":64,"\u4f5c\u4e3a\u96c6\u7fa4\u8bad\u7ec3\u7684\u5de5\u4f5c\u7a7a\u95f4":73,"\u4f5c\u4e3aboot_layer\u4f20\u7ed9\u4e0b\u4e00\u4e2a\u5b50\u53e5\u7684memori":61,"\u4f5c\u5bb6":90,"\u4f5c\u7528":60,"\u4f60\u4e5f\u53ef\u4ee5\u4f7f\u7528\u8fd9\u4e09\u4e2a\u503c":87,"\u4f60\u4e5f\u53ef\u4ee5\u5148\u8df3\u8fc7\u672c\u6587\u7684\u89e3\u91ca\u73af\u8282":89,"\u4f60\u4e5f\u53ef\u4ee5\u7b80\u5355\u7684\u8fd0\u884c\u4ee5\u4e0b\u7684\u547d\u4ee4":85,"\u4f60\u4e5f\u53ef\u4ee5\u901a\u8fc7\u5728\u547d\u4ee4\u884c\u53c2\u6570\u4e2d\u589e\u52a0\u4e00\u4e2a\u53c2\u6570\u5982":87,"\u4f60\u4e5f\u8bb8\u53ef\u4ee5\u5c1d\u8bd5\u66f4\u8001\u7684\u65b9\u6cd5":54,"\u4f60\u53ea\u9700\u5b8c\u6210":73,"\u4f60\u53ea\u9700\u8981\u5728\u547d\u4ee4\u884c\u8f93\u5165\u4ee5\u4e0b\u547d\u4ee4":89,"\u4f60\u53ea\u9700\u8981\u6309\u7167\u5982\u4e0b\u65b9\u5f0f\u7ec4\u7ec7\u6570\u636e":94,"\u4f60\u53ef\u4ee5\u4f7f\u7528":87,"\u4f60\u53ef\u4ee5\u4f7f\u7528\u4e0b\u9762\u7684\u811a\u672c\u4e0b\u8f7d":93,"\u4f60\u53ef\u4ee5\u4f7f\u7528\u5b83\u6765\u8bbe\u7f6e":73,"\u4f60\u53ef\u4ee5\u4f7f\u7528\u672c\u5730\u8bad\u7ec3\u4e2d\u7684\u76f8\u540c\u6a21\u578b\u6587\u4ef6\u8fdb\u884c\u96c6\u7fa4\u8bad\u7ec3":73,"\u4f60\u53ef\u4ee5\u5728\u4efb\u4f55\u65f6\u5019\u7528":91,"\u4f60\u53ef\u4ee5\u5bf9\u6a21\u578b\u8fdb\u884c\u8bc4\u4f30":91,"\u4f60\u53ef\u4ee5\u5c06\u7f51\u7edc\u914d\u7f6e\u6210\u67d0\u4e9b\u5c42\u4f7f\u7528gpu\u8ba1\u7b97":77,"\u4f60\u53ef\u4ee5\u6267\u884c\u4e0a\u8ff0\u547d\u4ee4\u6765\u4e0b\u8f7d\u6240\u6709\u7684\u6a21\u578b\u548c\u5747\u503c\u6587\u4ef6":87,"\u4f60\u53ef\u4ee5\u901a\u8fc7\u6267\u884c\u4e0b\u9762\u7684\u547d\u4ee4\u6765\u5f97\u5230resnet\u7f51\u7edc\u7684\u7ed3\u6784\u53ef\u89c6\u5316\u56fe":87,"\u4f60\u53ef\u4ee5\u9884\u6d4b\u4efb\u4f55\u7528\u6237\u5bf9\u4e8e\u4efb\u4f55\u4e00\u90e8\u7535\u5f71\u7684\u8bc4\u4ef7":91,"\u4f60\u5c06\u4f1a\u770b\u5230\u4ee5\u4e0b\u7684\u6a21\u578b\u7ed3\u6784":85,"\u4f60\u5c06\u4f1a\u770b\u5230\u5982\u4e0b\u6d88\u606f":94,"\u4f60\u5c06\u4f1a\u770b\u5230\u5982\u4e0b\u7ed3\u679c":87,"\u4f60\u5c06\u4f1a\u770b\u5230\u7279\u5f81\u5b58\u50a8\u5728":87,"\u4f60\u5c06\u4f1a\u770b\u5230\u8fd9\u6837\u7684\u6d88\u606f":94,"\u4f60\u5c06\u5728\u76ee\u5f55":93,"\u4f60\u5c06\u770b\u5230\u5982\u4e0b\u7684\u4fe1\u606f":91,"\u4f60\u7684\u5de5\u4f5c\u7a7a\u95f4\u5e94\u5982\u4e0b\u6240\u793a":73,"\u4f60\u8fd8\u53ef\u4ee5\u5c06\u7528\u6237\u548c":73,"\u4f60\u9700\u8981\u4e00\u4e9b\u66f4\u590d\u6742\u7684\u5355\u5143\u6d4b\u8bd5\u6765\u4fdd\u8bc1\u4f60\u5b9e\u73b0\u7684\u7f51\u7edc\u5c42\u662f\u6b63\u786e\u7684":67,"\u4f60\u9700\u8981\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u6307\u5b9a\u8bbe\u5907\u7684id\u53f7":77,"\u4f60\u9700\u8981\u5728\u914d\u7f6ecmake\u65f6\u5c06":67,"\u4f60\u9700\u8981\u5b89\u88c5python\u7684\u7b2c\u4e09\u65b9\u5e93":91,"\u4f60\u9700\u8981\u628a\u8be5\u6587\u4ef6\u52a0\u5165":67,"\u4f7f\u5176\u8f6c\u53d8\u4e3a\u7ef4\u5ea6\u4e3ahidden_dim\u7684\u65b0\u5411\u91cf":89,"\u4f7f\u5f97":52,"\u4f7f\u5f97\u4e24\u4e2a\u5b57\u5178\u6709\u76f8\u540c\u7684\u4e0a\u4e0b\u6587":94,"\u4f7f\u5f97\u5355\u5143\u6d4b\u8bd5\u6709\u4e00\u4e2a\u5e72\u51c0\u7684\u73af\u5883":46,"\u4f7f\u5f97\u642d\u6a21\u578b\u65f6\u66f4\u65b9\u4fbf":67,"\u4f7f\u5f97\u6700\u7ec8\u5f97\u5230\u7684\u6a21\u578b\u51e0\u4e4e\u4e0e\u771f\u5b9e\u6a21\u578b\u4e00\u81f4":52,"\u4f7f\u7528":[34,42,49,50,51,54,61,63,64,67,72,75,89,92,93],"\u4f7f\u75280\u53f7\u548c1\u53f7gpu\u8ba1\u7b97fc2\u5c42":77,"\u4f7f\u75280\u53f7gpu\u8ba1\u7b97fc2\u5c42":77,"\u4f7f\u752810\u4e2a\u88c1\u526a\u56fe\u50cf\u5757":87,"\u4f7f\u75281\u53f7gpu\u8ba1\u7b97fc3\u5c42":77,"\u4f7f\u75282\u53f7\u548c3\u53f7gpu\u8ba1\u7b97fc3\u5c42":77,"\u4f7f\u7528\u4e00\u4e2a\u5c3a\u5ea6\u4e3a":67,"\u4f7f\u7528\u4e00\u4e2a\u8bcd\u524d\u4e24\u4e2a\u8bcd\u548c\u540e\u4e24\u4e2a\u8bcd":49,"\u4f7f\u7528\u4e0a\u6587\u521b\u5efa\u7684yaml\u6587\u4ef6\u521b\u5efakubernet":79,"\u4f7f\u7528\u4e0b\u9762\u547d\u4ee4":21,"\u4f7f\u7528\u4e86\u540c\u6837\u7684parameter\u548cbia":51,"\u4f7f\u7528\u4e86\u57fa\u4e8e\u53e5\u6cd5\u7ed3\u6784\u7684\u9884\u5b9a\u4e49\u7279\u5f81\u6a21\u677f":92,"\u4f7f\u7528\u4ee5\u4e0b\u547d\u4ee4\u8fdb\u884c\u6a21\u578b\u8bad\u7ec3":85,"\u4f7f\u7528\u5177\u6709softmax\u6fc0\u6d3b\u7684\u5168\u8fde\u63a5\u524d\u9988\u5c42\u6765\u6267\u884c\u5206\u7c7b\u4efb\u52a1":93,"\u4f7f\u7528\u52a8\u6001\u5e93":33,"\u4f7f\u7528\u540c\u6837\u7684\u8bad\u7ec3\u6570\u636eblock":20,"\u4f7f\u7528\u591a\u5757\u663e\u5361\u8bad\u7ec3":49,"\u4f7f\u7528\u591a\u7ebf\u7a0b\u8bad\u7ec3":49,"\u4f7f\u7528\u5982\u4e0b\u53c2\u6570":86,"\u4f7f\u7528\u5982\u4e0b\u547d\u4ee4":85,"\u4f7f\u7528\u5b66\u4e60\u5b8c\u6210\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u6a21\u578b\u751f\u6210\u5e8f\u5217":64,"\u4f7f\u7528\u5bb9\u5668\u65b9\u5f0f\u8fd0\u884c\u8bad\u7ec3\u4efb\u52a1\u7684kubernet":80,"\u4f7f\u7528\u6211\u4eec\u4e4b\u524d\u6784\u9020\u7684\u955c\u50cf":79,"\u4f7f\u7528\u6570\u503c\u6cd5\u68c0\u6d4b\u68af\u5ea6\u6b63\u786e\u6027\u548c\u7a33\u5b9a\u6027":68,"\u4f7f\u7528\u6587\u6863":68,"\u4f7f\u7528\u663e\u5361\u8bad\u7ec3":49,"\u4f7f\u7528\u667a\u80fd\u6307\u9488\u7684\u539f\u56e0\u662f":34,"\u4f7f\u7528\u6848\u4f8b":76,"\u4f7f\u7528\u7684\u7f16\u8bd1\u5668\u4e3aarm":58,"\u4f7f\u7528\u76f8\u5bf9\u8def\u5f84\u7684\u5f15\u7528\u65b9\u5f0f":34,"\u4f7f\u7528\u8005\u4e0d\u9700\u8981\u5173\u5fc3":75,"\u4f7f\u7528\u8005\u53ea\u9700\u8981\u5173\u6ce8\u4e8e\u8bbe\u8ba1rnn\u5728\u4e00\u4e2a\u65f6\u95f4\u6b65\u4e4b\u5185\u5b8c\u6210\u7684\u8ba1\u7b97":63,"\u4f7f\u7528\u8005\u53ef\u4ee5\u4f7f\u7528\u4e0b\u9762\u7684python\u811a\u672c\u6765\u8bfb\u53d6\u53c2\u6570\u503c":87,"\u4f7f\u7528\u8005\u65e0\u9700\u5173\u5fc3\u8fd9\u4e2a\u53c2\u6570":75,"\u4f7f\u7528\u8005\u901a\u5e38\u65e0\u9700\u5173\u5fc3":75,"\u4f7f\u7528\u81ea\u52a8\u7684\u66ff\u8865\u6765\u66ff\u4ee3\u7ecf\u9a8c\u4e30\u5bcc\u7684\u4eba\u5de5\u8bc4\u5224":94,"\u4f7f\u7528\u8be5learning_rate_schedule\u65f6":51,"\u4f7f\u7528\u8c13\u8bcd\u4e0a\u4e0b\u6587":92,"\u4f7f\u7528\u8fd9\u4e2a\u795e\u7ecf\u7f51\u7edc\u53ef\u4ee5\u5b8c\u6210\u5bf9\u65b0\u6570\u636e\u7684\u9884\u6d4b":20,"\u4f7f\u7528\u8fd9\u4e2a\u811a\u672c\u524d\u8bf7\u786e\u8ba4\u5df2\u7ecf\u5b89\u88c5\u4e86pillow\u53ca\u76f8\u5173\u4f9d\u8d56\u6a21\u5757":86,"\u4f7f\u7528\u8fd9\u79cd\u65b9\u5f0f":61,"\u4f7f\u7528\u8fdc\u7a0b\u7a00\u758f\u65b9\u5f0f\u8bad\u7ec3\u65f6":67,"\u4f7f\u7528\u968f\u673a\u68af\u5ea6\u4e0b\u964d":93,"\u4f7f\u7528\u9759\u6001\u5e93\u548c\u52a8\u6001\u5e93\u96be\u5ea6\u5dee\u4e0d\u591a":33,"\u4f7f\u7528\u9884\u8bad\u7ec3\u7684\u6807\u51c6\u683c\u5f0f\u8bcd\u5411\u91cf\u6a21\u578b":85,"\u4f7f\u7528args\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u8bbe\u7f6e":2,"\u4f7f\u7528c":34,"\u4f7f\u7528c99\u505a\u63a5\u53e3":33,"\u4f7f\u7528c99\u800c\u4e0d\u4f7f\u7528c11\u7684\u539f\u56e0\u662f":33,"\u4f7f\u7528c99\u800c\u4e0d\u4f7f\u7528c89":33,"\u4f7f\u7528checkgrad\u6a21\u5f0f\u65f6\u7684\u53c2\u6570\u53d8\u5316\u5927\u5c0f":75,"\u4f7f\u7528cpu\u4e24\u7ebf\u7a0b\u8ba1\u7b97fc4\u5c42":77,"\u4f7f\u7528cpu\u8ba1\u7b97fc4\u5c42":77,"\u4f7f\u7528cpu\u8bad\u7ec3":93,"\u4f7f\u7528docker\u5c31\u4e0d\u7528\u914d\u7f6e\u4ea4\u53c9\u7f16\u8bd1\u73af\u5883\u4e86":65,"\u4f7f\u7528docker\u6784\u5efapaddlepaddle\u7684\u6587\u6863":70,"\u4f7f\u7528init":77,"\u4f7f\u7528lstm\u4f5c\u4e3aencod":61,"\u4f7f\u7528max":86,"\u4f7f\u7528memory\u7684rnn\u5b9e\u73b0\u4fbf\u5982\u4e0b\u56fe\u6240\u793a":61,"\u4f7f\u7528model":77,"\u4f7f\u7528paddlepaddl":89,"\u4f7f\u7528python\u6570\u636e\u63d0\u4f9b\u5668":86,"\u4f7f\u7528rdma\u8fd8\u662ftcp\u4f20\u8f93\u534f\u8bae":75,"\u4f7f\u7528regress":42,"\u4f7f\u7528swig\u53ea\u652f\u6301cpython\u89e3\u91ca\u5668":33,"\u4f7f\u7528swig\u9700\u8981\u591a\u8bed\u8a00\u7ed1\u5b9a\u7684\u5f00\u53d1\u4eba\u5458\u719f\u7ec3\u638c\u63e1swig\u914d\u7f6e":33,"\u4f7f\u7528void":33,"\u4f7f\u8be5\u5c42\u7684\u53c2\u6570\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u4fdd\u6301\u4e0d\u53d8":51,"\u4f7f\u8f93\u5165\u5c42\u5230\u9690\u85cf\u5c42\u7684\u795e\u7ecf\u5143\u662f\u5168\u90e8\u8fde\u63a5\u7684":86,"\u4f86":61,"\u4f8b\u5982":[2,21,33,34,42,49,50,53,56,61,64,67,72,73,74,75,77,80,87,89,91,93],"\u4f8b\u5982\u4e0a\u6587\u7684pod":78,"\u4f8b\u5982\u4e0a\u9762\u7684":52,"\u4f8b\u5982\u4ee5\u592a\u7f51\u7684":73,"\u4f8b\u5982\u4f7f\u7528":49,"\u4f8b\u5982\u5bf9\u4e8ejava\u6216\u8005python":33,"\u4f8b\u5982\u5bf9\u4e8ejava\u6765\u8bf4":33,"\u4f8b\u5982\u5bf9\u4e8epython":33,"\u4f8b\u5982\u5c06\u7b2c\u4e00\u6761\u6570\u636e\u8f6c\u5316\u4e3a":61,"\u4f8b\u5982\u6587\u672c\u5206\u7c7b\u4e2d":61,"\u4f8b\u5982\u672c\u4f8b\u4e2d\u7684\u4e24\u4e2a\u7279\u5f81":61,"\u4f8b\u5982\u673a\u5668\u4e0a\u67094\u5757gpu":49,"\u4f8b\u5982\u7b2c300\u4e2apass\u7684\u6a21\u578b\u4f1a\u88ab\u4fdd\u5b58\u5728":86,"\u4f8b\u5982c":33,"\u4f8b\u5982hostpath":78,"\u4f8b\u5982java\u4e0epython\u7684\u9519\u8bef\u5904\u7406\u662f\u76f4\u63a5\u6254\u51fa\u6765except":33,"\u4f8b\u5982output\u76ee\u5f55\u4e0b\u5c31\u5b58\u653e\u4e86\u8f93\u51fa\u7ed3\u679c":80,"\u4f8b\u5982python\u53ef\u4ee5\u4f7f\u7528":33,"\u4f8b\u5982python\u7684":33,"\u4f8b\u5982sigmoid":67,"\u4f8b\u5982sigmoid\u53d8\u6362":89,"\u4f8b\u5b50\u4e2d\u662f":67,"\u4f8b\u5b50\u4e2d\u662f0":67,"\u4f8b\u5b50\u4e2d\u662f100":67,"\u4f8b\u5b50\u4e2d\u662f4096":67,"\u4f8b\u5b50\u4e2d\u662f8192":67,"\u4f8b\u5b50\u4e2d\u662ffc":67,"\u4f8b\u5b50\u4e2d\u662fsoftmax":67,"\u4f8b\u5b50\u4f7f\u7528":78,"\u4f9bpaddlepaddle\u52a0\u8f7d":75,"\u4f9d\u636e\u5206\u7c7b\u9519\u8bef\u7387\u83b7\u5f97\u6700\u4f73\u6a21\u578b\u8fdb\u884c\u6d4b\u8bd5":93,"\u4f9d\u636e\u662f\u5426\u5305\u542bkernel":68,"\u4f9d\u6b21\u7c7b\u63a8":42,"\u4f9d\u8d56\u4e8epython\u7684":86,"\u4fbf\u4e8e\u5b58\u50a8\u8d44\u6e90\u7ba1\u7406\u548cpod\u5f15\u7528":78,"\u4fbf\u4e8e\u672c\u5730\u9a8c\u8bc1\u548c\u6d4b\u8bd5":78,"\u4fbf\u4e8e\u7528\u6237\u6d4f\u89c8c":54,"\u4fbf\u5229":61,"\u4fbf\u548c\u5355\u5c42rnn\u914d\u7f6e\u4e2d\u7684":61,"\u4fbf\u5b9c":61,"\u4fbf\u662f\u5c06\u9759\u6001\u5e93\u52a0\u5165jvm\u4e2d":33,"\u4fdd\u5b58\u6a21\u578b\u53c2\u6570\u7684\u76ee\u5f55":75,"\u4fdd\u5b58\u751f\u6210\u7ed3\u679c\u7684\u6587\u4ef6":94,"\u4fdd\u5b58\u7684\u53c2\u6570\u4e5f\u662ffloat\u7c7b\u578b":51,"\u4fdd\u5b58\u7f51\u7edc\u5c42\u8f93\u51fa\u7ed3\u679c\u7684\u76ee\u5f55":75,"\u4fdd\u5b58\u9884\u6d4b\u7ed3\u679c\u7684\u6587\u4ef6\u540d":75,"\u4fdd\u6301\u5bbd\u9ad8\u6bd4\u7f29\u653e\u5230\u77ed\u8fb9\u4e3a256":87,"\u4fdd\u6301\u5c3d\u91cf\u5c11\u7684commit":66,"\u4fe1\u53f7\u6765\u81ea\u52a8\u7ec8\u6b62\u5b83\u542f\u52a8\u7684\u6240\u6709\u8fdb\u7a0b":73,"\u4fee\u590d\u6240\u6709bug\u540e":42,"\u4fee\u590ddocker\u7f16\u8bd1\u955c\u50cf\u95ee\u9898":42,"\u4fee\u590dubuntu":42,"\u4fee\u6539":[78,79],"\u4fee\u6539\u542f\u52a8\u811a\u672c\u540e":79,"\u4fee\u6539\u6210":42,"\u4fee\u6539\u6210\u66f4\u5feb\u7684\u7248\u672c":72,"\u4fee\u6539\u6587\u6863":71,"\u503c\u5f97\u6ce8\u610f\u7684\u662f":[61,66],"\u503c\u5f97\u6df1\u5165\u5206\u6790":72,"\u503c\u7c7b\u578b":77,"\u5047\u5982\u6211\u4eec\u662f\u4e09\u5206\u7c7b\u95ee\u9898":51,"\u5047\u8bbe":67,"\u5047\u8bbe\u53d8\u91cf":52,"\u5047\u8bbe\u60a8\u5df2\u7ecf\u5b8c\u6210\u4e86\u4e00\u4e2aai\u8bad\u7ec3\u7684python\u7a0b\u5e8f":54,"\u5047\u8bbe\u635f\u5931\u51fd\u6570\u662f":67,"\u5047\u8bbe\u7b2c\u4e00\u4e2alayer\u7684\u8f93\u51faa\u662f\u4e00\u4e2a":49,"\u5047\u8bbe\u8bcd\u5411\u91cf\u7ef4\u5ea6\u4e3a32":85,"\u504f\u7f6e\u53c2\u6570":87,"\u504f\u7f6e\u53c2\u6570\u7684\u5927\u5c0f":67,"\u505a\u4e00\u4e2a\u4ecb\u7ecd":69,"\u505a\u53ea\u8bfb\u6302\u8f7d":21,"\u505a\u5982\u4e0b\u51e0\u4e2a\u64cd\u4f5c":42,"\u505a\u63a5\u53e3":33,"\u505a\u68af\u5ea6\u68c0\u6d4b":68,"\u505a\u68c0\u67e5":68,"\u505c\u6b62\u4fdd\u5b58\u68c0\u67e5\u70b9\u7684\u7ebf\u7a0b":20,"\u505c\u6b62\u52a0\u8f7d\u6570\u636e":75,"\u505c\u7535":61,"\u50cf":54,"\u513f\u7ae5\u7247":90,"\u5143\u7d20":60,"\u5143\u7d20\u4e4b\u95f4\u7684\u987a\u5e8f\u662f\u91cd\u8981\u7684\u8f93\u5165\u4fe1\u606f":60,"\u5148\u5b9e\u73b0\u6a21\u578b\u63a8\u65ad\u7684api":34,"\u5148\u627e\u51fa\u53c2\u6570":50,"\u5148\u67e5\u770b\u4e00\u4e0b\u662f\u5426\u66fe\u7ecf\u5b89\u88c5\u8fc7paddl":46,"\u5148\u8c03\u7528initializer\u51fd\u6570":89,"\u5168\u5bb6":61,"\u5168\u8fde\u63a5\u5c42":[52,85,86,91],"\u5168\u8fde\u63a5\u5c42\u4ee5\u4e00\u4e2a\u7ef4\u5ea6\u4e3a":67,"\u5168\u8fde\u63a5\u5c42\u5c06\u7535\u5f71\u7684\u6bcf\u4e2a\u7279\u5f81\u7ed3\u5408\u6210\u4e00\u4e2a\u7535\u5f71\u7279\u5f81":91,"\u5168\u8fde\u63a5\u5c42\u6743\u91cd":87,"\u5168\u8fde\u63a5\u5c42\u6ca1\u6709\u7f51\u7edc\u5c42\u914d\u7f6e\u7684\u8d85\u53c2\u6570":67,"\u5168\u8fde\u63a5\u5c42\u7684\u5b9e\u73b0\u4f4d\u4e8e":67,"\u5168\u8fde\u63a5\u5c42\u7684\u6bcf\u4e2a\u8f93\u51fa\u90fd\u8fde\u63a5\u5230\u4e0a\u4e00\u5c42\u7684\u6240\u6709\u7684\u795e\u7ecf\u5143\u4e0a":67,"\u5168\u8fde\u63a5\u5c42python\u5c01\u88c5\u7684\u4f8b\u5b50\u4e2d\u5305\u542b\u4e0b\u9762\u51e0\u6b65":67,"\u516b\u4e2a\u7279\u5f81\u5206\u522b\u8f6c\u6362\u4e3a\u5411\u91cf":92,"\u516c\u5f0f":54,"\u516c\u94a5\u5199\u5165":73,"\u516d\u4e2a\u7279\u5f81\u548c\u6807\u7b7e\u90fd\u662f\u7d22\u5f15\u69fd":92,"\u5171\u4eab\u4e00\u4e2aop\u5b9a\u4e49":68,"\u5171\u4eab\u4efb\u52a1\u4e2d\u8bbe\u7f6e\u7684\u6570\u636e\u8fdb\u884c\u8bad\u7ec3\u548c\u6d4b\u8bd5":92,"\u5171\u4eab\u540c\u4e00\u4e2akernel\u65f6":68,"\u5171\u4eab\u5b58\u50a8\u6302\u5728\u7684\u8def\u5f84":80,"\u5171\u670932":85,"\u5173\u4e8e\u5728paddlepaddle\u4e2d\u5982\u4f55\u4f7f\u7528eigen\u5e93":68,"\u5173\u4e8e\u5982\u4f55\u5b9a\u4e49\u7f51\u7edc\u4e2d\u7684\u5c42":86,"\u5173\u4e8e\u65f6\u95f4\u5e8f\u5217":61,"\u5173\u4e8e\u6784\u5efa\u548c\u6d4b\u8bd5\u7684\u66f4\u591a\u4fe1\u606f":66,"\u5173\u4e8eeigen":69,"\u5173\u4e8elstm":50,"\u5173\u4e8epaddlepaddle\u7684\u5206\u5e03\u5f0f\u8bad\u7ec3":80,"\u5173\u4e8epaddlepaddle\u7684\u66f4\u591a\u4f7f\u7528\u65b9\u6cd5\u8bf7\u53c2\u8003":56,"\u5173\u4e8eunbound":63,"\u5173\u4e8evgg\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u7684\u63cf\u8ff0\u53ef\u4ee5\u53c2\u8003":86,"\u5173\u952e\u8bcd\u5305\u62ec":66,"\u5176\u4e0b\u5b50\u6587\u4ef6\u5939\u7684\u7ed3\u6784\u5982\u4e0b":86,"\u5176\u4e2d":[2,33,42,49,51,52,56,58,59,64,67,85,86,87],"\u5176\u4e2d156\u548c285\u662f\u8fd9\u4e9b\u56fe\u50cf\u7684\u5206\u7c7b\u6807\u7b7e":87,"\u5176\u4e2d50000\u5f20\u56fe\u7247\u4f5c\u4e3a\u8bad\u7ec3\u96c6":86,"\u5176\u4e2d\u5206\u522b\u5305\u542b\u4e86cifar":86,"\u5176\u4e2d\u5305\u542b6":90,"\u5176\u4e2d\u5305\u542b\u4e86200\u79cd\u9e1f\u7c7b\u7684\u7167\u7247":86,"\u5176\u4e2d\u5305\u542b\u7b97\u6cd5\u548c\u7f51\u7edc\u914d\u7f6e":93,"\u5176\u4e2d\u5305\u62ec\u51fd\u6570":92,"\u5176\u4e2d\u5b9a\u4e49\u4e86\u6a21\u578b\u67b6\u6784\u548csolver\u914d\u7f6e":94,"\u5176\u4e2d\u6587\u672c\u8f93\u5165\u7c7b\u578b\u5b9a\u4e49\u4e3a\u6574\u6570\u65f6\u5e8f\u7c7b\u578binteger_value_sequ":89,"\u5176\u4e2d\u6bcf\u4e00\u884c\u5bf9\u5e94\u4e00\u4e2a\u6570\u636e\u6587\u4ef6\u5730\u5740":1,"\u5176\u4e2d\u6bcf\u4e2a\u5143\u7d20\u662f\u53cc\u5c42\u5e8f\u5217\u4e2d\u6bcf\u4e2asubseq\u6700\u540e\u4e00\u4e2a":60,"\u5176\u4e2d\u6bcf\u4e2a\u5411\u91cf\u5bf9\u5e94\u8f93\u5165\u8bed\u53e5\u4e2d\u7684\u4e00\u4e2a\u5143\u7d20":94,"\u5176\u4e2d\u6bcf\u6761pass\u82b1\u8d39\u4e867\u4e2a\u5c0f\u65f6":94,"\u5176\u4e2d\u6bcf\u884c\u6570\u636e\u4ee3\u8868\u4e00\u5f20\u56fe\u7247":2,"\u5176\u4e2d\u8be6\u7ec6\u8bf4\u660e\u4e86\u6a21\u578b\u67b6\u6784":94,"\u5176\u4e2d\u8f93\u5165\u56fe\u50cf\u7684\u989c\u8272\u901a\u9053\u987a\u5e8f\u4e3a":87,"\u5176\u4e2dbeam":94,"\u5176\u4e2dcheckgrad\u4e3b\u8981\u4e3a\u5f00\u53d1\u8005\u4f7f\u7528":75,"\u5176\u4e2dmean\u548cstd\u662f\u8bad\u7ec3\u914d\u7f6e\u4e2d\u7684\u53c2\u6570":75,"\u5176\u4e2dvalue\u5373\u4e3asoftmax\u5c42\u7684\u8f93\u51fa":4,"\u5176\u4e2dx\u8868\u793a\u8f93\u5165\u6570\u636e\u662f\u4e00\u4e2a\u7ef4\u5ea6\u4e3a2\u7684\u7a20\u5bc6\u5411\u91cf":56,"\u5176\u4e3b\u8981\u63a5\u53e3\u5982\u4e0b":69,"\u5176\u4ed6":90,"\u5176\u4ed6\u4eba\u53ef\u4ee5\u590d\u73b0\u95ee\u9898\u4ee5\u4fbf\u5e2e\u52a9":65,"\u5176\u4ed6\u4eba\u5f88\u5bb9\u6613\u770b\u61c2\u955c\u50cf\u751f\u6210\u8fc7\u7a0b":54,"\u5176\u4ed6\u516d\u884c\u5217\u51fa\u4e86\u96c6\u675f\u641c\u7d22\u7684\u7ed3\u679c":94,"\u5176\u4ed6\u5185\u5b58\u6742\u9879":49,"\u5176\u4ed6\u5185\u5b58\u6742\u9879\u662f\u6307paddlepaddle\u672c\u8eab\u6240\u7528\u7684\u4e00\u4e9b\u5185\u5b58":49,"\u5176\u4ed6\u51fd\u6570\u5747\u8fd4\u56de":34,"\u5176\u4ed6\u53c2\u6570\u4f7f\u7528":2,"\u5176\u4ed6\u53c2\u6570\u8bf7\u53c2\u8003":89,"\u5176\u4ed6\u6240\u6709\u5c42\u90fd\u4f1a\u4f7f\u7528gpu\u8ba1\u7b97":77,"\u5176\u4ed6\u7528\u6237\u7684fork\u7248\u672c\u5e93\u5e76\u4e0d\u9700\u8981\u4e25\u683c\u9075\u5b88":42,"\u5176\u4ed6\u914d\u7f6e\u53c2\u6570":[58,59],"\u5176\u4ed6\u9ad8\u7ea7\u529f\u80fd\u5305\u62ec\u5b9a\u4e49\u591a\u4e2amemori":64,"\u5176\u4f1a\u81ea\u52a8\u88ab\u52a0\u5165\u7f16\u8bd1\u5217\u8868":67,"\u5176\u4f59\u884c\u662f":85,"\u5176\u4f5c\u7528\u662f\u5c06\u6570\u636e\u4f20\u5165\u5185\u5b58\u6216\u663e\u5b58":1,"\u5176\u5177\u4f53\u8bf4\u660e\u4e86\u5b57\u6bb5\u7c7b\u578b\u548c\u6587\u4ef6\u540d\u79f0":91,"\u5176\u5185\u90e8\u7684\u6587\u4ef6\u4e5f\u4f1a\u968f\u4e4b\u6d88\u5931":78,"\u5176\u5305\u62ec\u4e24\u4e2a\u51fd\u6570":89,"\u5176\u53c2\u6570\u5982\u4e0b":2,"\u5176\u5b83\u90e8\u5206\u548c\u903b\u8f91\u56de\u5f52\u7f51\u7edc\u7ed3\u6784\u4e00\u81f4":89,"\u5176\u5b83layer\u7684\u8f93\u51fa":63,"\u5176\u5b9e\u4e5f\u662f\u548c\u6bcf\u4e2amini":49,"\u5176\u63d0\u4f9b\u5e94\u7528\u90e8\u7f72":78,"\u5176\u6b21":[2,61,89],"\u5176\u76ee\u7684\u662f\u5728\u7ed9\u5b9a\u7684\u8f93\u5165\u53e5\u5b50\u4e2d\u53d1\u73b0\u6bcf\u4e2a\u8c13\u8bcd\u7684\u8c13\u8bcd\u8bba\u5143\u7ed3\u6784":92,"\u5176\u8bf4\u660e\u5982\u4e0b":61,"\u5176\u8f93\u5165\u53c2\u6570\u5982\u4e0b":94,"\u5176\u8f93\u51fa\u88ab\u7528\u4f5cmemory\u7684\u521d\u59cb\u503c":64,"\u5176name\u7531\u53c2\u6570":50,"\u5177\u4f53\u4f7f\u7528\u65b9\u6cd5\u4e3a":[34,49],"\u5177\u4f53\u505a\u6cd5\u8bf7\u53c2\u8003":65,"\u5177\u4f53\u539f\u56e0\u53c2\u8003":34,"\u5177\u4f53\u53ef\u4ee5\u53c2\u8003":[2,49,67],"\u5177\u4f53\u53ef\u53c2\u8003\u6587\u6863":63,"\u5177\u4f53\u5982\u4e0b":54,"\u5177\u4f53\u60c5\u51b5\u56e0\u4eba\u800c\u5f02":72,"\u5177\u4f53\u64cd\u4f5c\u5982\u4e0b":46,"\u5177\u4f53\u6d41\u7a0b\u5982\u4e0b":89,"\u5177\u4f53\u7684\u683c\u5f0f\u8bf4\u660e":2,"\u5177\u4f53\u7684\u89e3\u51b3\u65b9\u6cd5\u662f":46,"\u5177\u4f53\u8bf7\u53c2\u7167\u793a\u4f8b":87,"\u5177\u4f53\u8bf7\u53c2\u8003":[2,34,66],"\u5177\u4f53\u8bf7\u89c1":66,"\u5177\u6709\u76f8\u540c\u7684\u7ed3\u679c\u4e86":61,"\u5177\u6709\u81ea\u5faa\u73af\u8fde\u63a5\u7684\u795e\u7ecf\u5143":93,"\u5185":64,"\u5185\u5b58":72,"\u5185\u5b58\u4e0d\u8db3":47,"\u5185\u5b58\u5bb9\u9650\u9608\u503c":75,"\u5185\u5bb9":[68,89],"\u5185\u5bb9\u5982\u4e0b":79,"\u5185\u5c42inner_step\u7684recurrent_group\u548c\u5355\u5c42\u5e8f\u5217\u7684\u51e0\u4e4e\u4e00\u6837":61,"\u5185\u5df2\u7ecf\u5305\u542bpaddlepaddle\u7684\u6267\u884c\u7a0b\u5e8f\u4f46\u662f\u8fd8\u6ca1\u4e0a\u8ff0\u529f\u80fd":80,"\u5185\u90e8":80,"\u5185\u90e8\u9a71\u52a8python\u89e3\u91ca\u5668\u8fdb\u884c\u6a21\u578b\u914d\u7f6e\u89e3\u6790\u548c\u6570\u636e\u8bfb\u53d6":33,"\u518d\u4e3apaddle\u7684\u8bad\u7ec3\u8fc7\u7a0b\u63d0\u4f9b\u6587\u4ef6\u5217\u8868":91,"\u518d\u4ee5":68,"\u518d\u5199\u5165\u7f51\u7edc\u53c2\u6570":51,"\u518d\u5728\u6bcf\u4e00\u4e2aapi\u4e2d\u81ea\u5df1\u68c0\u67e5\u7c7b\u578b":33,"\u518d\u57fa\u4e8e":42,"\u518d\u5b89\u88c5":46,"\u518d\u5bf9\u6bcf\u4e00\u4e2a\u5355\u5c42\u65f6\u95f4\u5e8f\u5217\u8fdb\u884c\u5904\u7406":61,"\u518d\u5bf9\u6bcf\u4e00\u53e5\u8bdd\u7684\u7f16\u7801\u5411\u91cf\u7528lstm\u7f16\u7801\u6210\u4e00\u4e2a\u6bb5\u843d\u7684\u5411\u91cf":61,"\u518d\u5bf9\u8fd9\u4e2a\u6bb5\u843d\u5411\u91cf\u8fdb\u884c\u5206\u7c7b":61,"\u518d\u6307\u5b9a":53,"\u518d\u6b21\u5bf9\u4ee3\u7801\u8fdb\u884c\u6027\u80fd\u5206\u6790":72,"\u518d\u7528\u8fd9\u4e2a\u68af\u5ea6\u53bb\u548c":67,"\u518d\u901a\u8fc7\u51fd\u6570":80,"\u5192\u9669\u7247":90,"\u5197\u4f59\u7b49\u529f\u80fd":78,"\u5199\u4ee3\u7801":33,"\u5199\u5165\u5feb\u7167\u6570\u636e":20,"\u5199\u5b8c\u6a21\u578b\u914d\u7f6e\u4e4b\u540e":94,"\u5199\u68af\u5ea6\u68c0\u67e5\u5355\u5143\u6d4b\u8bd5\u662f\u4e00\u4e2a\u9a8c\u8bc1\u65b0\u5b9e\u73b0\u7684\u5c42\u662f\u5426\u6b63\u786e\u7684\u76f8\u5bf9\u7b80\u5355\u7684\u529e\u6cd5":67,"\u519c\u6c11":90,"\u51c6\u5907":61,"\u51c6\u5907\u597d\u6570\u636e":91,"\u51c6\u5907\u9884\u6d4b\u6570\u636e":4,"\u51cf\u5c0f\u5e8f\u5217\u7684\u957f\u5ea6":49,"\u51cf\u5c0f\u8fd9\u4e2a\u5185\u5b58\u6c60\u5373\u53ef\u51cf\u5c0f\u5185\u5b58\u5360\u7528":49,"\u51cf\u5c0fbatch":49,"\u51e0\u4e2a\u57fa\u7840\u7684\u6982\u5ff5\u5e2e\u52a9\u7406\u89e3\u548c\u4f7f\u7528docker":54,"\u51fa\u53bb\u73a9":61,"\u51fa\u5dee":61,"\u51fa\u6765":61,"\u51fa\u73b0":46,"\u51fa\u73b0\u4ee5\u4e0b\u9519\u8bef":51,"\u51fa\u73b0\u8be5\u9519\u8bef\u7684\u539f\u56e0\u4e00\u822c\u662f\u7528\u6237\u5bf9\u4e0d\u540clayer\u7684\u53c2\u6570":50,"\u51fa\u73b0\u8fd9\u4e2a\u95ee\u9898\u7684\u4e3b\u8981\u539f\u56e0\u662f":46,"\u51fd\u6570":[2,52,64,67,72,92,93],"\u51fd\u6570\u4e2d\u4f7f\u7528":2,"\u51fd\u6570\u4e2d\u64cd\u4f5c\u7684\u91cd\u8981\u53d8\u91cf\u7684\u8be6\u7ec6\u89e3\u91ca":68,"\u51fd\u6570\u4e2d\u8bbe\u7f6e\u7684":73,"\u51fd\u6570\u5047\u8bbe":64,"\u51fd\u6570\u52a0\u5230\u4ee3\u7801\u4e2d":72,"\u51fd\u6570\u5373\u53ef\u5b8c\u6210\u8f6c\u6362":21,"\u51fd\u6570\u53ea\u5173\u6ce8\u4e8ernn\u4e00\u4e2a\u65f6\u95f4\u6b65\u4e4b\u5185\u7684\u8ba1\u7b97":63,"\u51fd\u6570\u540d\u4e3a":34,"\u51fd\u6570\u547d\u540d":33,"\u51fd\u6570\u5b9a\u4e49\u8f93\u5165":68,"\u51fd\u6570\u5bf9\u6bd4cpu":68,"\u51fd\u6570\u5c31\u662f\u6839\u636e\u8be5\u673a\u5236\u914d\u7f6e\u7684":2,"\u51fd\u6570\u5f97\u5230\u7684\u68af\u5ea6\u53bb\u5bf9\u6bd4":67,"\u51fd\u6570\u5fc5\u987b\u5148\u8c03\u7528\u57fa\u7c7b\u4e2d\u7684\u51fd\u6570":67,"\u51fd\u6570\u5fc5\u987b\u8fd4\u56de\u4e00\u4e2a\u6216\u591a\u4e2alayer\u7684\u8f93\u51fa":63,"\u51fd\u6570\u6307\u51fa\u4e86\u5728\u8bad\u7ec3\u65f6\u9700\u8981\u4ece\u53c2\u6570\u670d\u52a1\u5668\u53d6\u51fa\u7684\u884c":67,"\u51fd\u6570\u6765\u5c06\u4fe1\u606f\u8f93\u51fa\u5230\u754c\u9762\u4e2d":72,"\u51fd\u6570\u67e5\u8be2\u8f6f\u4ef6\u5305\u76f8\u5173api\u8bf4\u660e":4,"\u51fd\u6570\u7684":2,"\u51fd\u6570\u7684\u5b9e\u73b0\u662f\u6b63\u786e\u7684":67,"\u51fd\u6570\u7684\u5f00\u5934\u5fc5\u987b\u8c03\u7528":67,"\u51fd\u6570\u91cc\u5b9e\u73b0":68,"\u5206\u4e3a\u597d\u8bc4":89,"\u5206\u522b\u4e3a":85,"\u5206\u522b\u4e3atrain":94,"\u5206\u522b\u4ece\u8bcd\u8bed\u548c\u53e5\u5b50\u7ea7\u522b\u7f16\u7801\u8f93\u5165\u6570\u636e":63,"\u5206\u522b\u4f7f\u7528\u5355\u53cc\u5c42rnn\u4f5c\u4e3a\u7f51\u7edc\u914d\u7f6e\u7684\u6a21\u578b":61,"\u5206\u522b\u5305\u542b\u4e86\u6cd5\u8bed\u5230\u82f1\u8bed\u7684\u5e73\u884c\u8bed\u6599\u5e93\u7684\u8bad\u7ec3\u6570\u636e":94,"\u5206\u522b\u5b9a\u4e49\u5b50\u53e5\u7ea7\u522b\u548c\u8bcd\u8bed\u7ea7\u522b\u4e0a\u9700\u8981\u5b8c\u6210\u7684\u8fd0\u7b97":63,"\u5206\u522b\u5bf9\u5e94\u4e8e\u53d8\u91cf":52,"\u5206\u522b\u662f":60,"\u5206\u522b\u662frnn\u72b6\u6001\u548c\u8f93\u5165\u7684\u53d8\u6362\u77e9\u9635":64,"\u5206\u522b\u662fsentences\u548clabel":61,"\u5206\u522b\u662fwords\u548clabel":61,"\u5206\u522b\u8ba1\u7b97\u6bcf\u4e2a\u53c2\u6570\u7684\u68af\u5ea6":67,"\u5206\u522b\u8fdb\u884c\u5e8f\u5217\u64cd\u4f5c":61,"\u5206\u5272":[90,92],"\u5206\u5272\u6587\u4ef6\u7684\u65b9\u6cd5\u662f":91,"\u5206\u5e03\u5f0f\u5b58\u50a8\u670d\u52a1":20,"\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf":78,"\u5206\u6210\u4e24\u90e8\u5206":2,"\u5206\u652f":[42,66],"\u5206\u652f\u4e00\u65e6\u5efa\u7acb":42,"\u5206\u652f\u4e0a":66,"\u5206\u652f\u4e0a\u521b\u5efa\u65b0\u5206\u652f":66,"\u5206\u652f\u4e2d":42,"\u5206\u652f\u4e3a\u5f00\u53d1":42,"\u5206\u652f\u4e3a\u6bcf\u4e00\u6b21release\u65f6\u5efa\u7acb\u7684\u4e34\u65f6\u5206\u652f":42,"\u5206\u652f\u4e3a\u7a33\u5b9a":42,"\u5206\u652f\u529f\u80fd\u7684\u5c01\u95ed":42,"\u5206\u652f\u5408\u5165":42,"\u5206\u652f\u5408\u5165master\u5206\u652f":42,"\u5206\u652f\u540c\u6b65\u4e3b\u7248\u672c\u5e93\u7684":42,"\u5206\u652f\u540d":66,"\u5206\u652f\u540d\u4e3a":42,"\u5206\u652f\u5b58\u5728\u7684\u65f6\u5019":42,"\u5206\u652f\u6d3e\u751f\u51fa\u65b0\u7684\u5206\u652f":42,"\u5206\u652f\u7528\u6765\u6d4b\u8bd5\u53ea\u9700\u8981\u8ba1\u7b97\u4e00\u4e2a\u8f93\u5165\u68af\u5ea6\u7684\u60c5\u51b5":68,"\u5206\u652f\u7684\u7248\u672c\u90fd\u662f\u7ecf\u8fc7\u5355\u5143\u6d4b\u8bd5\u548c\u56de\u5f52\u6d4b\u8bd5\u7684\u7248\u672c":42,"\u5206\u652f\u7684\u7248\u672c\u90fd\u7ecf\u8fc7\u5355\u5143\u6d4b\u8bd5":42,"\u5206\u652f\u7684\u76ee\u6807\u673a\u7248openblas\u5e93":58,"\u5206\u652f\u89c4\u8303":66,"\u5206\u6790\u5f97\u5230\u7684\u4fe1\u606f\u7528\u4e8e\u534f\u52a9\u8fdb\u884c\u7a0b\u5e8f\u7684\u4f18\u5316":72,"\u5206\u7247":20,"\u5206\u7c7b\u6210\u6b63\u9762\u60c5\u7eea\u548c\u8d1f\u9762\u60c5\u7eea\u4e24\u7c7b":2,"\u5206\u7c7b\u8bef\u5dee\u662f0":93,"\u5206\u7c7b\u9519\u8bef\u7387\u548c\u6a21\u578b\u5927\u5c0f\u7531\u4e0b\u8868\u7ed9\u51fa":87,"\u5206\u8bcd\u5e8f\u5217\u7684\u5f00\u59cb":85,"\u5206\u8bcd\u5e8f\u5217\u7684\u7ed3\u675f":85,"\u5206\u8bcd\u98ce\u683c\u5982\u4e0b":85,"\u5206\u914d\u5230\u5f53\u524d\u6570\u636e\u5757\u6837\u672c\u6570\u7684\u56db\u5206\u4e4b\u4e00":75,"\u5206\u9694":[85,91],"\u5206\u9694\u7b26\u4e3a":90,"\u5207\u6362\u5230":66,"\u5207\u6362\u5230\u6240\u5efa\u5206\u652f":66,"\u5217\u8868":91,"\u5217\u8868\u5982\u4e0b":[2,56],"\u5219\u4e0d\u5728\u4e4e\u5185\u5b58\u6682\u5b58\u591a\u5c11\u6761\u6570\u636e":2,"\u5219\u4e0d\u9700\u8981\u91cd\u5199\u8be5\u51fd\u6570":67,"\u5219\u4f1a\u9884\u5148\u8bfb\u53d6\u5168\u90e8\u6570\u636e\u5230\u5185\u5b58\u4e2d":2,"\u5219\u4f7f\u7528\u533a\u57df\u6807\u8bb0":92,"\u5219\u4f7f\u7528\u540c\u6b65\u8bad\u7ec3":75,"\u5219\u4f7f\u7528\u542f\u52a8\u53c2\u6570\u5b9a\u4e49\u7684\u521d\u59cb\u5316\u65b9\u6cd5\u521d\u59cb\u5316\u53c2\u6570":20,"\u5219\u4f7f\u7528\u8be5\u53c2\u6570\u4f5c\u4e3a\u9ed8\u8ba4\u503c":75,"\u5219\u5148\u505a\u5d4c\u5165":91,"\u5219\u53ef\u4ee5\u50cf":73,"\u5219\u53ef\u8bbe\u7f6e":[58,59],"\u5219\u5b57\u4e0e\u5b57\u4e4b\u95f4\u7528\u7a7a\u683c\u5206\u9694":89,"\u5219\u5ffd\u7565":20,"\u5219\u603b\u4f1a\u663e\u793a\u963b\u9694\u6458\u8981\u4fe1\u606f":75,"\u5219\u628a\u53e6\u4e00\u4e2a\u6162\u901f\u7684kill\u6389":20,"\u5219\u63a8\u8350\u5927\u4e8e\u8bad\u7ec3\u65f6batch":2,"\u5219\u662f\u5e26gui\u7684nvidia\u53ef\u89c6\u5316\u6027\u80fd\u5206\u6790\u5de5\u5177":72,"\u5219\u663e\u793a\u963b\u9694\u6027\u80fd\u7684\u6458\u8981\u4fe1\u606f":75,"\u5219\u76f4\u63a5\u5f15\u5165\u53e6\u4e00\u79cd\u7c7b\u578b\u7684\u5934\u6587\u4ef6":34,"\u5219\u8bbe\u7f6e\u6210":[58,59],"\u5219\u9700\u8981\u4f7f\u7528\u7b49\u4e8e\u6743\u91cd\u53c2\u6570\u89c4\u6a21\u5927\u7ea65\u500d\u7684\u5185\u5b58":49,"\u5219\u9700\u8981\u56de\u6eda\u5230\u4e0a\u4e00\u4e2a\u68c0\u67e5\u70b9":20,"\u5219\u9700\u8981\u914d\u7f6e":78,"\u5219\u9700\u9009\u7528":59,"\u521b\u5efa\u4e00\u4e2akubernet":80,"\u521b\u5efa\u53cd\u5411op\u5bf9\u5e94\u7684\u524d\u5411op":68,"\u521b\u5efa\u548c\u53d1\u5e03\u81ea\u5df1\u7684ai\u7a0b\u5e8f\u955c\u50cf":54,"\u521b\u5efa\u5e76\u5207\u6362\u5230\u65b0\u5206\u652f":66,"\u521b\u5efa\u6210\u529f\u540e":80,"\u521b\u5efa\u8bad\u7ec3\u6570\u636e\u7684":94,"\u521b\u5efagener":2,"\u521d\u59cb\u5316\u4e4b\u540e":4,"\u521d\u59cb\u5316\u504f\u7f6e\u5411\u91cf":67,"\u521d\u59cb\u5316\u65f6\u8c03\u7528\u7684\u51fd\u6570":2,"\u521d\u59cb\u5316\u6743\u91cd\u8868":67,"\u521d\u59cb\u5316\u6a21\u578b\u7684\u8def\u5f84":75,"\u521d\u59cb\u5316\u6a21\u578b\u7684\u8def\u5f84\u914d\u7f6e\u4e3a":85,"\u521d\u59cb\u5316\u7236\u7c7b":67,"\u521d\u59cb\u5316biases_":67,"\u521d\u59cb\u5316paddlepaddle\u73af\u5883":4,"\u521d\u59cb\u72b6\u6001":63,"\u5220\u9664":66,"\u5220\u9664\u78c1\u76d8\u76ee\u5f55\u4e2d\u4e0d\u662f\u5f53\u524duuid\u7684\u5feb\u7167\u6587\u4ef6":20,"\u5229\u7528\u5206\u5e03\u5f0f\u8bad\u7ec3\u9a7e\u9a6d\u66f4\u591a\u7684\u8ba1\u7b97\u8d44\u6e90":49,"\u5229\u7528\u5355\u8bcdid\u67e5\u627e\u8be5\u5355\u8bcd\u5bf9\u5e94\u7684\u8fde\u7eed\u5411\u91cf":89,"\u5229\u7528\u66f4\u591a\u7684\u8ba1\u7b97\u8d44\u6e90\u53ef\u4ee5\u5206\u4e3a\u4e00\u4e0b\u51e0\u4e2a\u65b9\u5f0f\u6765\u8fdb\u884c":49,"\u5229\u7528\u6b63\u5219\u8868\u8fbe\u5f0f\u6765\u89e3\u6790\u8be5\u7279\u5f81":91,"\u5229\u7528\u8bad\u7ec3\u96c6\u751f\u6210\u7684\u5b57\u5178":93,"\u5229\u7528\u8fd9\u79cd\u7279\u6027":63,"\u5229\u7528\u903b\u8f91\u56de\u5f52\u6a21\u578b\u5bf9\u8be5\u5411\u91cf\u8fdb\u884c\u5206\u7c7b":89,"\u5229\u7528kubernetes\u80fd\u65b9\u4fbf\u5730\u7ba1\u7406\u8de8\u673a\u5668\u8fd0\u884c\u5bb9\u5668\u5316\u7684\u5e94\u7528":78,"\u5229\u843d":61,"\u522b\u4eba\u5e2e\u4e86\u5fd9":66,"\u522b\u5fd8\u4e86":65,"\u5230":[20,46,64],"\u5230\u6240\u6709\u8282\u70b9\u800c\u4e0d\u7528\u5bc6\u7801":73,"\u5230\u672c\u5730":66,"\u5230\u6b64":[54,68],"\u5230\u76ee\u524d\u4e3a\u6b62":92,"\u5236\u4f5c\u65b0\u955c\u50cf\u6765\u5b8c\u6210\u4ee5\u4e0a\u7684\u5de5\u4f5c":80,"\u5236\u4f5cpaddlepaddle\u5f00\u53d1\u955c\u50cf":54,"\u5236\u4f5cpaddlepaddle\u751f\u4ea7\u955c\u50cf":54,"\u5236\u4f5cpaddlepaddle\u955c\u50cf":80,"\u5237\u7259":61,"\u524d\u4e00\u7bc7\u6587\u7ae0\u4ecb\u7ecd\u4e86\u5982\u4f55\u5728kubernetes\u96c6\u7fa4\u4e0a\u542f\u52a8\u4e00\u4e2a\u5355\u673apaddlepaddle\u8bad\u7ec3\u4f5c\u4e1a":80,"\u524d\u4e09\u884cimport\u4e86\u5b9a\u4e49network":94,"\u524d\u53f0":61,"\u524d\u5411\u4f20\u64ad":67,"\u524d\u5411\u4f20\u64ad\u7ed9\u5b9a\u8f93\u5165":67,"\u524d\u5411\u548c\u540e\u5411":67,"\u524d\u5411op":68,"\u524d\u5411op\u5355\u5143\u6d4b\u8bd5\u7ee7\u627f\u81ea":68,"\u524d\u5411op\u5b9e\u73b0\u5b8c\u6210":68,"\u524d\u5411operator\u5355\u6d4b":68,"\u524d\u8005\u5728":49,"\u524d\u8005\u5b58\u50a8op\u7684\u8f93\u5165\u8f93\u51fa\u548c\u53c2\u6570\u5c5e\u6027":68,"\u524d\u8005\u622a\u65ad\u53ef\u5b66\u4e60\u53c2\u6570\u7684\u68af\u5ea6":49,"\u524d\u8005op\u7684\u5b9a\u4e49\u7ee7\u627f\u81ea":68,"\u524d\u81ea\u52a8\u68c0\u67e5\u4e00\u4e9b\u57fa\u672c\u4e8b\u5b9c":66,"\u5269\u4e0b\u7684pass\u4f1a\u76f4\u63a5\u4ece\u5185\u5b58\u91cc":2,"\u529f\u80fd":28,"\u529f\u80fd\u7684\u6b63\u786e\u6027\u5305\u62ec\u9a8c\u8bc1paddlepaddle\u76ee\u524d\u7684":42,"\u52a0\u4e0a\u504f\u7f6e\u5411\u91cf":67,"\u52a0\u4e86l2\u6b63\u5219\u548c\u68af\u5ea6\u622a\u65ad":89,"\u52a0\u5165":72,"\u52a0\u6743\u548c\u7528\u6765\u751f\u6210":64,"\u52a0\u6743\u7f16\u7801\u5411\u91cf":64,"\u52a0\u8f7d\u5177\u4f53\u7f51\u7edc\u53c2\u6570":51,"\u52a0\u8f7d\u6570\u636e":92,"\u52a0\u8f7d\u6a21\u578b":92,"\u52a0\u8f7d\u6a21\u578b\u53c2\u6570":94,"\u52a0\u8f7d\u9884\u8bad\u7ec3\u53c2\u6570":51,"\u52a0\u8f7dtest":75,"\u52a0\u901fpaddlepaddle\u8bad\u7ec3\u53ef\u4ee5\u8003\u8651\u4ece\u4ee5\u4e0b\u51e0\u4e2a\u65b9\u9762":49,"\u52a8\u4f5c\u7247":90,"\u52a8\u6001\u5e93":33,"\u52a8\u753b\u7247":90,"\u52a8\u8bcd":92,"\u52a9\u624b":67,"\u5305\u542b12":93,"\u5305\u542b20\u4e2a\u8bad\u7ec3\u6837\u4f8b":85,"\u5305\u542b3\u4e2a\u5c5e\u6027":85,"\u5305\u542b50":93,"\u5305\u542b\u4e86\u67d0\u79cd\u7c7b\u578b\u7684\u7c7b\u578b\u5b9a\u4e49\u548c\u66b4\u9732\u7684\u5168\u90e8\u51fd\u6570":34,"\u5305\u542b\u6240\u6709\u548c\u524d\u7aef\u7f51\u9875\u8bbe\u8ba1\u76f8\u5173\u7684\u6587\u4ef6":70,"\u5305\u542b\u7684\u6240\u6709\u4f9d\u8d56\u5047\u8bbe\u90fd\u53ef\u4ee5\u5728paddle\u7684\u8fd0\u884c\u5bb9\u5668\u4e2d":54,"\u5305\u542bkernel\u7684op\u548c\u4e0d\u5305\u542bkernel\u7684op":68,"\u5305\u62ec":[21,75,89,92,94],"\u5305\u62ec\u4ee5\u4e0b\u4e24\u79cd":2,"\u5305\u62ec\u53d1\u884c\u65f6\u95f4":90,"\u5305\u62ec\u5b57\u7b26\u4e32\u5206\u914d":49,"\u5305\u62ec\u6570\u636e\u8f93\u5165":52,"\u5305\u62ec\u6743\u91cdw\u548c\u504f\u7f6eb":20,"\u5305\u62ec\u751f\u6210cpu":53,"\u5305\u62ec\u795e\u7ecf\u7f51\u7edc\u62d3\u6251\u7ed3\u6784":56,"\u5305\u62ec\u7b80\u5355\u7684":89,"\u5305\u62ecbool":77,"\u5305\u62eclinux":54,"\u5305\u7684\u65b9\u6cd5\u662f":46,"\u533a\u522b\u662f\u540c\u65f6\u5904\u7406\u4e86\u4e24\u4e2a\u8f93\u5165":61,"\u533a\u522b\u662frnn\u4f7f\u7528\u4e24\u5c42\u5e8f\u5217\u6a21\u578b":61,"\u533b\u751f":90,"\u533b\u7597\u4fdd\u5065":90,"\u5341\u4e00":61,"\u5347\u5e8f\u6392\u5217":94,"\u534e\u6da6\u4e07\u5bb6":61,"\u534f\u540c\u5b8c\u6210releas":42,"\u5355\u4e2a\u503c":21,"\u5355\u4f4d\u662fmb":75,"\u5355\u5143\u6d4b\u8bd5":[65,69],"\u5355\u5143\u6d4b\u8bd5\u4f1a\u5f15\u7528site":46,"\u5355\u5143\u6d4b\u8bd5\u4f1a\u88ab\u81ea\u52a8\u52a0\u5165\u5de5\u7a0b\u8fdb\u884c\u7f16\u8bd1":68,"\u5355\u5143\u6d4b\u8bd5\u5728\u5185\u7684\u6240\u6709\u5355\u5143\u6d4b\u8bd5":65,"\u5355\u5143\u6d4b\u8bd5checkgrad_ep":74,"\u5355\u53cc\u5c42\u5e8f\u5217\u7684\u53e5\u5b50\u662f\u4e00\u6837\u7684":61,"\u5355\u53cc\u5c42rnn":62,"\u5355\u53d8\u91cf\u7684\u7ebf\u6027\u56de\u5f52":52,"\u5355\u5c42":63,"\u5355\u5c42\u4e0d\u7b49\u957frnn":61,"\u5355\u5c42\u548c\u53cc\u5c42\u5e8f\u5217\u7684\u4f7f\u7528\u548c\u793a\u4f8b2\u4e2d\u7684\u793a\u4f8b\u7c7b\u4f3c":61,"\u5355\u5c42\u5e8f\u5217":60,"\u5355\u5c42\u5e8f\u5217\u7684\u6bcf\u4e2a\u5143\u7d20":60,"\u5355\u5c42\u5e8f\u5217\u7b2ci\u4e2a\u5143\u7d20":60,"\u5355\u5c42\u6216\u53cc\u5c42":60,"\u5355\u5c42\u65f6\u95f4\u5e8f\u5217":61,"\u5355\u5c42rnn":[61,63],"\u5355\u5c42rnn\u548c\u53cc\u5c42rnn\u7684\u7f51\u7edc\u914d\u7f6e":61,"\u5355\u673acpu\u8bad\u7ec3":49,"\u5355\u673agpu\u8bad\u7ec3":49,"\u5355\u6b65\u51fd\u6570":64,"\u5355\u6b65\u51fd\u6570\u548c\u8f93\u51fa\u51fd\u6570\u5728":64,"\u5355\u6b65\u51fd\u6570\u548c\u8f93\u51fa\u51fd\u6570\u90fd\u975e\u5e38\u7b80\u5355":64,"\u5355\u6b65\u51fd\u6570\u7684\u5b9e\u73b0\u5982\u4e0b\u6240\u793a":64,"\u5355\u6d4b\u5305\u62ec\u5bf9\u6bd4\u524d\u5411op\u4e0d\u540c\u8bbe\u5907":68,"\u5355\u70b9\u6545\u969c":20,"\u5355\u8fdb\u5355\u51fa":63,"\u5360\u4f4d\u7b26":85,"\u536b\u751f":61,"\u5373":[34,49,50,52,68,70,80,89,93],"\u5373\u4e00\u4e2a\u5c06\u5355\u8bcd\u5b57\u7b26\u4e32\u6620\u5c04\u5230\u5355\u8bcdid\u7684\u5b57\u5178":2,"\u5373\u4e0a\u8ff0\u4ee3\u7801\u4e2d\u7684\u7b2c19\u884c":61,"\u5373\u4e0d\u5141\u8bb8\u5728":68,"\u5373\u4e0d\u8981\u5c06\u6bcf\u4e00\u4e2a\u6837\u672c\u90fd\u653e\u5165train":2,"\u5373\u4e0d\u9700\u8981\u4f7f\u7528memori":61,"\u5373\u4e3a\u4e00\u4e2a\u65f6\u95f4\u6b65":61,"\u5373\u4e3a\u5355\u5c42rnn\u5e8f\u5217\u7684\u4f7f\u7528\u4ee3\u7801":61,"\u5373\u4e3a\u65f6\u95f4\u5e8f\u5217\u7684\u8f93\u5165":61,"\u5373\u4e3a\u8fd9\u4e2a\u53cc\u5c42rnn\u7684\u7f51\u7edc\u7ed3\u6784":61,"\u5373\u4e3a\u8fd9\u4e2a\u6570\u636e\u6587\u4ef6\u7684\u540d\u5b57":2,"\u5373\u4e8c\u7ef4\u6570\u7ec4":61,"\u5373\u4f7f\u7528":[34,50],"\u5373\u4f7f\u7528\u6237\u76f4\u63a5\u5f15\u7528\u67d0\u79cd\u7c7b\u578b\u7684\u5934\u6587\u4ef6":34,"\u5373\u4f7f\u95f4\u9694\u5f88\u5c0f":75,"\u5373\u4f7fc":34,"\u5373\u4f7fprocess\u51fd\u6570\u91cc\u9762\u53ea\u6709\u4e00\u4e2ayield":2,"\u5373\u4f8b\u5982":34,"\u5373\u4fbf\u662f":65,"\u5373\u4fbf\u8bbe\u7f6e":46,"\u5373\u4fbfpaddl":34,"\u5373\u521d\u59cb\u72b6\u6001\u4e3a0":63,"\u5373\u5305\u542b\u65f6\u95f4\u6b65\u4fe1\u606f":2,"\u5373\u5355\u65f6\u95f4\u6b65\u6267\u884c\u7684\u51fd\u6570":64,"\u5373\u53cc\u5411lstm\u548c\u4e09\u5c42\u5806\u53e0lstm":93,"\u5373\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217":61,"\u5373\u53cc\u5c42rnn\u7684\u6bcf\u4e2a\u72b6\u6001":63,"\u5373\u53ef":[52,66],"\u5373\u53ef\u4ee5\u6781\u5927\u7684\u52a0\u901f\u6570\u636e\u8f7d\u5165\u6d41\u7a0b":49,"\u5373\u5728\u53cc\u5c42\u5e8f\u5217\u7684\u539f\u59cb\u6570\u636e\u4e2d":61,"\u5373\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d":49,"\u5373\u5927\u90e8\u5206\u503c\u4e3a0":[2,56],"\u5373\u5b8c\u6210\u67d0\u4e00\u4e2a\u4efb\u52a1\u7684\u6700\u5c11\u51fd\u6570":34,"\u5373\u5bf9\u7b2c\u4e09\u6b65\u8fdb\u884c\u66ff\u6362":89,"\u5373\u5c06\u4e00\u6bb5\u82f1\u6587\u6587\u672c\u6570\u636e":2,"\u5373\u5c06\u4e00\u6bb5\u8bdd\u8fdb\u884c\u5206\u7c7b":61,"\u5373\u5f53\u524d\u65f6\u95f4\u6b65\u4e0b\u7684\u795e\u7ecf\u7f51\u7edc\u4f9d\u8d56\u524d\u4e00\u4e2a\u65f6\u95f4\u6b65\u795e\u7ecf\u7f51\u7edc\u4e2d\u67d0\u4e00\u4e2a\u795e\u7ecf\u5143\u8f93\u51fa":61,"\u5373\u6211\u4eec\u53ef\u4ee5\u5148\u5b9a\u4e49\u4e00\u4e2atensor":69,"\u5373\u6211\u4eec\u7684\u8bad\u7ec3\u76ee\u6807":52,"\u5373\u628a\u5355\u5c42rnn\u751f\u6210\u540e\u7684subseq\u7ed9\u62fc\u63a5\u6210\u4e00\u4e2a\u65b0\u7684\u53cc\u5c42seq":63,"\u5373\u6574\u4e2a\u53cc\u5c42group\u662f\u5c06\u524d\u4e00\u4e2a\u5b50\u53e5\u7684\u6700\u540e\u4e00\u4e2a\u5411\u91cf":61,"\u5373\u6574\u4e2a\u8f93\u5165\u5e8f\u5217":60,"\u5373\u6574\u6570\u6570\u7ec4":61,"\u5373\u65f6\u95f4\u9012\u5f52\u795e\u7ecf\u7f51\u7edc":61,"\u5373\u662f\u8de8\u8d8a\u65f6\u95f4\u6b65\u7684\u7f51\u7edc\u8fde\u63a5":61,"\u5373\u66b4\u9732":34,"\u5373\u6b63\u9762\u548c\u8d1f\u9762":93,"\u5373\u6b63\u9762\u8bc4\u4ef7\u6807\u7b7e\u548c\u8d1f\u9762\u8bc4\u4ef7\u6807\u7b7e":93,"\u5373\u7279\u5f81\u7684\u6570\u7ec4":61,"\u5373\u7f51\u5361\u540d":80,"\u5373\u82e5\u5e72\u6570\u636e\u6587\u4ef6\u8def\u5f84\u7684\u67d0\u4e00\u4e2a":2,"\u5373\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u51fa\u73b0nan\u6216\u8005inf":49,"\u5373\u8bbe\u7f6e":49,"\u5373\u8fd9\u4e2a\u52a8\u6001\u5e93\u662f\u4e0d\u4f9d\u8d56\u4e8e\u5176\u4ed6\u4efb\u4f55\u6587\u4ef6\u7684":33,"\u5373define_py_data_sources2\u5e94\u6539\u4e3a":51,"\u5373input":63,"\u5373rnn\u4e4b\u95f4\u6709\u4e00\u6b21\u5d4c\u5957\u5173\u7cfb":61,"\u5377\u79ef\u5c42":86,"\u5377\u79ef\u5c42\u6743\u91cd":87,"\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u53ef\u4ee5\u8fa8\u8bc6\u56fe\u7247\u4e2d\u7684\u4e3b\u4f53":86,"\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u5728\u56fe\u7247\u5206\u7c7b\u4e0a\u6709\u7740\u60ca\u4eba\u7684\u6027\u80fd":86,"\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u662f\u4e00\u79cd\u4f7f\u7528\u5377\u79ef\u5c42\u7684\u524d\u5411\u795e\u7ecf\u7f51\u7edc":86,"\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\u80fd\u591f\u5f88\u597d\u7684\u8868\u793a\u8fd9\u4e24\u7c7b\u4fe1\u606f":86,"\u5377\u79ef\u7f51\u7edc\u662f\u4e00\u79cd\u7279\u6b8a\u7684\u4ece\u8bcd\u5411\u91cf\u8868\u793a\u5230\u53e5\u5b50\u8868\u793a\u7684\u65b9\u6cd5":89,"\u5378\u8f7dpaddlepaddle\u5305":46,"\u538b\u6241\u6210\u4e3aeigen\u7684\u4e00\u7ef4tensor":69,"\u538b\u7f29\u6210\u4e00\u4e2a\u5411\u91cf":61,"\u539f\u56e0":66,"\u539f\u56e0\u5728\u4e8e\u6ca1\u6709\u628a\u673a\u5668\u4e0acuda\u76f8\u5173\u7684\u9a71\u52a8\u548c\u5e93\u6620\u5c04\u5230\u5bb9\u5668\u5185\u90e8":46,"\u539f\u56e0\u662f\u6bcf\u4e2a\u56de\u590d\u90fd\u4f1a\u53d1\u9001\u4e00\u5c01\u90ae\u4ef6":66,"\u53bb\u8fc7":61,"\u53c2\u6570":[2,6,7,8,9,10,11,12,14,33,49,65,67,74,80,85,87,93],"\u53c2\u6570\u4e3a":68,"\u53c2\u6570\u5171\u4eab\u7684\u914d\u7f6e\u793a\u4f8b\u4e3a":51,"\u53c2\u6570\u521d\u59cb\u5316\u8def\u5f84":92,"\u53c2\u6570\u5373\u53ef":93,"\u53c2\u6570\u540d":87,"\u53c2\u6570\u6570\u91cf":89,"\u53c2\u6570\u670d\u52a1\u5668":74,"\u53c2\u6570\u670d\u52a1\u5668\u7684\u53c2\u6570\u5206\u5757\u5927\u5c0f":75,"\u53c2\u6570\u670d\u52a1\u5668\u7684\u76d1\u542c\u7aef\u53e3":75,"\u53c2\u6570\u670d\u52a1\u5668\u7684\u7f51\u7edc\u8bbe\u5907\u540d\u79f0":75,"\u53c2\u6570\u670d\u52a1\u5668\u7684ip\u5730\u5740":75,"\u53c2\u6570\u670d\u52a1\u5668\u7a00\u758f\u66f4\u65b0\u7684\u53c2\u6570\u5206\u5757\u5927\u5c0f":75,"\u53c2\u6570\u6765\u63a7\u5236\u7f13\u5b58\u65b9\u6cd5":49,"\u53c2\u6570\u6982\u8ff0":76,"\u53c2\u6570\u7684\u89e3\u6790":80,"\u53c2\u6570\u7ef4\u5ea6":85,"\u53c2\u6570\u884c":85,"\u53c2\u6570\u8bbe\u7f6e":48,"\u53c2\u6570\u8bbe\u7f6e\u4e86\u5916\u5c42":61,"\u53c2\u6570\u8f93\u5165":49,"\u53c2\u6570\u9700\u8981\u5b9e\u73b0":64,"\u53c2\u7167\u4e0a\u8ff0\u6b65\u9aa4\u66f4\u65b0":66,"\u53c2\u8003":[28,33,78],"\u53c2\u8003\u5f3a\u8c03\u90e8\u5206":72,"\u53c2\u8003\u6587\u732e":94,"\u53c2\u8003\u65f6\u95f4\u5e8f\u5217":61,"\u53c2\u8003\u955c\u50cf\u7684":80,"\u53c8":61,"\u53c8\u662f\u4e00\u4e2a\u5355\u5c42\u7684\u5e8f\u5217":60,"\u53c8\u8981\u4fdd\u8bc1\u6570\u636e\u662f\u968f\u673a\u7684":49,"\u53ca":67,"\u53cc\u5411\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u9690\u85cf\u72b6\u6001":64,"\u53cc\u5411\u9a8c\u8bc1":28,"\u53cc\u5c42":63,"\u53cc\u5c42\u4e0d\u7b49\u957frnn":61,"\u53cc\u5c42\u5e8f\u5217":60,"\u53cc\u5c42\u5e8f\u5217\u6216\u5355\u5c42\u5e8f\u5217":60,"\u53cc\u5c42\u5e8f\u5217\u6570\u636e\u4e00\u5171\u67094\u4e2a\u6837\u672c":61,"\u53cc\u5c42\u5e8f\u5217\u662f\u4e00\u4e2a\u5d4c\u5957\u7684\u5e8f\u5217":60,"\u53cc\u5c42\u5e8f\u5217\u662fpaddlepaddle\u652f\u6301\u7684\u4e00\u79cd\u975e\u5e38\u7075\u6d3b\u7684\u6570\u636e\u7ec4\u7ec7\u65b9\u5f0f":63,"\u53cc\u5c42\u5e8f\u5217\u6bcf\u4e2asubseq\u4e2d\u6bcf\u4e2a\u5143\u7d20":60,"\u53cc\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u53d8\u6210\u4e00\u4e2a0\u5c42\u5e8f\u5217":60,"\u53cc\u5c42\u6216\u8005\u5355\u5c42":60,"\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217\u7684dataprovider\u7684\u4ee3\u7801":61,"\u53cc\u5c42rnn":63,"\u53cc\u5c42rnn\u6570\u636e\u968f\u610f\u52a0\u4e86\u4e00\u4e9b\u9694\u65ad":61,"\u53cc\u5c42rnn\u987e\u540d\u601d\u4e49":61,"\u53cc\u8fdb\u5355\u51fa":63,"\u53cc\u8fdb\u53cc\u51fa":63,"\u53cd\u4e4b\u5219":92,"\u53cd\u5411\u4f20\u64ad":67,"\u53cd\u5411\u4f20\u64ad\u6839\u636e\u8f93\u51fa\u7684\u68af\u5ea6":67,"\u53cd\u5411\u5355\u5143\u6d4b\u8bd5\u51fd\u6570\u9700\u8981\u4ee5":68,"\u53cd\u5411op\u5355\u5143\u6d4b\u8bd5\u7ee7\u627f\u81ea":68,"\u53cd\u5411op\u7684\u68af\u5ea6\u6d4b\u8bd5":68,"\u53cd\u5411op\u7c7b":68,"\u53cd\u5411op\u7c7b\u7684\u5b9a\u4e49":68,"\u53cd\u5411operator\u5355\u6d4b":68,"\u53cd\u5411opkernel\u7684\u5b9a\u4e49\u4e0e\u524d\u5411op\u7c7b\u4f3c":68,"\u53d1\u5e03":54,"\u53d1\u5e03\u5230dockerhub":42,"\u53d1\u5e03\u5230github":42,"\u53d1\u6563\u5230\u4e86\u4e00\u4e2a\u6570\u503c\u7279\u522b\u5927\u7684\u5730\u65b9":49,"\u53d1\u884c\u548c\u7ef4\u62a4":66,"\u53d1\u9001\u53c2\u6570\u7684\u7aef\u53e3\u53f7":75,"\u53d6\u503c\u76f8\u540c\u7684layer":50,"\u53d6\u51b3\u4e8e":68,"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230cuda\u5de5\u5177\u94fe":53,"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230gtest":53,"\u53d6\u51b3\u4e8e\u662f\u5426\u5bfb\u627e\u5230swig":53,"\u53d8\u6362\u77e9\u9635":67,"\u53d8\u91cf\u6765\u8bbe\u7f6e\u5185\u5b58\u4e2d\u6682\u5b58\u7684\u6570\u636e\u6761":2,"\u53e3\u5934":61,"\u53e3\u7edf\u8ba1\u5b66\u4fe1\u606f\u7684\u7528\u6237\u624d\u88ab\u5305\u542b\u5728\u6570\u636e\u96c6\u4e2d":90,"\u53e5\u5b50":93,"\u53e5\u5b50\u4e2d\u7684\u7ec4\u5757\u5c06\u4f1a\u626e\u6f14\u67d0\u4e9b\u8bed\u4e49\u89d2\u8272":92,"\u53e5\u5b50\u8868\u793a\u7684\u8ba1\u7b97\u66f4\u65b0\u4e3a\u4e24\u6b65":89,"\u53e6\u4e00\u4e2a\u4f8b\u5b50\u662f\u901a\u8fc7\u5206\u6790\u6bcf\u65e5twitter\u535a\u5ba2\u7684\u6587\u672c\u5185\u5bb9\u6765\u9884\u6d4b\u80a1\u7968\u53d8\u52a8":93,"\u53e6\u4e00\u4e2a\u597d\u5904\u662f\u6211\u4eec\u53ef\u4ee5\u628apaddlepaddle\u5bb9\u5668\u8fd0\u884c\u5728\u8fdc\u7a0b\u670d\u52a1\u5668\u4e0a":54,"\u53e6\u4e00\u4e2a\u662f\u5185\u5b58\u64cd\u4f5c\u91cf":72,"\u53e6\u4e00\u4e2a\u662f\u521b\u5efadockerfile\u5e76\u8fd0\u884cdock":54,"\u53e6\u4e00\u4e2a\u662f\u6bcf\u6761\u5e8f\u5217":49,"\u53e6\u4e00\u4e2a\u7ec8\u7aef\u8fd0\u884cpython":54,"\u53e6\u4e00\u65b9\u9762":93,"\u53e6\u4e00\u79cd\u65b9\u5f0f\u662f\u5c06\u7f51\u7edc\u5c42\u5212\u5206\u5230\u4e0d\u540c\u7684gpu\u4e0a\u53bb\u8ba1\u7b97":77,"\u53e6\u5916":[61,65],"\u53e6\u5916\u4e24\u4e2a\u5206\u522b\u662f\u6ed1\u52a8\u5747\u503c\u548c\u65b9\u5dee":87,"\u53ea\u4f5c\u4e3aread":63,"\u53ea\u4fdd\u5b58\u6700\u540e\u4e00\u8f6e\u7684\u53c2\u6570":75,"\u53ea\u5141\u8bb8\u6574\u6570\u7684\u661f\u7ea7":90,"\u53ea\u5728\u7b2c\u4e00\u6b21cmake\u7684\u65f6\u5019\u6709\u6548":53,"\u53ea\u5bf9\u7279\u6b8a\u5728\u7ebf\u7cfb\u7edf\u8003\u8651\u4e24\u53f0\u4ee5\u4e0a\u540c\u65f6\u6545\u969c\u7684\u5bb9\u707e":20,"\u53ea\u622a\u53d6\u4e2d\u5fc3\u65b9\u5f62\u7684\u56fe\u50cf\u533a\u57df":87,"\u53ea\u662f\u53cc\u5c42\u5e8f\u5217\u5c06\u5176\u53c8\u505a\u4e86\u5b50\u5e8f\u5217\u5212\u5206":61,"\u53ea\u662f\u5c06\u53e5\u5b50\u7528\u8fde\u7eed\u5411\u91cf\u8868\u793a\u66ff\u6362\u4e3a\u7528\u7a00\u758f\u5411\u91cf\u8868\u793a":89,"\u53ea\u662f\u8bf4\u660e\u6570\u636e\u7684\u987a\u5e8f\u662f\u91cd\u8981\u7684":2,"\u53ea\u662f\u8bf7\u4e0d\u8981\u5fd8\u8bb0\u63d0\u524d\u5728\u7269\u7406\u673a\u4e0a\u5b89\u88c5gpu\u6700\u65b0\u9a71\u52a8":54,"\u53ea\u66b4\u9732\u6982\u5ff5\u7684\u63a5\u53e3":34,"\u53ea\u6709":61,"\u53ea\u67092\u4e2a\u914d\u7f6e\u4e0d\u4e00\u6837":85,"\u53ea\u6709\u542b\u6709\u4eba":90,"\u53ea\u6709\u5f53\u8bbe\u7f6e\u4e86spars":75,"\u53ea\u7528\u4e8e\u5728\u5e8f\u5217\u751f\u6210\u4efb\u52a1\u4e2d\u6307\u5b9a\u8f93\u5165\u6570\u636e":63,"\u53ea\u80fd\u5728recurrent_group\u4e2d\u4f5c\u4e3astep":50,"\u53ea\u80fd\u6d4b\u8bd5\u5355\u4e2a\u6a21\u578b":77,"\u53ea\u80fd\u8bbf\u95ee\u5b83\u4eec\u7684\u8f93\u51fa\u503c":50,"\u53ea\u80fd\u8c03\u7528paddle\u7684\u52a8\u6001\u5e93":33,"\u53ea\u8981\u4e00\u7cfb\u5217\u7279\u5f81\u6570\u636e\u4e2d\u7684":61,"\u53ea\u8981\u51fa\u73b0\u6d6e\u70b9\u6570\u5f02\u5e38":49,"\u53ea\u8981\u5728docker\u91cc\u542f\u52a8paddlepaddle\u7684\u65f6\u5019\u7ed9\u5b83\u4e00\u4e2a\u540d\u5b57":54,"\u53ea\u8bfbmemory\u8f93\u5165":63,"\u53ea\u9488\u5bf9\u5185\u5b58":49,"\u53ea\u9700\u4e2d\u65ad":73,"\u53ea\u9700\u4f7f\u7528":73,"\u53ea\u9700\u5220\u9664\u6700\u540e\u4e00\u884c\u4e2d\u7684\u6ce8\u91ca\u5e76\u628a":93,"\u53ea\u9700\u5728linux\u4e0b\u8fd0\u884c\u5982\u4e0b\u547d\u4ee4":94,"\u53ea\u9700\u7528\u4f60\u5b9a\u4e49\u7684\u76ee\u5f55\u4fee\u6539":73,"\u53ea\u9700\u77e5\u9053\u8fd9\u662f\u4e00\u4e2a\u6807\u8bb0\u5c5e\u6027\u7684\u65b9\u6cd5\u5c31\u53ef\u4ee5\u4e86":2,"\u53ea\u9700\u8981":64,"\u53ea\u9700\u8981\u4e00\u884c\u4ee3\u7801\u5c31\u53ef\u4ee5\u8c03\u7528\u8fd9\u4e2apydataprovider2":2,"\u53ea\u9700\u8981\u5728\u51fd\u6570\u4e2d\u8c03\u7528\u591a\u6b21yield\u5373\u53ef":2,"\u53ea\u9700\u8981\u6062\u590d\u8fd9\u53f0\u8282\u70b9":20,"\u53ea\u9700\u8981\u7b80\u5355\u5730\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4":94,"\u53ea\u9700\u8981\u7b80\u5355\u7684\u8fd0\u884c\u4e0b\u9762\u7684\u547d\u4ee4\u5373\u53ef":91,"\u53ea\u9700\u8981\u8fd0\u884c":91,"\u53ef\u4ee5":[61,66,73],"\u53ef\u4ee5\u4ee5\u540e\u53f0\u8fdb\u7a0b\u65b9\u5f0f\u8fd0\u884c\u5bb9\u5668":54,"\u53ef\u4ee5\u4f20\u5165\u4e00\u4e2a\u51fd\u6570":2,"\u53ef\u4ee5\u4f30\u8ba1\u51fa\u5982\u679c\u6a21\u578b\u91c7\u7528\u4e0d\u53d8\u7684\u8f93\u51fa\u6700\u5c0f\u7684cost0\u662f\u591a\u5c11":51,"\u53ef\u4ee5\u4f7f\u7528":51,"\u53ef\u4ee5\u4f7f\u7528\u5982\u4e0b\u4ee3\u7801":51,"\u53ef\u4ee5\u4f7f\u7528\u76f8\u5e94\u6570\u636e\u7c7b\u578b\u7684":51,"\u53ef\u4ee5\u4f7f\u7528\u8be5\u53c2\u6570":75,"\u53ef\u4ee5\u4f7f\u7528kubernetes\u7684\u547d\u4ee4\u884c\u5de5\u5177\u521b\u5efajob":80,"\u53ef\u4ee5\u4f7f\u7528python\u7684":4,"\u53ef\u4ee5\u51cf\u5c0f\u7cfb\u7edf\u590d\u6742\u6027":20,"\u53ef\u4ee5\u51cf\u5c11\u7f13\u5b58\u6c60\u7684\u5927\u5c0f":49,"\u53ef\u4ee5\u521b\u5efa\u4e00\u4e2a":79,"\u53ef\u4ee5\u521b\u5efa\u975e":68,"\u53ef\u4ee5\u53c2\u7167\u4e0b\u9762\u7684\u547d\u4ee4\u8fdb\u884c\u5b89\u88c5":86,"\u53ef\u4ee5\u53c2\u8003":[61,64,65,66,78,80,94],"\u53ef\u4ee5\u53c2\u8003\u4e0b\u9762\u7684\u6b65\u9aa4\u6392\u67e5":47,"\u53ef\u4ee5\u53c2\u8003\u4fdd\u5b58\u5728":85,"\u53ef\u4ee5\u53c2\u8003paddlepaddl":56,"\u53ef\u4ee5\u540c\u65f6\u5728cpu":69,"\u53ef\u4ee5\u544a\u8bc9\u60a8\u67d0\u4e2a\u64cd\u4f5c\u5230\u5e95\u82b1\u4e86\u591a\u957f\u65f6\u95f4":72,"\u53ef\u4ee5\u5728":54,"\u53ef\u4ee5\u5728\u4efb\u4f55\u673a\u5668\u4e0a\u6267\u884c\u7684":33,"\u53ef\u4ee5\u5728\u5171\u4eab\u5b58\u50a8\u4e0a\u67e5\u770b\u8f93\u51fa\u7684\u65e5\u5fd7\u548c\u6a21\u578b":80,"\u53ef\u4ee5\u5728\u5f88\u5927\u7a0b\u5ea6\u4e0a\u6d88\u9664\u6b67\u4e49":92,"\u53ef\u4ee5\u5728\u7f51\u7ad9\u4e0a\u627e\u5230":92,"\u53ef\u4ee5\u5728\u8fd9\u4e2a":66,"\u53ef\u4ee5\u5728kubernetes\u4e2d\u6309\u7167":78,"\u53ef\u4ee5\u5b9e\u73b0\u4ecepaddl":69,"\u53ef\u4ee5\u5c06\u67d0\u4e00\u4e2a\u51fd\u6570\u6807\u8bb0\u6210\u4e00\u4e2apydataprovider2":2,"\u53ef\u4ee5\u5c06\u78c1\u76d8\u4e0a\u67d0\u4e2a\u76ee\u5f55\u5171\u4eab\u7ed9\u7f51\u7edc\u4e2d\u5176\u4ed6\u673a\u5668\u8bbf\u95ee":78,"\u53ef\u4ee5\u5c06memory\u7406\u89e3\u4e3a\u4e00\u4e2a\u65f6\u5ef6\u64cd\u4f5c":63,"\u53ef\u4ee5\u5c06op\u5206\u4e3a\u4e24\u79cd":68,"\u53ef\u4ee5\u5e2e\u60a8\u63d0\u4f9b\u4e00\u4e9b\u5b9a\u4f4d\u6027\u80fd\u74f6\u9888\u7684\u5efa\u8bae":72,"\u53ef\u4ee5\u5e76\u884c\u7f16\u8bd1\u5417":65,"\u53ef\u4ee5\u5feb\u901f\u8fed\u4ee3":54,"\u53ef\u4ee5\u6267\u884c":[46,54],"\u53ef\u4ee5\u6267\u884c\u4ee5\u4e0b\u547d\u4ee4\u7f16\u8bd1\u751f\u6210\u6587\u6863":70,"\u53ef\u4ee5\u628a\u5b83\u60f3\u8c61\u4e3a\u4e00\u4e2a\u7c7b\u4f3c":65,"\u53ef\u4ee5\u628a\u672c\u5730\u7684\u6570\u636e\u4e0a\u4f20\u5230\u5b58\u50a8\u96c6\u7fa4\u4e2d":21,"\u53ef\u4ee5\u6307\u5b9a\u54ea\u4e00\u4e2a\u8f93\u5165\u548c\u8f93\u51fa\u5e8f\u5217\u4fe1\u606f\u4e00\u81f4":61,"\u53ef\u4ee5\u6309\u5982\u4e0b\u7684\u7ed3\u6784\u6765\u51c6\u5907\u6570\u6910":93,"\u53ef\u4ee5\u662f\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":[60,63],"\u53ef\u4ee5\u662f\u4e00\u4e2a\u975e\u5e8f\u5217":63,"\u53ef\u4ee5\u662f\u4ee5\u4e0b\u51e0\u79cd":67,"\u53ef\u4ee5\u663e\u793a\u5730\u6307\u5b9a\u4e00\u4e2alayer\u7684\u8f93\u51fa\u7528\u4e8e\u521d\u59cb\u5316memori":63,"\u53ef\u4ee5\u6709\u4ee5\u4e0b\u4e24\u79cd":63,"\u53ef\u4ee5\u6709\u6548\u51cf\u5c0f\u7f51\u7edc\u7684\u963b\u585e":75,"\u53ef\u4ee5\u6709\u6548\u7684\u907f\u514dparamet":20,"\u53ef\u4ee5\u67e5\u770b":80,"\u53ef\u4ee5\u67e5\u770b\u6b64pod\u8fd0\u884c\u7684\u5bbf\u4e3b\u673a":79,"\u53ef\u4ee5\u6d4b\u8bd5\u591a\u4e2a\u6a21\u578b":77,"\u53ef\u4ee5\u7528":[28,65],"\u53ef\u4ee5\u7528\u4e8e\u4ece\u5b98\u65b9\u7f51\u7ad9\u4e0a\u4e0b\u8f7dcifar":86,"\u53ef\u4ee5\u7528\u4e8e\u5c0f\u91cf\u6570\u636e\u7684\u9a8c\u8bc1":78,"\u53ef\u4ee5\u7528\u4e8e\u63a5\u6536\u548cpydataprovider2\u4e00\u6837\u7684\u8f93\u5165\u6570\u636e\u5e76\u8f6c\u6362\u6210\u9884\u6d4b\u63a5\u53e3\u6240\u9700\u7684\u6570\u636e\u7c7b\u578b":4,"\u53ef\u4ee5\u7528\u4ee5\u4e0b\u6307\u4ee4":21,"\u53ef\u4ee5\u7528\u5982\u4e0b\u547d\u4ee4":66,"\u53ef\u4ee5\u7528\u6765\u8ba1\u7b97cpu\u51fd\u6570\u6216cuda\u5185\u6838\u7684\u65f6\u95f4\u6d88\u8017":72,"\u53ef\u4ee5\u770b\u4f5c\u662f\u4e00\u4e2a\u975e\u5e8f\u5217\u8f93\u5165":60,"\u53ef\u4ee5\u7cbe\u786e\u8bf4\u660e\u4e00\u4e2a\u957f\u8017\u65f6\u64cd\u4f5c\u7684\u5177\u4f53\u539f\u56e0":72,"\u53ef\u4ee5\u7ee7\u7eed\u5728\u81ea\u5df1\u7684\u529f\u80fd\u5206\u652f\u63d0\u4ea4\u4ee3\u7801":42,"\u53ef\u4ee5\u7f16\u5199":54,"\u53ef\u4ee5\u8003\u8651\u4f7f\u7528\u4e00\u4e9b\u4f18\u5316\u7b97\u6cd5":49,"\u53ef\u4ee5\u8054\u7cfbop":47,"\u53ef\u4ee5\u8054\u7cfbop\u662f\u5426\u53ef\u4ee5\u66f4\u6362\u96c6\u7fa4\u6216\u5347\u7ea7\u5f53\u524d\u96c6\u7fa4":47,"\u53ef\u4ee5\u88c5\u7684\u662f":65,"\u53ef\u4ee5\u8bbe\u7f6e":[58,59,86],"\u53ef\u4ee5\u8bbf\u95ee\u7531recurr":50,"\u53ef\u4ee5\u8c03\u7528resize\u63a5\u53e3\u8fdb\u884c\u6539\u53d8":69,"\u53ef\u4ee5\u8f7b\u677e\u5730\u5b8c\u6210\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e":56,"\u53ef\u4ee5\u8fd0\u884c\u4e0b\u9762\u7684\u547d\u4ee4\u6765\u751f\u6210":91,"\u53ef\u4ee5\u8fd0\u884c\u811a\u672ctrain":86,"\u53ef\u4ee5\u9009\u62e9\u662f\u5426\u4f7f\u7528\u53c2\u6570":77,"\u53ef\u4ee5\u901a\u8fc7":[54,66,78],"\u53ef\u4ee5\u901a\u8fc7\u4fee\u6539\u8fd9\u4e24\u4e2a\u51fd\u6570\u6765\u5b9e\u73b0\u590d\u6742\u7684\u7f51\u7edc\u914d\u7f6e":64,"\u53ef\u4ee5\u901a\u8fc7\u8c03\u7528":4,"\u53ef\u4ee5\u901a\u8fc7\u9636\u6bb5\u6027\u7684\u4fdd\u5b58\u6bcf\u4e2aparamet":20,"\u53ef\u4ee5\u901a\u8fc7show_parameter_stats_period\u8bbe\u7f6e\u6253\u5370\u53c2\u6570\u4fe1\u606f\u7b49":89,"\u53ef\u4ee5\u91cd\u547d\u540d\u8fd9\u4e2awhl\u5305\u4e3a":46,"\u53ef\u7528\u4e8e\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u89e3\u6790\u8fd9\u4e9b\u53c2\u6570":77,"\u53ef\u7528\u5728\u6d4b\u8bd5\u6216\u8bad\u7ec3\u65f6\u6307\u5b9a\u521d\u59cb\u5316\u6a21\u578b":89,"\u53ef\u80fd\u4f1a\u5bfc\u81f4\u51fa\u9519":80,"\u53ef\u80fd\u4f1a\u9020\u6210\u7f51\u7edc\u62e5\u585e":20,"\u53ef\u80fd\u7684\u4ee3\u7801\u4e3a":49,"\u53ef\u80fd\u7684\u539f\u56e0\u662f":51,"\u53ef\u80fd\u7684\u60c5\u51b5\u4e0b":72,"\u53ef\u80fd\u9700\u8981\u6ce8\u610f\u7ed9\u8fd9\u4e2a\u865a\u62df\u673a\u591a\u5206\u914d\u4e00\u4e9b":65,"\u53ef\u89c1\u8be5\u8ba1\u7b97\u7531\u4e24\u4e2a\u8f93\u5165":68,"\u53ef\u8bbe\u7f6e":58,"\u53ef\u9009":[2,67],"\u53ef\u91c7\u7528\u7b2c\u4e8c\u79cd\u65b9\u5f0f":50,"\u53f3\u8fb9\u662f":87,"\u5403":61,"\u5403\u996d":61,"\u5404\u65b9\u9762":61,"\u5404\u9879\u53c2\u6570\u7684\u8be6\u7ec6\u8bf4\u660e\u53ef\u4ee5\u5728\u547d\u4ee4\u884c\u53c2\u6570\u76f8\u5173\u6587\u6863\u4e2d\u627e\u5230":86,"\u5404\u9879\u66f4\u52a0\u5177\u4f53\u7684\u5355\u5143\u6d4b\u8bd5\u5728":68,"\u5408":61,"\u5408\u5e76":94,"\u5408\u5e76\u6bcf\u4e2a":94,"\u5408\u7406":61,"\u540c\u65f6":[46,49,72],"\u540c\u65f6\u4e5f\u4f1a\u8bfb\u53d6\u76f8\u5173\u8def\u5f84\u53d8\u91cf\u6765\u8fdb\u884c\u641c\u7d22":53,"\u540c\u65f6\u4e5f\u53ef\u4ee5\u52a0\u901f\u5f00\u59cb\u8bad\u7ec3\u524d\u6570\u636e\u8f7d\u5165\u7684\u8fc7\u7a0b":49,"\u540c\u65f6\u4e5f\u53ef\u4ee5\u901a\u8fc7":66,"\u540c\u65f6\u4e5f\u80fd\u591f\u5f15\u5165\u66f4\u52a0\u590d\u6742\u7684\u8bb0\u5fc6\u673a\u5236":63,"\u540c\u65f6\u4f1a\u8ba1\u7b97\u5206\u7c7b\u51c6\u786e\u7387":89,"\u540c\u65f6\u4f60\u53ef\u4ee5\u4f7f\u7528":87,"\u540c\u65f6\u4f7f\u7528\u4e86l2\u6b63\u5219":89,"\u540c\u65f6\u5176\u5185\u90e8\u5b9e\u73b0\u53ef\u4ee5\u907f\u514d\u7eafcpu\u7248\u672cpaddlepaddle\u5728\u6267\u884c\u672c\u8bed\u53e5\u65f6\u53d1\u751f\u5d29\u6e83":72,"\u540c\u65f6\u518d\u5c06":42,"\u540c\u65f6\u53ef\u4ee5\u4f7f\u7528\u6237\u53ea\u5173\u6ce8\u5982\u4f55\u4ece\u6587\u4ef6\u4e2d\u8bfb\u53d6\u6bcf\u4e00\u6761\u6570\u636e":2,"\u540c\u65f6\u5728\u5185\u5b58\u91cc\u76f4\u63a5\u968f\u5373\u9009\u53d6\u6570\u636e\u6765\u505ashuffl":49,"\u540c\u65f6\u5c06\u53c2\u6570\u521d\u59cb\u5316\u4e3a":51,"\u540c\u65f6\u628a\u5f53\u524d\u76ee\u5f55":65,"\u540c\u65f6\u63d0\u8d77":42,"\u540c\u65f6\u7528\u6237\u9700\u8981\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u6307\u5b9a":77,"\u540c\u65f6\u8bbe\u7f6e\u5185\u5b58\u7f13\u5b58\u529f\u80fd":49,"\u540c\u65f6\u8bbe\u7f6e\u5b83\u7684input_types\u5c5e\u6027":2,"\u540c\u65f6\u8f93\u51fa\u5e8f\u5217\u5c42\u548c\u975e\u5e8f\u5217\u5c42":49,"\u540c\u65f6\u9884\u6d4b\u7f51\u7edc\u901a\u5e38\u76f4\u63a5\u8f93\u51fa\u6700\u540e\u4e00\u5c42\u7684\u7ed3\u679c\u800c\u4e0d\u662f\u50cf\u8bad\u7ec3\u7f51\u7edc\u4e00\u6837\u518d\u63a5\u4e00\u5c42cost":4,"\u540c\u6837":56,"\u540c\u6837\u4e5f\u53ef\u4ee5\u5728\u6d4b\u8bd5\u6a21\u5f0f\u4e2d\u6307\u5b9a\u6a21\u578b\u8def\u5f84":75,"\u540c\u6837\u53ef\u4ee5\u6267\u884c":54,"\u540c\u6837\u53ef\u4ee5\u6269\u5c55\u5230\u53cc\u5c42\u5e8f\u5217\u7684\u5904\u7406\u4e0a":63,"\u540c\u6837\u53ef\u83b7\u53d6\u5230\u8f93\u5165\u8f93\u51fa\u548c\u5c5e\u6027\u53c2\u6570":68,"\u540c\u6b65\u6267\u884c\u64cd\u4f5c\u7684\u7ebf\u7a0b\u6570":75,"\u540c\u7406":68,"\u540d\u5b57\u4fee\u9970":33,"\u540d\u79f0":89,"\u540e":[51,53,58,59,66,80,93],"\u540e\u5411\u4f20\u64ad":67,"\u540e\u5411\u4f20\u64ad\u7ed9\u5b9a\u8f93\u51fa\u7684\u68af\u5ea6":67,"\u540e\u8005\u5728\u6fc0\u6d3b\u51fd\u6570\u53cd\u5411\u8ba1\u7b97\u65f6\u88ab\u8c03\u7528":49,"\u540e\u8005\u622a\u65ad\u56de\u4f20\u7ed9\u524d\u5c42\u7684\u68af\u5ea6":49,"\u540e\u8005\u7528\u4e8e\u68c0\u67e5\u53c2\u6570\u5c5e\u6027\u7684\u5408\u6cd5\u6027":68,"\u540e\u8005\u7ee7\u627f\u81ea":68,"\u540e\u9762\u8fde\u5168\u8fde\u63a5\u5c42\u548csoftmax\u5c42":93,"\u5411\u6307\u5b9a\u7684\u76ee\u5f55\u4e2d\u4e00\u4e2a\u65b0\u7684\u6587\u4ef6":20,"\u5411\u91cfenable_parallel_vector":74,"\u5411paddlepaddle\u7684\u4e3b\u7248\u672c\u5e93\u63d0\u4ea4":42,"\u5417":65,"\u5426":53,"\u5426\u5219":[1,58,59,68,73,91],"\u5426\u5219\u4f1a\u628a":66,"\u5426\u5219\u4f60\u9700\u8981\u81ea\u5df1\u4e0b\u8f7d":94,"\u5426\u5219\u4f7f\u7528\u591a\u673a\u8bad\u7ec3":75,"\u5426\u5219\u4f7f\u7528cpu\u6a21\u5f0f":75,"\u5426\u5219\u4f7f\u7528gpu":77,"\u5426\u5219\u5b83\u4ee5\u4e00\u4e2a\u5e8f\u5217\u8f93\u5165":64,"\u5426\u5219\u5f97\u628apaddle\u9759\u6001\u5e93\u94fe\u63a5\u5230\u89e3\u91ca\u5668\u91cc":33,"\u5426\u5219\u9891\u7e41\u7684\u591a\u8282\u70b9\u5de5\u4f5c\u7a7a\u95f4\u90e8\u7f72\u53ef\u80fd\u4f1a\u5f88\u9ebb\u70e6":73,"\u5426\u5b9a":92,"\u542b\u4e49":[87,93],"\u542b\u6709":90,"\u542b\u6709\u5e8f\u5217\u4fe1\u606f\u548c\u5b50\u5e8f\u5217\u4fe1\u606f\u7684\u7a20\u5bc6\u5411\u91cf":67,"\u542b\u6709\u5e8f\u5217\u4fe1\u606f\u7684\u6574\u6570":67,"\u542b\u6709\u5e8f\u5217\u4fe1\u606f\u7684\u7a20\u5bc6\u5411\u91cf":67,"\u542f\u52a8\u4e00\u4e2a\u65b0\u7684\u7ebf\u7a0b\u5f00\u59cb\u4fdd\u5b58\u68c0\u67e5\u70b9":20,"\u542f\u52a8\u5bb9\u5668\u5f00\u59cb\u8bad\u7ec3":80,"\u542f\u52a8\u5e76\u884c\u5411\u91cf\u7684\u9608\u503c":75,"\u542f\u52a8\u5feb\u901f\u5e94\u7b54":75,"\u542f\u7528\u68af\u5ea6\u53c2\u6570\u7684\u9608\u503c":75,"\u5440":61,"\u544a\u8bc9paddle\u54ea\u4e2a\u6587\u4ef6\u662f\u795e\u7ecf\u7f51\u7edc\u7684\u914d\u7f6e\u6587\u4ef6":91,"\u544a\u8bc9paddle\u5c06\u6a21\u578b\u4fdd\u5b58\u5728":91,"\u5468\u56f4":61,"\u547d\u4ee4":73,"\u547d\u4ee4\u4e3a":[46,79],"\u547d\u4ee4\u521b\u5efa\u65b0\u955c\u50cf":79,"\u547d\u4ee4\u5220\u9664":[58,59],"\u547d\u4ee4\u53ef\u4ee5\u8bbe\u7f6e":53,"\u547d\u4ee4\u6307\u5b9a\u7684\u53c2\u6570\u4f1a\u4f20\u5165\u7f51\u7edc\u914d\u7f6e\u4e2d":89,"\u547d\u4ee4\u6709\u65f6\u5019\u4f1a\u4ea7\u751f\u4e00\u4e9b\u4e2d\u95f4\u7ed3\u679c":65,"\u547d\u4ee4\u770b\u5230\u505c\u6b62\u540e\u4f46\u662f\u6ca1\u6709\u5220\u9664\u7684":65,"\u547d\u4ee4\u7f16\u8bd1\u6e90\u7801\u5373\u53ef":65,"\u547d\u4ee4\u884c\u53c2\u6570\u6587\u6863":89,"\u547d\u4ee4\u8bbe\u7f6e\u8be5\u7c7b\u7f16\u8bd1\u9009\u9879":53,"\u547d\u4ee4\u9009\u9879\u5e76\u4e14":73,"\u547d\u4ee4\u91cc\u90fd\u7528\u4e86":65,"\u547d\u540d\u4e3a":66,"\u547d\u540d\u7a7a\u95f4":78,"\u547d\u540d\u7a7a\u95f4\u4e3b\u8981\u4e3a\u4e86\u5bf9\u8c61\u8fdb\u884c\u903b\u8f91\u4e0a\u7684\u5206\u7ec4\u4fbf\u4e8e\u7ba1\u7406":78,"\u547d\u540d\u89c4\u8303":68,"\u547d\u540d\u8bf7\u9075\u5b88":68,"\u548c":[21,33,34,42,49,50,51,52,53,58,59,61,64,65,66,67,68,69,70,72,73,77,78,85,86,89,91,94],"\u548c\u4e00\u4e2a\u5df2\u7ecf\u5206\u8bcd\u540e\u7684\u53e5\u5b50":61,"\u548c\u4e09\u79cd\u5e8f\u5217\u6a21\u5f0f":[2,56],"\u548c\u4e0b\u9762\u5c06\u8981\u4ecb\u7ecd\u7684\u6ce8\u518c\u51fd\u6570\u4e00\u8d77\u653e\u5728":68,"\u548c\u4e2d\u6587\u6587\u6863":70,"\u548c\u4e4b\u524d\u51cf\u5c0f\u901a\u8fc7\u51cf\u5c0f\u7f13\u5b58\u6c60\u6765\u51cf\u5c0f\u5185\u5b58\u5360\u7528\u7684\u539f\u7406\u4e00\u81f4":49,"\u548c\u504f\u7f6e\u5411\u91cf":67,"\u548c\u5185\u5b58":65,"\u548c\u533a\u57df\u6807\u8bb0":92,"\u548c\u53cc\u5c42\u5e8f\u5217\u542b\u6709subseq":60,"\u548c\u56fd\u5185\u955c\u50cf":54,"\u548c\u5728":2,"\u548c\u5bf9\u8c61\u5b58\u50a8api":78,"\u548c\u5dee\u8bc4":89,"\u548c\u5e8f\u5217\u4e2d\u542b\u6709\u5143\u7d20\u7684\u6570\u76ee\u540c":60,"\u548c\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u8f93\u5165":64,"\u548c\u64cd\u4f5c\u7cfb\u7edf\u4e0a\u76f4\u63a5\u8fd0\u884c\u7684":65,"\u548c\u68af\u5ea6\u622a\u65ad":89,"\u548c\u6a21\u578b\u8def\u5f84":93,"\u548c\u771f\u5b9e":52,"\u548c\u793a\u4f8b2\u4e2d\u7684\u914d\u7f6e\u7c7b\u4f3c":61,"\u548c\u79bb\u7ebf\u6570\u636e\u7684\u65b9\u5f0f":21,"\u548c\u7b2c6\u884c\u7684":94,"\u548c\u90e8\u5206layer":63,"\u548cadam\u5b66\u4e60\u65b9\u6cd5":94,"\u548cargument":92,"\u548cpool":60,"\u548cpython\u63a5\u53e3\u6765\u63d0\u53d6\u7279\u5f81":87,"\u54c1\u8d28":61,"\u54ea\u4e2atrainer\u5148\u5b8c\u6210block\u7684\u8bad\u7ec3":20,"\u54ea\u4e9b\u4e0d\u662f":61,"\u5546\u52a1":61,"\u554a":61,"\u559c\u5267\u7247":90,"\u5668":89,"\u56db\u79cd\u6570\u636e\u7c7b\u578b":[2,56],"\u56de\u5f52\u8bef\u5dee\u4ee3\u4ef7\u5c42":52,"\u56e0\u4e3a\u5168\u8fde\u63a5\u5c42\u7684\u6fc0\u6d3b\u53ef\u4ee5\u662fsoftmax":67,"\u56e0\u4e3a\u5355\u4e2a\u8c13\u8bcd\u4e0d\u80fd\u7cbe\u786e\u5730\u63cf\u8ff0\u8c13\u8bcd\u4fe1\u606f":92,"\u56e0\u4e3a\u53c2\u6570":77,"\u56e0\u4e3a\u5b83\u4eec\u7684\u8ba1\u7b97\u6548\u7387\u6bd4":64,"\u56e0\u4e3a\u5b83\u6bd4":64,"\u56e0\u4e3a\u5b98\u65b9\u955c\u50cf":80,"\u56e0\u4e3a\u5bb9\u5668\u5185\u7684\u6587\u4ef6\u90fd\u662f\u6682\u65f6\u5b58\u5728\u7684":78,"\u56e0\u4e3a\u6211\u4eec\u4f1a\u628a\u6240\u6709\u7f16\u8bd1\u5de5\u5177\u90fd\u5b89\u88c5\u8fdb\u4e00\u4e2a":65,"\u56e0\u4e3a\u6e90\u7801\u5c31\u5728\u672c\u673a\u4e0a":65,"\u56e0\u4e3a\u8be5\u6587\u4ef6\u53ef\u9002\u7528\u4e8e\u9884\u6d4b":86,"\u56e0\u4e3a\u8fd9\u6837\u505a\u4e5f\u6ca1\u6cd5\u4fdd\u8bc1\u6d88\u9664\u968f\u673a\u6027":20,"\u56e0\u4e3adocker\u80fd\u5728\u6240\u6709\u4e3b\u8981\u64cd\u4f5c\u7cfb\u7edf":54,"\u56e0\u4e3apython\u7684\u641c\u7d22\u8def\u5f84\u662f\u4f18\u5148\u5df2\u7ecf\u5b89\u88c5\u7684python\u5305":46,"\u56e0\u4e3aswig\u5728\u7b2c\u4e09\u65b9\u8bed\u8a00\u4e2d\u66b4\u9732\u7684\u51fd\u6570\u540d":33,"\u56e0\u6b64":[1,2,61,63,67,68],"\u56e0\u6b64\u4f7f\u7528":2,"\u56e0\u6b64\u53cc\u5c42\u5e8f\u5217\u7684\u914d\u7f6e\u4e2d":61,"\u56e0\u6b64\u53ef\u4ee5\u4f7f\u7528\u8be5\u9009\u9879":85,"\u56e0\u6b64\u53ef\u80fd\u4f1a\u6709\u4e00\u4e9b\u9519\u8bef\u548c\u4e0d\u4e00\u81f4\u53d1\u751f":90,"\u56e0\u6b64\u53ef\u80fd\u6d4b\u8bd5\u4e0d\u591f\u5b8c\u5907":69,"\u56e0\u6b64\u5728\u8f6c\u6362\u65f6\u9700\u8981\u663e\u793a\u7684\u6307\u5b9a":69,"\u56e0\u6b64\u5982\u679c\u8fd9\u4e2a\u811a\u672c\u8fd0\u884c\u5931\u8d25":86,"\u56e0\u6b64\u5b83\u662finteger_value_sub_sequ":61,"\u56e0\u6b64\u6211\u4eec\u91c7\u7528\u8f93\u51fa\u7684\u52a0\u6743\u548c":67,"\u56e0\u6b64\u6709\u4e24\u79cd\u89e3\u51b3\u65b9\u6848":2,"\u56e0\u6b64\u7528\u6237\u5e76\u4e0d\u9700\u8981\u5173\u5fc3\u5b83\u4eec":74,"\u56e0\u6b64\u8be5\u5c42\u4e2d\u6ca1\u6709\u504f\u7f6e":87,"\u56e0\u6b64\u9519\u8bef\u7684\u4f7f\u7528\u4e8c\u8fdb\u5236\u53d1\u884c\u7248\u53ef\u80fd\u4f1a\u5bfc\u81f4\u8fd9\u79cd\u9519\u8bef":46,"\u56e0\u6b64init_hook\u5c3d\u91cf\u4f7f\u7528":2,"\u56e2\u8d2d\u7f51\u7ad9":93,"\u56fe":[87,93],"\u56fe2\u662f\u53cc\u5411lstm\u7f51\u7edc":93,"\u56fe3\u662f\u4e09\u5c42lstm\u7ed3\u6784":93,"\u56fe\u50cf\u5206\u7c7b":[42,88],"\u56fe\u50cf\u5927\u5c0f\u4e3a3":87,"\u56fe\u50cf\u63cf\u8ff0":94,"\u56fe\u7247\u5206\u4e3a10\u7c7b":86,"\u56fe\u7684\u5e95\u90e8\u662fword":93,"\u56fe\u8868":[54,93],"\u5728":[2,34,42,60,61,64,65,66,68,73,87,89,90,92],"\u57280":54,"\u5728\u4e00\u4e2a\u4e0d\u53ef\u4e2d\u65ad\u5e76\u7f3a\u5c11\u5907\u4efd\u7684\u8bad\u7ec3\u4efb\u52a1\u4e2d":20,"\u5728\u4e00\u4e2a\u529f\u80fd\u9f50\u5168\u7684kubernetes\u673a\u7fa4\u91cc":79,"\u5728\u4e00\u4e2a\u53c2\u6570\u7684\u68af\u5ea6\u88ab\u66f4\u65b0\u540e":67,"\u5728\u4e00\u4e2a\u5468\u671f\u5185\u6d4b\u8bd5\u6240\u6709\u6570\u636e":92,"\u5728\u4e00\u8f6e\u4e2d\u6bcfsave":75,"\u5728\u4e0a\u56fe\u4e2d\u663e\u793a\u4e86\u5728\u4e00\u4e2a\u5b9e\u9645\u751f\u4ea7\u73af\u5883\u4e2d\u7684\u5e94\u7528":21,"\u5728\u4e0a\u9762\u4ee3\u7801\u4e2d":61,"\u5728\u4e0a\u9762\u7684\u4ee3\u7801\u4e2d":68,"\u5728\u4e0b\u4e00\u7bc7\u4e2d":79,"\u5728\u4e0b\u9762\u4f8b\u5b50\u91cc":89,"\u5728\u4e0b\u9762\u7684\u4f8b\u5b50\u4e2d":86,"\u5728\u4e0d\u540c\u64cd\u4f5c\u7cfb\u7edf":78,"\u5728\u4e4b\u540e\u7684":49,"\u5728\u4efb\u610f\u957f\u5ea6\u8bed\u53e5\u7ffb\u8bd1\u7684\u573a\u666f\u4e0b\u90fd\u53ef\u4ee5\u89c2\u5bdf\u5230\u5176\u6548\u679c\u7684\u63d0\u5347":94,"\u5728\u4f7f\u7528\u5b83\u4e4b\u524d\u8bf7\u5b89\u88c5paddlepaddle\u7684python":93,"\u5728\u4f7f\u7528\u8be5\u6587\u6863\u4e4b\u524d":56,"\u5728\u4f7f\u7528twine\u4e0a\u4f20\u4e4b\u524d":42,"\u5728\u5168\u8fde\u63a5\u5c42\u4e2d":67,"\u5728\u5177\u4f53\u7684\u8ba1\u7b97\u4e2d":69,"\u5728\u51c6\u5907\u53d1\u8d77":66,"\u5728\u51fa\u73b0\u5355\u70b9\u6545\u969c\u65f6":20,"\u5728\u51fd\u6570":80,"\u5728\u5206\u5e03\u5f0f\u73af\u5883\u4e2d\u6d4b\u8bd5":75,"\u5728\u5206\u5e03\u5f0f\u8bad\u7ec3\u4e2d":75,"\u5728\u521b\u5efaparameters\u540e":51,"\u5728\u5355\u5c42\u6570\u636e\u7684\u57fa\u7840\u4e0a":61,"\u5728\u53c2\u6570\u670d\u52a1\u5668\u4e0a\u52a0\u8f7d\u548c\u4fdd\u5b58\u53c2\u6570":75,"\u5728\u53c2\u6570\u670d\u52a1\u5668\u7ec8\u7aef\u6bcflog":75,"\u5728\u53cc\u5c42rnn\u4e2d\u7684\u7ecf\u5178\u60c5\u51b5\u662f\u5c06\u5185\u5c42\u7684\u6bcf\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217\u6570\u636e":61,"\u5728\u53cd\u5411\u4f20\u9012\u7684\u65f6\u5019":49,"\u5728\u53d8\u6362\u65f6\u9700\u8981\u5c06\u8f93\u5165\u5e8f\u5217\u4f20\u5165":61,"\u5728\u540c\u4e00\u4e2a\u547d\u540d\u7a7a\u95f4\u4e2d":78,"\u5728\u542f\u52a8job\u4e4b\u524d":80,"\u5728\u56de\u590d\u8bc4\u5ba1\u4eba\u610f\u89c1\u65f6":66,"\u5728\u58f0\u660edataprovider\u7684\u65f6\u5019\u4f20\u5165dictionary\u4f5c\u4e3a\u53c2\u6570":2,"\u5728\u591acpu\u8bad\u7ec3\u65f6\u5171\u4eab\u8be5\u53c2\u6570":75,"\u5728\u5b8c\u6210\u795e\u7ecf\u7f51\u7edc\u7684\u642d\u5efa\u4e4b\u540e":56,"\u5728\u5b9a\u4e49\u8f93\u5165layer\u4e4b\u540e":56,"\u5728\u5b9e\u73b0\u8fc7\u7a0b\u4e2d":34,"\u5728\u5b9e\u9645\u5e94\u7528\u4e2d":50,"\u5728\u5bb9\u5668\u521b\u5efa\u540e":80,"\u5728\u5bf9\u5bb9\u5668\u7684\u63cf\u8ff0":80,"\u5728\u5c42\u4e2d\u6307\u5b9a":77,"\u5728\u5e8f\u5217\u751f\u6210\u4efb\u52a1\u4e2d":63,"\u5728\u5f00\u59cb\u8bad\u7ec3\u4e4b\u524d":[21,86],"\u5728\u5f02\u6784\u96c6\u7fa4\u4e2d":20,"\u5728\u5f15\u5165\u5176\u4ed6\u7c7b\u578b\u7684\u5934\u6587\u4ef6\u65f6":34,"\u5728\u5f53\u524d\u7684\u5b9e\u73b0\u65b9\u5f0f\u4e0b":67,"\u5728\u5f97\u5230":80,"\u5728\u5feb\u7167\u5199\u5165\u5b8c\u6210\u540e":20,"\u5728\u60a8\u7684\u5b9e\u9645\u73af\u5883\u4e2d":20,"\u5728\u6211\u4eec\u7684\u4f8b\u5b50\u4e2d":64,"\u5728\u6211\u4eec\u7684\u6d4b\u8bd5\u4e2d":93,"\u5728\u6267\u884c\u65f6":69,"\u5728\u63d0\u4ea4":66,"\u5728\u642d\u5efa\u795e\u7ecf\u7f51\u7edc\u7684\u8fc7\u7a0b\u4e2d":56,"\u5728\u6570\u636e\u52a0\u8f7d\u548c\u7f51\u7edc\u914d\u7f6e\u5b8c\u6210\u4e4b\u540e":89,"\u5728\u6587\u4ef6":91,"\u5728\u6709\u65b0\u7684\u5355\u8bcd\u6765\u4e34\u7684\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u9aa4\u5185":93,"\u5728\u672c\u4f8b\u4e2d":[61,66,77],"\u5728\u672c\u4f8b\u4e2d\u6ca1\u6709\u4f7f\u7528":2,"\u5728\u672c\u6559\u7a0b\u4e2d":[64,86],"\u5728\u672c\u6587\u4e2d":73,"\u5728\u672c\u6587\u4e2d\u4f7f\u7528\u7684":73,"\u5728\u672c\u6587\u6863\u4e2d":28,"\u5728\u672c\u6f14\u793a\u4e2d":93,"\u5728\u672c\u793a\u4f8b\u4e2d":[61,93],"\u5728\u672c\u8282\u4e2d":64,"\u5728\u673a\u7fa4\u4e0a\u8fd0\u884c\u8f6c\u6362\u7a0b\u5e8f":21,"\u5728\u6811\u7684\u6bcf\u4e00\u5c42\u4e0a":75,"\u5728\u6837\u4f8b\u4e2d":34,"\u5728\u6a21\u578b\u6587\u4ef6\u7684":73,"\u5728\u6a21\u578b\u914d\u7f6e\u4e2d\u901a\u8fc7":89,"\u5728\u6b64":[74,77],"\u5728\u6b64\u4e3a\u65b9\u4fbf\u5bf9\u6bd4\u4e0d\u540c\u7f51\u7edc\u7ed3\u6784":89,"\u5728\u6b64\u611f\u8c22":85,"\u5728\u6bcf\u4e2a\u65f6\u95f4\u6b65\u4e2d":64,"\u5728\u6bcf\u4e2a\u65f6\u95f4\u6b65\u7684\u5b50\u5e8f\u5217\u957f\u5ea6\u53ef\u4ee5\u4e0d\u76f8\u7b49":61,"\u5728\u6bcf\u4e2a\u65f6\u95f4\u6b65\u957f":64,"\u5728\u6bcf\u4e2apod\u4e0a\u90fd\u901a\u8fc7volume\u65b9\u5f0f\u6302\u8f7d\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u7684\u4e00\u4e2a\u76ee\u5f55\u7528\u4e8e\u4fdd\u5b58\u8bad\u7ec3\u6570\u636e\u548c\u8f93\u51fa\u7ed3\u679c":80,"\u5728\u6bcf\u8bad\u7ec3":91,"\u5728\u6d4b\u8bd5\u9636\u6bb5":75,"\u5728\u6d4b\u8bd5\u9636\u6bb5\u5b83\u4eec\u5c06\u4f1a\u88ab\u52a0\u8f7d\u5230\u6a21\u578b\u4e2d":87,"\u5728\u6e90\u7801\u76ee\u5f55\u6811\u7684\u6839\u76ee\u5f55\u4e2d\u8fd0\u884c":66,"\u5728\u6f14\u793a\u4e2d":92,"\u5728\u7269\u7406\u673a\u4e0a\u624b\u52a8\u90e8\u7f72":78,"\u5728\u751f\u6210\u65f6":64,"\u5728\u751f\u6210\u8fc7\u7a0b\u4e2d":94,"\u5728\u751f\u6210\u8fc7\u7a0b\u4e2d\u6211\u4eec\u4f7f\u7528sgd\u8bad\u7ec3\u7b97\u6cd5":94,"\u5728\u7528\u6237\u4f7f\u7528c":34,"\u5728\u7528\u6237\u6587\u4ef6user":91,"\u5728\u7528\u6237\u719f\u6089\u7684\u5f00\u53d1\u5e73\u53f0":58,"\u5728\u7535\u5f71\u6587\u4ef6movi":91,"\u5728\u76f8\u5e94\u7684\u4f18\u5316\u7b97\u6cd5\u91cc\u8bbe\u7f6elearning_rate_schedule\u53ca\u76f8\u5173\u53c2\u6570":51,"\u5728\u76f8\u5e94layer\u7684":50,"\u5728\u793a\u4f8b\u4e2d\u6211\u4eec\u4f7f\u7528attention\u7248\u672c\u7684gru\u7f16\u89e3\u7801\u7f51\u7edc":94,"\u5728\u793a\u4f8b\u4e2d\u6211\u4eec\u4f7f\u7528sgd\u8bad\u7ec3\u7b97\u6cd5":94,"\u5728\u793a\u4f8b\u4e2d\u6211\u4eec\u5b9a\u4e49\u4e86\u4e00\u4e2a\u5e8f\u5217\u5230\u5e8f\u5217\u7684\u751f\u6210\u6570\u636e":94,"\u5728\u793a\u4f8b\u4e2d\u6211\u4eec\u5b9a\u4e49\u4e86\u4e00\u4e2a\u5e8f\u5217\u5230\u5e8f\u5217\u7684\u8bad\u7ec3\u548c\u6d4b\u8bd5\u6570\u636e":94,"\u5728\u7a0b\u5e8f\u5f00\u59cb\u9636\u6bb5":4,"\u5728\u7b2c\u4e00\u884c\u4e2d\u6211\u4eec\u8f7d\u5165\u7528\u4e8e\u5b9a\u4e49\u7f51\u7edc\u7684\u51fd\u6570":86,"\u5728\u7ebf\u6a21\u578b\u9884\u6d4b\u670d\u52a1":21,"\u5728\u7ec4\u5408\u65f6":56,"\u5728\u7f16\u8bd1\u5bbf\u4e3b\u673a\u7248protoc\u53ef\u6267\u884c\u6587\u4ef6\u548c\u76ee\u6807\u673a\u7248openblas\u5e93\u65f6\u9700\u8981\u7528\u5230":[58,59],"\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d":67,"\u5728\u7f51\u7edc\u914d\u7f6e\u91cc":2,"\u5728\u7ffb\u8bd1\u6cd5\u8bed\u53e5\u5b50\u4e4b\u524d":94,"\u5728\u811a\u672c":91,"\u5728\u81ea\u7136\u8bed\u8a00\u5904\u7406\u4efb\u52a1\u4e2d":60,"\u5728\u8bad\u7ec3\u4e2d":64,"\u5728\u8bad\u7ec3\u4e4b\u524d":80,"\u5728\u8bad\u7ec3\u4e86":91,"\u5728\u8bad\u7ec3\u4e86\u51e0\u4e2a\u8f6e\u6b21\u4ee5\u540e":91,"\u5728\u8bad\u7ec3\u5b8c\u6210\u540e":86,"\u5728\u8bad\u7ec3\u6570\u96c6\u4e0a\u8bad\u7ec3\u751f\u6210\u8bcd\u5411\u91cf\u5b57\u5178":85,"\u5728\u8bad\u7ec3\u65f6":79,"\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d":[80,94],"\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u6bcfshow":75,"\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u8fdb\u884c\u6d4b\u8bd5":1,"\u5728\u8bbe\u7f6e":[58,59],"\u5728\u8bc4\u5ba1\u8fc7\u7a0b\u4e2d":42,"\u5728\u8be5\u793a\u4f8b\u4e2d":51,"\u5728\u8be5\u914d\u7f6e\u76847":61,"\u5728\u8bed\u8a00\u751f\u6210\u9886\u57df\u4e2d":94,"\u5728\u8d2d\u7269\u7f51\u7ad9\u4e0a":89,"\u5728\u8f6f\u4ef6\u5de5\u7a0b\u7684\u8303\u7574\u91cc":72,"\u5728\u8f93\u51fa\u7684\u8fc7\u7a0b\u4e2d":63,"\u5728\u8fd0\u884c":93,"\u5728\u8fd0\u884c\u795e\u7ecf\u7f51\u7edc\u8ba1\u7b97\u56fe\u65f6":69,"\u5728\u8fd9\u4e2a":42,"\u5728\u8fd9\u4e2a\u4efb\u52a1\u4e2d":94,"\u5728\u8fd9\u4e2a\u4f8b\u5b50\u4e2d":[52,93],"\u5728\u8fd9\u4e2a\u4f8b\u5b50\u91cc":[67,79],"\u5728\u8fd9\u4e2a\u51fd\u6570\u4e2d":61,"\u5728\u8fd9\u4e2a\u52a8\u6001\u5e93\u4e2d\u4e0d\u5d4c\u5165\u4efb\u4f55\u5176\u4ed6\u8bed\u8a00\u7684\u89e3\u91ca\u5668":33,"\u5728\u8fd9\u4e2a\u6559\u7a0b\u4e2d":72,"\u5728\u8fd9\u4e2a\u6a21\u578b\u4e2d":64,"\u5728\u8fd9\u4e2a\u793a\u4f8b\u4e2d":92,"\u5728\u8fd9\u4e2a\u9636\u6bb5\u7684\u4ee3\u7801\u6b63\u5728\u7ecf\u5386\u56de\u5f52\u6d4b\u8bd5":42,"\u5728\u8fd9\u4e9b\u5934\u6587\u4ef6\u4e2d":34,"\u5728\u8fd9\u4e9b\u6587\u4ef6\u4e2d":34,"\u5728\u8fd9\u4e9b\u7f51\u7edc\u4e2d":91,"\u5728\u8fd9\u4e9blayer\u4e2d":61,"\u5728\u8fd9\u65f6\u771f\u6b63\u7684\u5206\u914d\u5185\u5b58":69,"\u5728\u8fd9\u6b65\u4efb\u52a1\u4e2d":93,"\u5728\u8fd9\u6bb5\u4ee3\u7801\u4e2d":69,"\u5728\u8fd9\u79cd\u60c5\u51b5\u4e0b":[64,67],"\u5728\u8fd9\u79cd\u7ed3\u6784\u4e2d":63,"\u5728\u8fd9\u7bc7\u6587\u6863\u91cc":79,"\u5728\u8fd9\u7bc7\u6587\u7ae0\u91cc":80,"\u5728\u8fd9\u91cc":63,"\u5728\u8fd9\u91cc\u6211\u4eec\u4f7f\u7528\u5168\u8fde\u63a5\u5c42\u4f5c\u4e3a\u4f8b\u5b50\u6765\u5c55\u793a\u5b9e\u73b0\u65b0\u7f51\u7edc\u5c42\u6240\u9700\u8981\u7684\u56db\u4e2a\u6b65\u9aa4":67,"\u5728\u8fd9\u91cc\u7528eigenvector\u6765\u8868\u793a":69,"\u5728\u8fd9\u91cc\u9700\u8981\u6ce8\u610f\u7684\u662f":69,"\u5728\u8fdb\u884c\u7f51\u7edc\u914d\u7f6e\u4e4b\u524d":56,"\u5728\u914d\u7f6e\u4e2d\u9700\u8981\u8bfb\u53d6\u5916\u90e8\u5b57\u5178":2,"\u5728\u914d\u7f6e\u6587\u4ef6\u4e2d\u7684":87,"\u5728\u91c7\u7528sgd":51,"\u5728\u96c6\u7fa4\u4e0a\u8bad\u7ec3\u4e00\u4e2a\u7a00\u758f\u6a21\u578b\u9700\u8981\u52a0\u4e0a\u4e0b\u9762\u7684\u53c2\u6570":77,"\u5728\u975e\u5e8f\u5217\u8f93\u5165\u65f6":49,"\u5728\u9884\u5904\u7406\u542b\u6709\u591a\u884c\u6570\u6910\u7684\u6587\u672c\u6587\u4ef6\u65f6\u53c2\u6570\u8bbe\u7f6e\u7a0d\u6709\u4e0d\u540c":93,"\u5728\u9884\u6d4b\u5e8f\u5217\u6216\u6bb5\u843d\u7684\u60c5\u611f\u4e2d\u8d77\u4e3b\u8981\u4f5c\u7528":93,"\u5728android\u5e73\u53f0\u4e0a\u4e0d\u652f\u6301\u901a\u8fc7swig\u8c03\u7528\u6765\u8bad\u7ec3\u6216\u8005\u9884\u6d4b":58,"\u5728android\u5e73\u53f0\u4e0a\u53ea\u652f\u6301\u4f7f\u7528c":58,"\u5728aws\u4e0a\u5feb\u901f\u90e8\u7f72\u96c6\u7fa4":78,"\u5728build\u76ee\u5f55\u4e0b\u6267\u884c":46,"\u5728c":33,"\u5728c\u7684\u5934\u6587\u4ef6":33,"\u5728cub":86,"\u5728eigen\u4e2d":69,"\u5728generator\u7684\u4e0a\u4e0b\u6587\u4e2d\u5c3d\u91cf\u7559\u4e0b\u975e\u5e38\u5c11\u7684\u53d8\u91cf\u5f15\u7528":2,"\u5728kubernetes\u4e2d\u521b\u5efa\u7684\u6240\u6709\u8d44\u6e90\u5bf9\u8c61":78,"\u5728linux\u4e0b":94,"\u5728meta\u6587\u4ef6\u4e2d\u6709\u4e24\u79cd\u7279\u5f81":91,"\u5728movielen":91,"\u5728paddl":80,"\u5728paddle\u4e2d":77,"\u5728paddle\u4e4b\u4e0a\u8fd0\u884c\u7684\u6df1\u5ea6\u5b66\u4e60\u8bad\u7ec3\u8f93\u51fa\u7684\u6a21\u578b\u4f1a\u63d0\u4f9b\u7ed9\u5728\u7ebf\u4eba\u8138\u8bc6\u522b\u7684\u5e94\u7528\u4f7f\u7528":21,"\u5728paddlepaddle\u4e2d":[56,63],"\u5728paddlepaddle\u4e2d\u4f7f\u7528dropout\u6709\u4e24\u79cd\u65b9\u5f0f":50,"\u5728paddlepaddle\u4e2d\u5305\u542b\u4ee5\u4e0b":50,"\u5728paddlepaddle\u7684\u6587\u6863\u4e2d":61,"\u5728paddlepaddle\u91cc":52,"\u5728paramet":20,"\u5728python\u811a\u672c\u4e2d\u5b9e\u73b0\u4e0e\u524d\u5411operator\u76f8\u540c\u7684\u8ba1\u7b97\u903b\u8f91":68,"\u5728step\u51fd\u6570\u4e2d\u5b9a\u4e49":63,"\u5728step\u51fd\u6570\u4e2d\u5b9a\u4e49memori":63,"\u5728trainer":77,"\u5730\u5740\u4e5f\u53ef\u4ee5\u4e3ahdfs\u6587\u4ef6\u8def\u5f84":1,"\u5730\u6bb5":61,"\u5730\u7406\u4f4d\u7f6e":61,"\u5730\u94c1\u7ad9":61,"\u5747\u4f1a\u88ab\u5b89\u88c5\u5230includ":34,"\u5747\u503c\u56fe\u50cf\u6587\u4ef6":87,"\u5747\u5300\u5206\u5e03":51,"\u5747\u5300\u5206\u5e03\u7684\u8303\u56f4\u662f":75,"\u5747\u662f\u5728":34,"\u5747\u6709\u4e09\u4e2a\u5b50\u5e8f\u5217":61,"\u5747\u6709\u4e24\u7ec4\u7279\u5f81":61,"\u57fa\u4e8e\u53cc\u5c42\u5e8f\u5217\u8f93\u5165":63,"\u57fa\u4e8e\u5b57\u6bcd\u7684\u8bcd\u5d4c\u5165\u7279\u5f81":91,"\u57fa\u4e8e\u7c98\u6027\u4f1a\u8bdd\u7684\u8d1f\u8f7d\u5747\u8861\u529f\u80fd":28,"\u57fa\u4e8epython\u7684\u6a21\u578b\u9884\u6d4b":4,"\u57fa\u4e8epython\u7684\u9884\u6d4b":[3,89],"\u57fa\u672c\u4e0a\u548cmnist\u6837\u4f8b\u4e00\u81f4":2,"\u57fa\u672c\u4f7f\u7528\u6982\u5ff5":57,"\u57fa\u672c\u76f8\u540c":85,"\u57fa\u7c7b":68,"\u586b\u5199":66,"\u589e\u52a0":68,"\u589e\u52a0\u4e86\u4e00\u6761cd\u547d\u4ee4":79,"\u589e\u52a0\u4e86\u8bbe\u5907\u7c7b\u578b":68,"\u589e\u52a0\u5982\u4e0b\u53c2\u6570":77,"\u589e\u52a0\u68af\u5ea6\u68c0\u6d4b\u7684\u5355\u5143\u6d4b\u8bd5":67,"\u58f0\u660epython\u6570\u636e\u6e90":91,"\u5904\u7406\u5668\u6709\u4e24\u4e2a\u5173\u952e\u6027\u80fd\u9650\u5236":72,"\u5904\u7406\u6570\u636e\u7684python\u811a\u672c\u6587\u4ef6":89,"\u5904\u7406\u7684\u8f93\u5165\u5e8f\u5217\u4e3b\u8981\u5206\u4e3a\u4ee5\u4e0b\u4e09\u79cd\u7c7b\u578b":63,"\u5904\u7406\u76f8\u4f3c\u5ea6\u56de\u5f52":91,"\u5904\u7406\u8fc7\u7a0b\u4e2d\u6570\u636e\u5b58\u50a8\u683c\u5f0f":86,"\u5907\u6ce8":72,"\u590d\u6742\u5ea6\u6216\u65f6\u95f4\u590d\u6742\u5ea6":72,"\u5916\u5c42memory\u662f\u4e00\u4e2a\u5143\u7d20":61,"\u5916\u5c42outer_step\u4e2d":61,"\u591a\u4e2a\u503c":21,"\u591a\u4e2a\u5c42\u7684\u8f93\u51fa\u77e9\u9635\u7684\u9ad8\u5ea6\u4e0d\u4e00\u81f4\u5bfc\u81f4\u62fc\u63a5\u5931\u8d25":49,"\u591a\u4e2a\u8f93\u51fa\u5c42\u5904\u7406\u591a\u4e2a\u4e0d\u540c\u957f\u5ea6\u7684\u5e8f\u5217":49,"\u591a\u4e2ainput\u4ee5list\u65b9\u5f0f\u8f93\u5165":89,"\u591a\u4e2aparamet":20,"\u591a\u53e5\u8bdd\u8fdb\u4e00\u6b65\u6784\u6210\u4e86\u6bb5\u843d":63,"\u591a\u673a\u8bad\u7ec3":49,"\u591a\u7ebf\u7a0b\u7684\u6570\u636e\u8bfb\u53d6":2,"\u591a\u8f6e\u5bf9\u8bdd\u7b49\u66f4\u4e3a\u590d\u6742\u7684\u8bed\u8a00\u6570\u636e":63,"\u5927\u578b\u7535\u5f71\u8bc4\u8bba\u6570\u636e\u96c6":93,"\u5927\u591a\u6570\u5c42\u4e0d\u9700\u8981\u8fdc\u7a0b\u7a00\u758f\u8bad\u7ec3\u51fd\u6570":67,"\u5927\u591a\u6570\u5c42\u9700\u8981\u8bbe\u7f6e\u4e3a":67,"\u5927\u591a\u6570\u6210\u529f\u7684srl\u7cfb\u7edf\u662f\u5efa\u7acb\u5728\u67d0\u79cd\u5f62\u5f0f\u7684\u53e5\u6cd5\u5206\u6790\u7ed3\u679c\u4e4b\u4e0a\u7684":92,"\u5927\u591a\u6570\u7f51\u7edc\u5c42\u4e0d\u9700\u8981\u652f\u6301\u8fdc\u7a0b\u7a00\u758f\u66f4\u65b0":67,"\u5927\u591a\u6570\u8bed\u8a00\u90fd\u652f\u6301\u4f7f\u7528c\u8bed\u8a00api":33,"\u5927\u5b66\u751f":90,"\u5927\u5bb6\u53ef\u4ee5\u7528\u628a\u5f00\u53d1\u5de5\u5177\u5b89\u88c5\u8fdb\u5165":65,"\u5927\u5bb6\u53ef\u4ee5\u901a\u8fc7\u5b83\u5236\u4f5c\u548c\u5206\u4eab\u5e26\u6709\u4ee3\u7801":54,"\u5927\u5c0f":73,"\u5927\u5c0f\u4e0d\u4e00\u6837\u65f6":49,"\u5927\u6982\u82b1\u5341\u5206\u949f\u770b\u4e00\u4e0b":65,"\u5929":61,"\u5929\u4e00\u5e7f\u573a":61,"\u5929\u4e00\u9601":61,"\u5929\u732b":93,"\u5934\u4fe1\u606f\u4e2d":51,"\u5934\u6587\u4ef6\u4e2d\u628a\u53c2\u6570\u5b9a\u4e49\u4e3a\u7c7b\u7684\u6210\u5458\u53d8\u91cf":67,"\u5934\u6587\u4ef6\u5982\u4e0b":67,"\u5947\u5e7b\u7247":90,"\u597d":61,"\u597d\u5403":61,"\u597d\u8bc4":89,"\u5982":[2,64,66,68,73,77],"\u59822":73,"\u5982\u4e0a\u4e00\u5c0f\u8282\u6240\u793a":69,"\u5982\u4e0b":[2,91,93],"\u5982\u4e0b\u56fe\u6240\u793a":[61,72,86],"\u5982\u4e0b\u6240\u793a":[77,87,91],"\u5982\u4e0b\u662f\u4e00\u6bb5\u4f7f\u7528mnist":4,"\u5982\u4e0b\u8868\u683c":89,"\u5982\u4f55":91,"\u5982\u4f55\u5b58\u50a8\u7b49\u7b49":2,"\u5982\u4f55\u89e3\u6790\u8be5\u5730\u5740\u4e5f\u662f\u7528\u6237\u81ea\u5b9a\u4e49dataprovider\u65f6\u9700\u8981\u8003\u8651\u7684\u5730\u65b9":1,"\u5982\u4f55\u8d21\u732e":71,"\u5982\u4f55\u8d21\u732e\u4ee3\u7801":71,"\u5982\u4f55\u8fdb\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3":89,"\u5982\u4fe1\u606f\u63d0\u53d6":92,"\u5982\u56fe2\u6240\u793a":93,"\u5982\u56fe\u4e2dtrainer":20,"\u5982\u5f62\u5bb9\u8bcd\u548c\u526f\u8bcd":93,"\u5982\u60f3\u4e86\u89e3\u66f4\u591a\u8be6\u7ec6\u7684\u89e3\u91ca":94,"\u5982\u6709":68,"\u5982\u672c\u4f8b\u4e2d":2,"\u5982\u672c\u4f8b\u7684":2,"\u5982\u679c\u4e00\u4e2a\u7f51\u7edc\u5c42\u9700\u8981\u914d\u7f6e\u7684\u8bdd":67,"\u5982\u679c\u4e0a\u9762\u4e24\u6b65\u51fa\u73b0\u9519\u8bef":20,"\u5982\u679c\u4e0b\u8f7d\u6210\u529f":87,"\u5982\u679c\u4e0d\u4e3a0":75,"\u5982\u679c\u4e0d\u4e86\u89e3":2,"\u5982\u679c\u4e0d\u5207\u8bcd":89,"\u5982\u679c\u4e0d\u6536\u655b":51,"\u5982\u679c\u4e0d\u662f\u5e8f\u5217":91,"\u5982\u679c\u4e3a":2,"\u5982\u679c\u4e3a0":75,"\u5982\u679c\u4e3afals":75,"\u5982\u679c\u4e3atrue":[2,75],"\u5982\u679c\u4e4b\u540e\u60f3\u8981\u91cd\u65b0\u8bbe\u7f6e":53,"\u5982\u679c\u4ec5\u4ec5\u4fee\u6539\u4e00\u4e2a\u6587\u4ef6\u4f46\u63d0\u4ea4\u4e86\u5341\u51e0\u4e2acommit":66,"\u5982\u679c\u4ed4\u7ec6\u8bbe\u7f6e\u7684\u8bdd":75,"\u5982\u679c\u4f60\u4e0d\u9700\u8981\u8fd9\u4e2a\u64cd\u4f5c":93,"\u5982\u679c\u4f60\u53ea\u9700\u8981\u4f7f\u7528\u7b80\u5355\u7684rnn":64,"\u5982\u679c\u4f60\u5b89\u88c5gpu\u7248\u672c\u7684paddlepaddl":93,"\u5982\u679c\u4f60\u60f3\u4f7f\u7528\u8fd9\u4e9b\u7279\u6027":77,"\u5982\u679c\u4f60\u60f3\u8981\u4fdd\u5b58\u67d0\u4e9b\u5c42\u7684\u7279\u5f81\u56fe":75,"\u5982\u679c\u4f60\u60f3\u8fdb\u884c\u8bf8\u5982\u8bed\u4e49\u8f6c\u8ff0":94,"\u5982\u679c\u4f60\u6267\u884c\u5176\u5b83\u7684\u7528\u60c5\u611f\u5206\u6790\u6765\u5206\u7c7b\u6587\u672c\u7684\u4efb\u52a1":93,"\u5982\u679c\u4f60\u66fe\u7ecf\u5728\u6e90\u7801\u76ee\u5f55\u4e0b\u7f16\u8bd1\u8fc7\u5176\u4ed6\u5e73\u53f0\u7684paddlepaddle\u5e93":[58,59],"\u5982\u679c\u4f60\u6b63\u5728\u5904\u7406\u5e8f\u5217\u6807\u8bb0\u4efb\u52a1":64,"\u5982\u679c\u4f60\u6ca1\u6709gpu\u73af\u5883":86,"\u5982\u679c\u4f60\u8981\u4e3a\u4e86\u6d4b\u8bd5\u800c\u589e\u52a0\u65b0\u7684\u6587\u4ef6":67,"\u5982\u679c\u4f7f\u7528":[73,85],"\u5982\u679c\u4f7f\u7528nvidia":54,"\u5982\u679c\u4f7f\u7528ssl\u8ba4\u8bc1":78,"\u5982\u679c\u4f7f\u7528swig\u6211\u4eec\u9700\u8981\u5c06\u5728interface\u6587\u4ef6\u91cc":33,"\u5982\u679c\u51fa\u73b0\u4ee5\u4e0bpython\u76f8\u5173\u7684\u5355\u5143\u6d4b\u8bd5\u90fd\u8fc7\u4e0d\u4e86\u7684\u60c5\u51b5":46,"\u5982\u679c\u53c2\u6570\u4fdd\u5b58\u4e0b\u6765\u7684\u6a21\u578b\u76ee\u5f55":49,"\u5982\u679c\u53c2\u6570\u6a21\u578b\u6587\u4ef6\u7f3a\u5931":85,"\u5982\u679c\u53d1\u73b0\u6700\u65e9\u7684\u62a5\u9519\u5c31\u662f\u7f51\u7edc\u901a\u4fe1\u7684\u95ee\u9898":47,"\u5982\u679c\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u672a\u8bbe\u7f6easync":75,"\u5982\u679c\u5728\u8bad\u7ec3\u671f\u95f4\u540c\u65f6\u53d1\u8d77\u53e6\u5916\u4e00\u4e2a\u8fdb\u7a0b\u8fdb\u884c\u6d4b\u8bd5":75,"\u5982\u679c\u5728\u8bad\u7ec3\u914d\u7f6e\u4e2d\u8bbe\u7f6ebatch":75,"\u5982\u679c\u5728\u8bad\u7ec3nlp\u76f8\u5173\u6a21\u578b\u65f6":51,"\u5982\u679c\u591a\u4e2aop\u4f9d\u8d56\u4e00\u4e9b\u5171\u7528\u7684\u51fd\u6570":68,"\u5982\u679c\u5931\u8d25":42,"\u5982\u679c\u5b58\u5728\u67d0\u4e9btrainer\u6267\u884c\u901f\u5ea6\u8fc7\u6162\u4f1a\u5f71\u54cd\u6574\u4f53\u96c6\u7fa4\u7684\u901f\u5ea6":20,"\u5982\u679c\u5b83\u4f4d\u4e8e\u8c13\u8bcd\u4e0a\u4e0b\u6587\u533a\u57df\u4e2d":92,"\u5982\u679c\u5c06\u8fd9\u4e2a\u5185\u5b58\u6c60\u51cf\u5c0f":49,"\u5982\u679c\u5df2\u5b89\u88c5":92,"\u5982\u679c\u5df2\u7ecf\u6709pod\u8fd0\u884c":80,"\u5982\u679c\u5df2\u7ecf\u6b63\u5728\u6267\u884c\u4fdd\u5b58\u68c0\u67e5\u70b9\u7684\u7ebf\u7a0b":20,"\u5982\u679c\u5f00\u542f\u4f1a\u5bfc\u81f4\u8fd0\u884c\u7565\u6162":53,"\u5982\u679c\u5f53\u524dmpi\u96c6\u7fa4\u5e76\u4e0d\u652f\u6301\u4efb\u52a1\u72ec\u5360\u6a21\u5f0f":47,"\u5982\u679c\u60a8\u5728\u56fd\u5185":54,"\u5982\u679c\u60a8\u60f3\u8981\u66f4\u6df1\u5165\u4e86\u89e3deep":54,"\u5982\u679c\u60a8\u6709\u597d\u7684\u5efa\u8bae\u6765":91,"\u5982\u679c\u60a8\u6ca1\u6709\u542c\u8bf4":65,"\u5982\u679c\u60a8\u7684gpu\u7406\u8bba\u53ef\u4ee5\u8fbe\u52306":72,"\u5982\u679c\u60f3\u4e3a\u4e00\u4e2a\u6570\u636e\u6587\u4ef6\u8fd4\u56de\u591a\u6761\u6837\u672c":2,"\u5982\u679c\u60f3\u4f7f\u7528\u53ef\u89c6\u5316\u7684\u5206\u6790\u5668":72,"\u5982\u679c\u60f3\u5f88\u597d\u7684\u7406\u89e3\u7a0b\u5e8f\u7684\u884c\u4e3a":72,"\u5982\u679c\u60f3\u6539\u53d8\u539f\u6709tensor\u7684shape\u4fe1\u606f":69,"\u5982\u679c\u60f3\u8981\u4e86\u89e3\u53cc\u5c42rnn\u5728\u5177\u4f53\u95ee\u9898\u4e2d\u7684\u4f7f\u7528":61,"\u5982\u679c\u60f3\u8981\u542f\u7528paddlepaddle\u7684\u5185\u7f6e\u5b9a\u65f6\u5668":72,"\u5982\u679c\u6211\u4eec\u53ea\u9700\u8981\u7f16\u8bd1\u4e00\u4e2a\u53ea\u652f\u6301":65,"\u5982\u679c\u6211\u77e5\u9053\u5185\u6838\u82b1\u4e8610ms\u6765\u79fb\u52a81gb\u6570\u636e":72,"\u5982\u679c\u6267\u884c\u5931\u8d25":78,"\u5982\u679c\u6267\u884c\u6210\u529f":87,"\u5982\u679c\u6307\u5b9a\u4e862\u4e2alayer\u4f5c\u4e3a\u8f93\u51fa\u5c42":49,"\u5982\u679c\u63d0\u793a\u6b63\u786e":70,"\u5982\u679c\u652f\u6301\u589e\u52a0\u6b64\u53c2\u6570\u63d0\u4ea4":47,"\u5982\u679c\u6570\u636e\u6587\u4ef6\u5b58\u4e8e\u672c\u5730\u78c1\u76d8":1,"\u5982\u679c\u6570\u6910\u83b7\u53d6\u6210\u529f":93,"\u5982\u679c\u662f\u4f7f\u7528\u975essl\u65b9\u5f0f\u8bbf\u95ee":78,"\u5982\u679c\u662f\u5176\u5b83\u7c7b\u578b":21,"\u5982\u679c\u662f\u5e8f\u5217":91,"\u5982\u679c\u6709\u591a\u4e2a\u8f93\u5165":63,"\u5982\u679c\u6709\u591a\u4e2a\u8f93\u5165\u5e8f\u5217":63,"\u5982\u679c\u6709\u66f4\u590d\u6742\u7684\u4f7f\u7528":1,"\u5982\u679c\u6709\u9700\u8981\u4fee\u6539\u7684\u5730\u65b9":66,"\u5982\u679c\u6709bugfix\u7684\u884c\u4e3a":42,"\u5982\u679c\u672a\u8bbe\u7f6e":75,"\u5982\u679c\u672a\u8bbe\u7f6egpu":77,"\u5982\u679c\u67d0\u4e00\u4e2a\u7c7b\u578b\u9700\u8981\u5f15\u7528\u53e6\u4e00\u4e2a\u7c7b\u578b":34,"\u5982\u679c\u67d0\u4e00\u4e2apaddl":34,"\u5982\u679c\u67d0\u4e00\u4e2apaddle\u6982\u5ff5\u5fc5\u987b\u8981\u66b4\u9732":34,"\u5982\u679c\u67d0\u4e00\u5757\u6839\u672c\u5c31\u4e0d\u600e\u4e48\u8017\u65f6":72,"\u5982\u679c\u68c0\u67e5\u5230\u5206\u914d\u5728\u4e0d\u540c\u53c2\u6570\u670d\u52a1\u5668\u4e0a\u7684\u53c2\u6570\u7684\u5206\u5e03\u4e0d\u5747\u5300\u6b21\u6570\u5927\u4e8echeck":75,"\u5982\u679c\u6ca1\u6709\u5b9a\u4e49memori":63,"\u5982\u679c\u6ca1\u6709\u8bbe\u7f6e":94,"\u5982\u679c\u6ca1\u6709\u8bbe\u7f6etest":1,"\u5982\u679c\u6ca1\u8fc7":66,"\u5982\u679c\u6d88\u606f\u6570\u636e\u592a\u5c0f":75,"\u5982\u679c\u6ee1\u8db3\u6761\u4ef6":20,"\u5982\u679c\u7528\u516c\u7528\u7684\u7535\u8111\u5f00\u53d1":65,"\u5982\u679c\u7528\u6237\u4e0d\u663e\u793a\u6307\u5b9a\u8fd4\u56de\u6570\u636e\u7684\u5bf9\u5e94\u5173\u7cfb":2,"\u5982\u679c\u7528\u6237\u4e0d\u9700\u8981\u8bbf\u95eelstm\u7684\u4e2d\u95f4\u53d8\u91cf":50,"\u5982\u679c\u7528\u6237\u60f3\u8981\u4e86\u89e3\u8be6\u7ec6\u7684\u6570\u636e\u96c6\u7684\u683c\u5f0f":85,"\u5982\u679c\u7528\u6237\u60f3\u8981\u81ea\u5b9a\u4e49\u521d\u59cb\u5316\u65b9\u5f0f":51,"\u5982\u679c\u7528\u6237\u8981\u628apaddle\u7684\u9759\u6001\u5e93":33,"\u5982\u679c\u7528\u81ea\u5df1\u7684\u7535\u8111\u5f00\u53d1":65,"\u5982\u679c\u771f\u60f3\u6316\u6398\u5185\u6838\u6df1\u5904\u7684\u67d0\u4e2a\u79d8\u5bc6":72,"\u5982\u679c\u7a0b\u5e8f\u5d29\u6e83\u4f60\u4e5f\u53ef\u4ee5\u624b\u52a8\u7ec8\u6b62":73,"\u5982\u679c\u7cfb\u7edf\u5b89\u88c5\u4e86\u591a\u4e2apython\u7248\u672c":46,"\u5982\u679c\u7cfb\u7edf\u652f\u6301":46,"\u5982\u679c\u7cfb\u7edf\u652f\u6301\u7684\u662f":46,"\u5982\u679c\u7f16\u8bd1\u7684\u65f6\u5019\u6211\u4eec\u7528\u4e86":65,"\u5982\u679c\u7f51\u7edc\u5c42\u4e0d\u9700\u8981\u8fdc\u7a0b\u7a00\u758f\u66f4\u65b0":67,"\u5982\u679c\u7f51\u7edc\u67b6\u6784\u7b80\u5355":64,"\u5982\u679c\u8981\u4e0a\u4f20gpu\u7248\u672c\u7684\u5305":42,"\u5982\u679c\u8981\u4f7f\u7528\u53cc\u5411lstm":93,"\u5982\u679c\u8981\u4f7f\u7528gpu":54,"\u5982\u679c\u8981\u542f\u7528gpu":73,"\u5982\u679c\u8981\u8fd0\u884c\u6240\u6709\u7684\u5355\u5143\u6d4b\u8bd5":66,"\u5982\u679c\u89e3\u51b3\u4e86\u67d0\u4e2aissue\u7684\u95ee\u9898":66,"\u5982\u679c\u8bad\u7ec3\u4e00\u4e2apass":51,"\u5982\u679c\u8bad\u7ec3\u8fc7\u7a0b\u542f\u52a8\u6210\u529f\u7684\u8bdd":91,"\u5982\u679c\u8bad\u7ec3\u8fc7\u7a0b\u7684\u7684cost\u660e\u663e\u9ad8\u4e8e\u8fd9\u4e2a\u5e38\u6570\u8f93\u51fa\u7684cost":51,"\u5982\u679c\u8bbe\u7f6e":2,"\u5982\u679c\u8bbe\u7f6e\u8be5\u53c2\u6570":75,"\u5982\u679c\u8bc4\u5ba1\u610f\u89c1\u6bd4\u8f83\u591a":66,"\u5982\u679c\u8bf4\u4e00\u4e2adocker\u955c\u50cf\u5c31\u662f\u4e00\u4e2a\u7a0b\u5e8f":54,"\u5982\u679c\u8c03\u7528\u9759\u6001\u5e93\u53ea\u80fd\u5c06\u9759\u6001\u5e93\u4e0e\u89e3\u91ca\u5668\u94fe\u63a5":33,"\u5982\u679c\u8f93\u51fa\u662fno":54,"\u5982\u679c\u8fd0\u884c":46,"\u5982\u679c\u8fd0\u884c\u6210\u529f":[87,93],"\u5982\u679c\u8fd8\u4e0d\u884c":46,"\u5982\u679c\u96c6\u7fa4\u8282\u70b9\u6570\u91cf\u5c11":73,"\u5982\u679c\u9700\u8981\u5305\u542b\u66f4\u591a\u7684\u4f9d\u8d56":54,"\u5982\u679c\u9700\u8981\u6269\u5927\u77e9\u9635":67,"\u5982\u679c\u9700\u8981\u7f29\u51cf\u77e9\u9635":67,"\u5982\u679cgpu":68,"\u5982\u679clearning_rate\u592a\u5927":51,"\u5982\u679clearning_rate\u592a\u5c0f":51,"\u5982\u679cop\u6ca1\u6709\u5b9e\u73b0gpu":68,"\u5982\u679cop\u7684\u67d0\u4e2a\u8f93\u5165\u4e0d\u53c2\u4e0e\u53cd\u5411\u68af\u5ea6\u7684\u8ba1\u7b97":68,"\u5982\u679cpaddlepaddle\u5305\u5df2\u7ecf\u5728python\u7684sit":46,"\u5982\u679cparamet":20,"\u5982\u6bcf\u4e2a\u6587\u4ef6\u53ea\u6709\u4e00\u4e2a":66,"\u5982\u795e\u7ecf\u5143\u6fc0\u6d3b\u503c\u7b49":49,"\u5982\u8981build\u8fd9\u4e2a\u5f00\u53d1\u955c\u50cf":66,"\u5982\u9ad8\u4eae\u90e8\u5206":72,"\u5b50":61,"\u5b50\u53e5":63,"\u5b50\u53e5\u7684\u5355\u8bcd\u6570\u548c\u6307\u5b9a\u7684\u4e00\u4e2a\u8f93\u5165\u5e8f\u5217\u4e00\u81f4":63,"\u5b50\u76ee\u5f55":65,"\u5b57\u5178":94,"\u5b57\u5178\u4f1a\u5305\u542b\u8f93\u5165\u6570\u636e\u96c6\u4e2d\u7684\u6240\u6709\u5355\u8bcd":94,"\u5b57\u5178\u5171\u5305\u542b":85,"\u5b57\u5178\u6587\u4ef6":[92,93],"\u5b57\u5178\u91c7\u7528utf8\u7f16\u7801":85,"\u5b57\u5178imdb":93,"\u5b57\u6bb5\u4e2d":80,"\u5b57\u6bb5\u4e3a\u4f8b":49,"\u5b57\u6bb5\u8868\u793a\u5bb9\u5668\u7684\u73af\u5883\u53d8\u91cf":80,"\u5b57\u6bb5\u8868\u793a\u8fd9\u4e2ajob\u4f1a\u540c\u65f6\u5f00\u542f3\u4e2apaddlepaddle\u8282\u70b9":80,"\u5b57\u7b26\u4e32":21,"\u5b58\u50a8":21,"\u5b58\u50a8\u5377":78,"\u5b58\u50a8\u5728\u8bb0\u5fc6\u5355\u5143\u533a\u5757\u7684\u5386\u53f2\u4fe1\u606f\u88ab\u66f4\u65b0\u7528\u6765\u8fed\u4ee3\u7684\u5b66\u4e60\u5355\u8bcd\u4ee5\u5408\u7406\u7684\u5e8f\u5217\u7a0b\u73b0":93,"\u5b58\u50a8\u6a21\u578b\u7684\u8def\u5f84":94,"\u5b58\u50a8\u7740\u7535\u5f71\u6216\u7528\u6237\u4fe1\u606f":91,"\u5b58\u5165settings\u5bf9\u8c61":2,"\u5b58\u5728\u6216\u66f4\u6539\u4e3a\u5176\u5b83\u6a21\u578b\u8def\u5f84":93,"\u5b66\u4e60":65,"\u5b66\u4e60\u6210\u672c\u9ad8":33,"\u5b66\u4e60\u7387\u4e3a":51,"\u5b66\u4e60\u7b97\u6cd5":52,"\u5b66\u672f":90,"\u5b81\u6ce2":61,"\u5b83\u4e0d\u4ec5\u80fd\u591f\u5904\u7406imdb\u6570\u636e":93,"\u5b83\u4eec\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4f5c\u4e3a\u7f51\u7edc\u7684\u51fa\u53e3":52,"\u5b83\u4eec\u7684\u6587\u4ef6\u540d\u662f":21,"\u5b83\u4eec\u8fd8\u53ef\u4ee5\u4f9b\u90a3\u4e9b\u8fd0\u884c\u66f4\u590d\u6742\u7684\u96c6\u7fa4\u7ba1\u7406\u7cfb\u7edf":73,"\u5b83\u4f1a\u5728dataprovider\u521b\u5efa\u7684\u65f6\u5019\u6267\u884c":2,"\u5b83\u4f7f\u752850\u5c42\u7684resnet\u6a21\u578b\u6765\u5bf9":87,"\u5b83\u5305\u542b\u4e86\u8fd9\u4e2a\u8f6f\u4ef6\u672c\u8eab\u548c\u5b83\u6240\u4f9d\u8d56\u7684\u8fd0\u884c\u73af\u5883":54,"\u5b83\u5305\u542b\u4ee5\u4e0b\u51e0\u6b65":67,"\u5b83\u5305\u542b\u4ee5\u4e0b\u53c2\u6570":67,"\u5b83\u5305\u542b\u7684\u5c5e\u6027\u53c2\u6570\u5982\u4e0b":2,"\u5b83\u5305\u62ec\u4e86\u4e00\u4e2a\u53cc\u5411\u7684gru\u4f5c\u4e3a\u7f16\u7801\u5668\u548c\u89e3\u7801\u5668":94,"\u5b83\u53eb\u505a":64,"\u5b83\u53ef\u4ee5\u5728\u53e5\u5b50\u7ea7\u522b\u5229\u7528\u53ef\u6269\u5c55\u7684\u4e0a\u4e0b\u6587":93,"\u5b83\u53ef\u4ee5\u5e2e\u52a9\u51cf\u5c11\u5206\u53d1\u5ef6\u8fdf":73,"\u5b83\u53ef\u4ee5\u5e2e\u52a9\u6211\u4eec\u683c\u5f0f\u5316\u6e90\u4ee3\u7801":66,"\u5b83\u53ef\u4ee5\u6307\u6d4b\u91cf\u4e00\u4e2a\u7a0b\u5e8f\u7684\u7a7a\u95f4":72,"\u5b83\u53ef\u4ee5\u88ab\u5e94\u7528\u4e8e\u8fdb\u884c\u673a\u5668\u7ffb\u8bd1":94,"\u5b83\u53ef\u80fd\u6709\u4e0d\u6b62\u4e00\u4e2a\u6743\u91cd":67,"\u5b83\u540c\u65f6\u5b66\u4e60\u6392\u5217":94,"\u5b83\u548c\u6570\u636e\u4f20\u5165\u51fd\u6570\u7684\u7b2c\u4e00\u4e2a\u53c2\u6570":2,"\u5b83\u5b58\u50a8\u5f53\u524d\u8282\u70b9\u6240\u6709\u8bad\u7ec3":73,"\u5b83\u5b9a\u4e49\u4e86":64,"\u5b83\u5b9a\u4e49\u4e86\u6a21\u578b\u53c2\u6570\u6539\u53d8\u7684\u89c4\u5219":52,"\u5b83\u5b9a\u4e49\u89e3\u7801\u7f51\u7edc\u7684":64,"\u5b83\u5c06\u88ab\u5206\u53d1\u5230":73,"\u5b83\u5c06\u8f93\u5165\u8bed\u53e5\u7f16\u7801\u4e3a\u5411\u91cf\u7684\u5e8f\u5217":94,"\u5b83\u5c06\u8fd4\u56de\u5982\u4e0b\u7684\u5b57\u5178":87,"\u5b83\u5c31\u4f1a\u5728\u6e90\u8bed\u53e5\u4e2d\u641c\u7d22\u51fa\u6700\u76f8\u5173\u4fe1\u606f\u7684\u4f4d\u7f6e\u7684\u96c6\u5408":94,"\u5b83\u5e76\u4e0d\u662f\u4e00\u4e2a\u5b8c\u6574\u7684recurr":50,"\u5b83\u652f\u6301\u591a\u7ebf\u7a0b\u66f4\u65b0":67,"\u5b83\u662finteger_value\u7c7b\u578b\u7684":61,"\u5b83\u662finteger_value_sequence\u7c7b\u578b\u7684":61,"\u5b83\u6709\u52a9\u4e8e\u5e2e\u52a9\u9891\u7e41\u4fee\u6539\u548c\u8bbf\u95ee\u5de5\u4f5c\u533a\u6587\u4ef6\u7684\u7528\u6237\u51cf\u5c11\u8d1f\u62c5":73,"\u5b83\u6a21\u62df\u4e86\u89e3\u7801\u7ffb\u8bd1\u8fc7\u7a0b\u4e2d\u5728\u6e90\u8bed\u53e5\u4e2d\u7684\u641c\u7d22":94,"\u5b83\u7684":64,"\u5b83\u7684\u529f\u80fd\u662f":68,"\u5b83\u7684\u6536\u655b\u901f\u5ea6\u6bd4":93,"\u5b83\u7684\u6bcf\u4e00\u4e2a\u5143\u7d20":60,"\u5b83\u7684\u76ee\u7684\u662f\u9884\u6d4b\u5728\u4e00\u4e2a\u5e8f\u5217\u4e2d\u8868\u8fbe\u7684\u60c5\u611f\u6001\u5ea6":93,"\u5b83\u7684\u8f93\u5165\u4e0e\u7ecf\u8fc7\u5b66\u4e60\u7684\u53c2\u6570\u505a\u5185\u79ef\u5e76\u52a0\u4e0a\u504f\u7f6e":67,"\u5b83\u76f4\u63a5\u5b66\u4e60\u6bb5\u843d\u8868\u793a":93,"\u5b83\u80fd\u591f\u4ece\u8bcd\u7ea7\u5230\u5177\u6709\u53ef\u53d8\u4e0a\u4e0b\u6587\u957f\u5ea6\u7684\u4e0a\u4e0b\u6587\u7ea7\u522b\u6765\u603b\u7ed3\u8868\u793a":93,"\u5b83\u8868\u793a":65,"\u5b83\u8bfb\u5165\u6570\u636e\u5e76\u5c06\u5b83\u4eec\u4f20\u8f93\u5230\u63a5\u4e0b\u6765\u7684\u7f51\u7edc\u5c42":52,"\u5b83\u8fd4\u56degen":94,"\u5b83\u8fd4\u56detrain":94,"\u5b83\u9700\u8981\u5728\u8fd9\u91cc\u6307\u5b9a":93,"\u5b83\u9996\u5148\u8c03\u7528\u57fa\u6784\u9020\u51fd\u6570":67,"\u5b89\u6392":61,"\u5b89\u88c5\u4e0e\u7f16\u8bd1":57,"\u5b89\u88c5\u540e\u7684\u76ee\u5f55\u7ed3\u6784\u4e3a":34,"\u5b89\u88c5\u597ddocker\u4e4b\u540e\u53ef\u4ee5\u4f7f\u7528\u6e90\u7801\u76ee\u5f55\u4e0b\u7684\u811a\u672c\u6784\u5efa\u6587\u6863":70,"\u5b89\u88c5\u5f00\u53d1\u5de5\u5177\u5230":65,"\u5b89\u88c5\u6587\u6863":56,"\u5b89\u88c5\u6d41\u7a0b":89,"\u5b89\u88c5\u8be5\u8f6f\u4ef6\u5305\u5c31\u53ef\u4ee5\u5728python\u73af\u5883\u4e0b\u5b9e\u73b0\u6a21\u578b\u9884\u6d4b":4,"\u5b89\u88c5paddlepaddl":89,"\u5b89\u88c5pillow":86,"\u5b89\u9759":61,"\u5b8c\u6210":66,"\u5b8c\u6210\u4e00\u4e2a\u4f20\u8f93\u52a8\u4f5c\u5b8c\u6210\u7684\u65f6\u95f4\u4e5f\u6bd4\u8f83\u77ed":28,"\u5b8c\u6210\u4efb\u610f\u7684\u8fd0\u7b97\u903b\u8f91":63,"\u5b8c\u6210\u540evolume\u4e2d\u7684\u6587\u4ef6\u5185\u5bb9\u5927\u81f4\u5982\u4e0b":80,"\u5b8c\u6210\u5f00\u53d1":54,"\u5b8c\u6210\u6570\u636e\u7684\u9884\u5904\u7406":21,"\u5b8c\u6210\u76f8\u5e94\u7684\u8ba1\u7b97":60,"\u5b8c\u6210paddlepaddle\u7684\u5b89\u88c5":56,"\u5b8c\u6574\u4ee3\u7801\u53ef\u4ee5\u53c2\u8003\u793a\u4f8b":49,"\u5b8c\u6574\u6e90\u7801\u53ef\u53c2\u8003":51,"\u5b8c\u6574\u7684\u4ee3\u7801\u89c1":4,"\u5b8c\u6574\u7684\u53c2\u6570\u77e9\u9635\u88ab\u5206\u5e03\u5728\u4e0d\u540c\u7684\u53c2\u6570\u670d\u52a1\u5668\u4e0a":67,"\u5b8c\u6574\u7684\u914d\u7f6e\u6587\u4ef6\u5728":64,"\u5b98\u65b9\u6587\u6863":53,"\u5b9a\u4e49\u4e00\u4e2a\u65f6\u95f4\u6b65\u4e4b\u5185rnn\u5355\u5143\u5b8c\u6210\u7684\u8ba1\u7b97":63,"\u5b9a\u4e49\u4e00\u4e2apython\u7684":2,"\u5b9a\u4e49\u4e86\u4e00\u4e2a\u53ea\u8bfb\u7684memori":63,"\u5b9a\u4e49\u4e86\u7f51\u7edc\u7684\u6570\u636e\u69fd":92,"\u5b9a\u4e49\u4e86\u7f51\u7edc\u7ed3\u6784":86,"\u5b9a\u4e49\u4e86\u7f51\u7edc\u7ed3\u6784\u5e76\u4fdd\u5b58\u4e3a":52,"\u5b9a\u4e49\u4e86lstm\u5355\u5143\u5728\u4e00\u4e2a\u65f6\u95f4\u6b65\u5185\u7684\u8ba1\u7b97\u8fc7\u7a0b":50,"\u5b9a\u4e49\u4f4d\u7f6e":68,"\u5b9a\u4e49\u4fdd\u6301\u4e00\u81f4":68,"\u5b9a\u4e49\u5728\u5916\u5c42":63,"\u5b9a\u4e49\u5f02\u6b65\u8bad\u7ec3\u7684\u957f\u5ea6":75,"\u5b9a\u4e49\u6570\u636e\u6765\u6e90":52,"\u5b9a\u4e49\u6e90\u8bed\u53e5\u7684\u6570\u636e\u5c42":64,"\u5b9a\u4e49\u7c7b\u578b":68,"\u5b9a\u4e49\u89e3\u7801\u5668\u7684memori":64,"\u5b9a\u4e49\u8bad\u7ec3\u6570\u6910\u548c\u6d4b\u8bd5\u6570\u6910\u63d0\u4f9b\u8005":93,"\u5b9a\u4e49\u8f93\u5165":68,"\u5b9a\u4e49\u8f93\u5165\u6570\u636e\u7684\u7c7b\u578b":52,"\u5b9a\u4e49\u8f93\u51fa":68,"\u5b9a\u4e49\u8f93\u51fa\u51fd\u6570":64,"\u5b9a\u4e49\u95e8\u63a7\u5faa\u73af\u5355\u5143\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u5355\u6b65\u51fd\u6570":64,"\u5b9d\u5854\u7684\u5e95\u7aef\u9700\u8981\u575a\u5b9e\u7684\u57fa\u5ea7\u6765\u652f\u6491":56,"\u5b9e\u4f8b\u5982\u4e0b":92,"\u5b9e\u73b0\u4e24\u4e2a\u5b8c\u5168\u7b49\u4ef7\u7684\u5168\u8fde\u63a5rnn":61,"\u5b9e\u73b0\u524d\u5411\u4f20\u64ad\u7684\u90e8\u5206\u6709\u4e0b\u9762\u51e0\u4e2a\u6b65\u9aa4":67,"\u5b9e\u73b0\u5355\u6b65\u51fd\u6570":64,"\u5b9e\u73b0\u540e\u5411\u4f20\u64ad\u7684\u90e8\u5206\u6709\u4e0b\u9762\u51e0\u4e2a\u6b65\u9aa4":67,"\u5b9e\u73b0\u5728":68,"\u5b9e\u73b0\u5bf9":69,"\u5b9e\u73b0\u6570\u636e\u8f93\u5165\u51fd\u6570":2,"\u5b9e\u73b0\u65b0\u7684op\u90fd\u6dfb\u52a0\u81f3\u76ee\u5f55":68,"\u5b9e\u73b0\u6784\u9020\u51fd\u6570":67,"\u5b9e\u73b0\u7684":50,"\u5b9e\u73b0\u7b80\u5355":33,"\u5b9e\u73b0\u7ec6\u8282":67,"\u5b9e\u73b0\u7f51\u7edc\u5c42\u7684\u524d\u5411\u4f20\u64ad":67,"\u5b9e\u73b0\u7f51\u7edc\u5c42\u7684\u540e\u5411\u4f20\u64ad":67,"\u5b9e\u73b0\u8bcd\u8bed\u548c\u53e5\u5b50\u4e24\u4e2a\u7ea7\u522b\u7684\u53cc\u5c42rnn\u7ed3\u6784":63,"\u5b9e\u73b0\u8be5\u5c42\u7684c":67,"\u5b9e\u9645\u4e0a":54,"\u5b9e\u9645\u4e0a\u4f7f\u7528\u4e86":50,"\u5b9e\u9645\u4e0a\u53ea\u6709":87,"\u5b9e\u9645\u4e0a\u662fcsv\u6587\u4ef6":90,"\u5b9e\u9645\u4e0a\u9700\u8981\u7684\u8f93\u51fa\u7ed3\u679c\u662f\u4e24\u4e2a\u77e9\u9635":49,"\u5ba2\u6237":61,"\u5ba2\u6237\u670d\u52a1":90,"\u5bb6":61,"\u5bb9\u5668":[54,78],"\u5bb9\u5668\u4e0d\u4f1a\u4fdd\u7559\u5728\u8fd0\u884c\u65f6\u751f\u6210\u7684\u6570\u636e":78,"\u5bb9\u5668\u4f7f\u7528":54,"\u5bb9\u5668\u8fd0\u884c\u90fd\u8fd0\u884c":80,"\u5bbd\u5ea6\u7b49\u4e8e\u914d\u7f6e\u4e2dlayer\u7684s":49,"\u5bbf\u4e3b\u673a\u662flinux":59,"\u5bbf\u4e3b\u673a\u7684c":[58,59],"\u5bbf\u4e3b\u673a\u76ee\u5f55":78,"\u5bc4\u5b58\u5668\u4f7f\u7528\u60c5\u51b5\u548c\u5171\u4eab\u5185\u5b58\u4f7f\u7528\u60c5\u51b5\u80fd\u8ba9\u6211\u4eec\u5bf9gpu\u7684\u6574\u4f53\u4f7f\u7528\u6709\u66f4\u597d\u7684\u7406\u89e3":72,"\u5bf9":61,"\u5bf9\u4e00\u4e2a5\u7ef4\u975e\u5e8f\u5217\u7684\u7a00\u758f01\u5411\u91cf":[2,56],"\u5bf9\u4e00\u4e2a5\u7ef4\u975e\u5e8f\u5217\u7684\u7a00\u758f\u6d6e\u70b9\u5411\u91cf":[2,56],"\u5bf9\u4e8e":64,"\u5bf9\u4e8e\u4e0d\u540c\u8bed\u8a00":33,"\u5bf9\u4e8e\u4e24\u79cd\u4e0d\u540c\u7684\u8f93\u5165\u6570\u636e\u7c7b\u578b":61,"\u5bf9\u4e8e\u4e60\u60ef\u4f7f\u7528windows\u548cmacos\u7684\u5f00\u53d1\u8005\u6765\u8bf4":65,"\u5bf9\u4e8e\u5185\u5b58\u8f83\u5c0f\u7684\u673a\u5668":2,"\u5bf9\u4e8e\u5355\u5c42rnn":61,"\u5bf9\u4e8e\u5355\u5c42rnn\u7684\u6570\u636e\u4e00\u5171\u6709\u4e24\u4e2a\u6837\u672c":61,"\u5bf9\u4e8e\u53cc\u5c42rnn":61,"\u5bf9\u4e8e\u540c\u4e00\u6bb5c":33,"\u5bf9\u4e8e\u540c\u6837\u7684\u6570\u636e":61,"\u5bf9\u4e8e\u591a\u8bed\u8a00\u63a5\u53e3":33,"\u5bf9\u4e8e\u5927\u591a\u6570\u8bed\u8a00":33,"\u5bf9\u4e8e\u6211\u4eec\u652f\u6301\u7684\u5168\u90e8\u77e9\u9635\u64cd\u4f5c":67,"\u5bf9\u4e8e\u6811\u7684\u6bcf\u4e00\u5c42":94,"\u5bf9\u4e8e\u6bb5\u843d\u7684\u6587\u672c\u5206\u7c7b":61,"\u5bf9\u4e8e\u6bcf\u4e00\u4e2a\u5355\u5c42rnn\u7684\u6570\u636e":61,"\u5bf9\u4e8e\u6bcf\u4f4d\u7528\u6237":91,"\u5bf9\u4e8e\u6bcf\u79cd\u7c7b\u578b":34,"\u5bf9\u4e8e\u6bcf\u79cdc":34,"\u5bf9\u4e8e\u7ed9\u5b9a\u7684\u4e00\u6761\u6587\u672c":89,"\u5bf9\u4e8e\u914d\u5907\u6709\u6ce8\u610f\u529b\u673a\u5236\u7684\u89e3\u7801\u5668":64,"\u5bf9\u4e8eamazon":89,"\u5bf9\u4e8erasspberri":59,"\u5bf9\u4ee3\u7801\u8fdb\u884c\u6027\u80fd\u5206\u6790":72,"\u5bf9\u5168\u8fde\u63a5\u5c42\u6765\u8bf4":67,"\u5bf9\u52a0\u8f7d\u9884\u8bad\u7ec3\u53c2\u6570\u7684\u5c42":51,"\u5bf9\u56fe\u7247\u8fdb\u884c\u9884\u5904\u7406":86,"\u5bf9\u5df2\u7ecfpush\u5230\u8fdc\u7a0b\u4ed3\u5e93\u7684\u591a\u4e2acommit":66,"\u5bf9\u5e94\u4e00\u4e2a\u5b50\u53e5":63,"\u5bf9\u5e94\u4e00\u4e2a\u8bcd":63,"\u5bf9\u5e94\u4e8e\u5b57\u5178":85,"\u5bf9\u5e94\u7684":2,"\u5bf9\u5e94\u7684\u68af\u5ea6op\u8ba1\u7b97\u4e4b\u4e2d":68,"\u5bf9\u6570\u636e\u96c6\u8fdb\u884c\u9884\u5904\u7406\u7684\u57fa\u672c\u547d\u4ee4\u662f":94,"\u5bf9\u6574\u4e2a\u65b0\u5411\u91cf\u96c6\u5408\u7684\u6bcf\u4e00\u4e2a\u7ef4\u5ea6\u53d6\u6700\u5927\u503c\u6765\u8868\u793a\u6700\u540e\u7684\u53e5\u5b50":89,"\u5bf9\u6587\u6863\u5904\u7406\u540e\u5f62\u6210\u7684\u5355\u8bcd\u5411\u91cf":93,"\u5bf9\u673a\u5668\u7ffb\u8bd1\u7684\u4eba\u5de5\u8bc4\u4f30\u5de5\u4f5c\u5f88\u5e7f\u6cdb\u4f46\u4e5f\u5f88\u6602\u8d35":94,"\u5bf9\u6bcf\u4e2a\u8f93\u5165":67,"\u5bf9\u6bcf\u4e2a\u8f93\u5165\u4e58\u4e0a\u53d8\u6362\u77e9\u9635":67,"\u5bf9\u6bd4":33,"\u5bf9\u6bd4\u53cd\u5411op\u4e0d\u540c\u8bbe\u5907":68,"\u5bf9\u6fc0\u6d3b\u6c42\u5bfc":67,"\u5bf9\u7528\u6237\u6765\u8bf4":2,"\u5bf9\u8bad\u7ec3\u6570\u636e\u8fdb\u884cshuffl":2,"\u5bf9\u8bc4\u5ba1\u610f\u89c1\u4e0d\u540c\u610f\u7684":66,"\u5bf9\u8bc4\u5ba1\u610f\u89c1\u540c\u610f\u4e14\u6309\u5176\u4fee\u6539\u5b8c\u7684":66,"\u5bf9\u8be5\u5411\u91cf\u8fdb\u884c\u975e\u7ebf\u6027\u53d8\u6362":89,"\u5bf9\u8c61":51,"\u5bf9\u8c61\u5b58\u50a8\u4e3a\u6587\u4ef6":91,"\u5bf9\u8f93\u5165\u53c2\u6570\u7684\u5b89\u5168\u6027\u8fdb\u884c\u4e86\u5fc5\u8981\u7684\u5224\u65ad":34,"\u5bf9\u8f93\u51fa\u7684\u5408\u5e76":63,"\u5bf9\u8fd9\u4e2a\u7248\u672c\u7684\u63d0\u4ea4":42,"\u5bf9\u9762":61,"\u5bf9check":2,"\u5bf9sparse_binary_vector\u548csparse_float_vector":[2,56],"\u5bfc\u51fa\u8fd9\u4e9b\u63a5\u53e3":34,"\u5bfc\u81f4\u4e86\u6d6e\u70b9\u6570\u6ea2\u51fa":49,"\u5bfc\u81f4\u53c2\u6570\u6536\u655b\u5230\u4e86\u4e00\u4e9b\u5947\u5f02\u7684\u60c5\u51b5":49,"\u5bfc\u81f4\u53c2\u6570\u7d2f\u52a0":49,"\u5bfc\u81f4\u7f16\u8bd1paddlepaddle\u5931\u8d25":46,"\u5bfc\u81f4\u8bad\u7ec3\u65f6\u95f4\u8fc7\u957f":51,"\u5c01\u88c5\u4e86":72,"\u5c01\u88c5\u8be5\u5c42\u7684python\u63a5\u53e3":67,"\u5c06":[2,42,51,72,89],"\u5c06\u4e00\u4e2a\u795e\u7ecf\u7f51\u7edc\u53c2\u6570\u62c6\u5206\u6210\u591a\u4efd":20,"\u5c06\u4e0a\u4e00\u65f6\u95f4\u6b65\u6240\u751f\u6210\u7684\u8bcd\u7684\u5411\u91cf\u6765\u4f5c\u4e3a\u5f53\u524d\u65f6\u95f4\u6b65\u7684\u8f93\u5165":64,"\u5c06\u4e3b\u673a\u4e0a\u7684\u6587\u4ef6\u6216\u76ee\u5f55\u6302\u8f7d\u5230":54,"\u5c06\u4ed6\u4eec\u79fb\u52a8\u5230\u76ee\u5f55":91,"\u5c06\u4f1a\u5c06\u7528\u6237\u4f20\u8fdb\u6765\u7684\u914d\u7f6e\u53c2\u6570\u4f20\u9012cmake\u7cfb\u7edf":58,"\u5c06\u4f1a\u81ea\u52a8\u8ba1\u7b97\u51fa\u4e00\u4e2a\u5408\u9002\u7684\u503c":75,"\u5c06\u5176\u8bbe\u7f6e\u6210":49,"\u5c06\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217\u6570\u636e\u5148\u53d8\u6362\u6210\u5355\u5c42\u65f6\u95f4\u5e8f\u5217\u6570\u636e":61,"\u5c06\u542b\u6709\u5b50\u53e5":63,"\u5c06\u542b\u6709\u8bcd\u8bed\u7684\u53e5\u5b50\u5b9a\u4e49\u4e3a\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":63,"\u5c06\u56fe\u7247\u6309\u7167\u4e0a\u8ff0\u7ed3\u6784\u5b58\u50a8\u597d\u4e4b\u540e":86,"\u5c06\u5728":86,"\u5c06\u5728\u8fd0\u884c\u65f6\u62a5\u9519":73,"\u5c06\u5916\u90e8\u7684\u5b58\u50a8\u670d\u52a1\u5728kubernetes\u4e2d\u63cf\u8ff0\u6210\u4e3a\u7edf\u4e00\u7684\u8d44\u6e90\u5f62\u5f0f":78,"\u5c06\u591a\u53e5\u8bdd\u770b\u6210\u4e00\u4e2a\u6574\u4f53\u540c\u65f6\u4f7f\u7528encoder\u538b\u7f29":61,"\u5c06\u591a\u53f0\u673a\u5668\u7684\u6d4b\u8bd5\u7ed3\u679c\u5408\u5e76":75,"\u5c06\u5927\u91cf\u7684":33,"\u5c06\u5b57\u5178\u7684\u5730\u5740\u4f5c\u4e3aargs\u4f20\u7ed9dataprovid":51,"\u5c06\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u524d\u5411\u548c\u53cd\u5411\u90e8\u5206\u6df7\u5408\u5728\u4e00\u8d77":64,"\u5c06\u6570\u636e\u5904\u7406\u6210\u89c4\u8303\u683c\u5f0f":85,"\u5c06\u6570\u636e\u7ec4\u5408\u6210batch\u8fdb\u884c\u8bad\u7ec3":2,"\u5c06\u6570\u636e\u8f6c\u6362\u4e3apaddle\u7684\u683c\u5f0f":86,"\u5c06\u65b0\u5206\u652f\u7684\u7248\u672c\u6253\u4e0atag":42,"\u5c06\u65b0\u5efa\u7684\u6743\u91cd\u52a0\u5165\u6743\u91cd\u8868":67,"\u5c06\u65e5\u5fd7\u5199\u5165\u6587\u4ef6":91,"\u5c06\u660e\u6587\u53c2\u6570\u8f6c\u5316\u4e3apaddlepaddle\u53ef\u52a0\u8f7d\u7684\u6a21\u578b\u53c2\u6570\u65f6":51,"\u5c06\u672c\u5730\u7684\u4fee\u6539\u63a8\u9001\u5230":66,"\u5c06\u6837\u672c\u7684\u5730\u5740\u653e\u5165\u53e6\u4e00\u4e2a\u6587\u672c\u6587\u4ef6":2,"\u5c06\u6b64\u76ee\u5f55\u6302\u8f7d\u4e3a\u5bb9\u5668\u7684":80,"\u5c06\u6bcf\u4e2a\u6e90\u8bed\u8a00\u5230\u76ee\u6807\u8bed\u8a00\u7684\u5e73\u884c\u8bed\u6599\u5e93\u6587\u4ef6\u5408\u5e76\u4e3a\u4e00\u4e2a\u6587\u4ef6":94,"\u5c06\u73af\u5883\u53d8\u91cf\u8f6c\u6362\u6210paddle\u7684\u547d\u4ee4\u884c\u53c2\u6570":80,"\u5c06\u7528\u4e8epython":68,"\u5c06\u7ed3\u679c\u4fdd\u5b58\u5230\u6b64\u76ee\u5f55\u91cc":80,"\u5c06\u884c\u4e2d\u7684\u6570\u636e\u8f6c\u6362\u6210\u4e0einput_types\u4e00\u81f4\u7684\u683c\u5f0f":2,"\u5c06\u88ab\u5206\u4e3a":85,"\u5c06\u8bad\u7ec3\u6587\u4ef6\u4e0e\u5207\u5206\u597d\u7684\u6570\u636e\u4e0a\u4f20\u5230\u5171\u4eab\u5b58\u50a8":80,"\u5c06\u8be5\u53e5\u8bdd\u5305\u542b\u7684\u6240\u6709\u5355\u8bcd\u5411\u91cf\u6c42\u5e73\u5747":89,"\u5c06\u8df3\u8fc7\u5206\u53d1\u9636\u6bb5\u76f4\u63a5\u542f\u52a8\u6240\u6709\u8282\u70b9\u7684\u96c6\u7fa4\u4f5c\u4e1a":73,"\u5c06\u8fd9\u79cd\u8de8\u8d8a\u65f6\u95f4\u6b65\u7684\u8fde\u63a5\u7528\u4e00\u4e2a\u7279\u6b8a\u7684\u795e\u7ecf\u7f51\u7edc\u5355\u5143\u5b9e\u73b0":61,"\u5c06\u8fdc\u7a0b\u4ed3\u5e93":66,"\u5c06\u900f\u660e":73,"\u5c06\u9700\u8981\u8f93\u51fa\u7684\u5c42\u4f5c\u4e3a":49,"\u5c06ip\u6392\u5e8f\u751f\u6210\u7684\u5e8f\u53f7\u4f5c\u4e3atrain":80,"\u5c06master\u5206\u652f\u7684\u5408\u5165commit\u6253\u4e0atag":42,"\u5c06paddlepaddle\u4fdd\u5b58\u7684\u6a21\u578b\u53c2\u6570\u8fd8\u539f\u56de\u660e\u6587\u65f6":51,"\u5c06recurr":50,"\u5c11\u4e8e5":73,"\u5c1a\u53ef":61,"\u5c31":61,"\u5c31\u4f1a\u751f\u6210\u975e\u5e38\u591a\u7684gener":2,"\u5c31\u53ef\u4ee5\u518d\u8fd0\u884c\u53e6\u4e00\u4e2anginx":54,"\u5c31\u53ef\u4ee5\u5c06\u6570\u636e\u4f20\u9001\u7ed9paddlepaddle\u4e86":2,"\u5c31\u53ef\u4ee5\u5c06\u8fd9\u4e9b\u6587\u4ef6\u6301\u4e45\u5316\u5b58\u50a8":78,"\u5c31\u53ef\u4ee5\u6309":65,"\u5c31\u5f88\u5bb9\u6613\u5bfc\u81f4\u5185\u5b58\u8d85\u9650":49,"\u5c31\u662f":61,"\u5c31\u662f\u6a21\u578b\u7684\u53c2\u6570":52,"\u5c31\u662f\u7528\u4e8e\u5c55\u793a\u4e0a\u8ff0\u5206\u6790\u5de5\u5177\u7684\u7528\u6cd5":72,"\u5c31\u80fd\u591f\u5f88\u65b9\u4fbf\u7684\u5b8c\u6210\u6570\u636e\u4e0b\u8f7d\u548c\u76f8\u5e94\u7684\u9884\u5904\u7406\u5de5\u4f5c":89,"\u5c31\u8fd9\u4e48\u7b80\u5355":54,"\u5c31\u901a\u5e38\u7684gpu\u6027\u80fd\u5206\u6790\u6765\u8bf4":72,"\u5c31\u9700\u8981\u5bf9\u8fd9\u4e2a\u7b2c\u4e09\u65b9\u8bed\u8a00\u589e\u52a0\u4e00\u4e9b\u5b9a\u4e49":33,"\u5c31\u9700\u8981\u9009\u62e9\u4f7f\u7528no":54,"\u5c3a\u5bf8":87,"\u5c3d\u65e9\u62a5\u9519":68,"\u5c40\u90e8\u5173\u8054\u6027\u8d28\u548c\u7a7a\u95f4\u4e0d\u53d8\u6027\u8d28":86,"\u5c42\u540e\u5f97\u5230\u6df1\u5ea6":92,"\u5c42\u548c\u8f93\u5165\u7684\u914d\u7f6e":67,"\u5c42\u6743\u91cd":87,"\u5c42\u6b21\u5316\u7684rnn":63,"\u5c42\u7279\u5f81":87,"\u5c42\u7684\u540d\u79f0\u4e0e":64,"\u5c42\u7684\u5927\u5c0f":67,"\u5c42\u7684\u7279\u5f81":87,"\u5c42\u7684\u7c7b\u578b":67,"\u5c42\u7684\u8f93\u5165":92,"\u5c42\u7684\u8f93\u5165\u548c\u8f93\u51fa\u4f5c\u4e3a\u4e0b\u4e00\u4e2a":92,"\u5c42\u7684\u8f93\u51fa\u88ab\u7528\u4f5c\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684":64,"\u5c42\u7ec4\u6210\u4e00\u5bf9":92,"\u5c45\u7136":61,"\u5c55\u793a\u4e86\u4e00\u79cd\u65b9\u6cd5":94,"\u5c55\u793a\u4e86\u4e0a\u8ff0\u7f51\u7edc\u6a21\u578b\u7684\u8bad\u7ec3\u6548\u679c":89,"\u5c55\u793a\u4e86\u5982\u4f55\u5c06\u6bcf\u4e2a\u7279\u5f81\u6620\u5c04\u5230\u4e00\u4e2a\u5411\u91cf":91,"\u5c5e\u4e8e\u8fd9\u4e00\u7c7b\u7684\u5b9e\u73b0":50,"\u5c5e\u6027":[68,92],"\u5d4c\u5165\u5c42":91,"\u5d4c\u5165\u7279\u5f81\u5b57\u5178":91,"\u5d4c\u5165\u7f16\u53f7\u4f1a\u6839\u636e\u5355\u8bcd\u6392\u5e8f":91,"\u5de5\u4f5c\u6a21\u5f0f":75,"\u5de5\u4f5c\u7a7a\u95f4":73,"\u5de5\u4f5c\u7a7a\u95f4\u4e2d\u7684":73,"\u5de5\u4f5c\u7a7a\u95f4\u6839\u76ee\u5f55":73,"\u5de5\u4f5c\u7a7a\u95f4\u76ee\u5f55\u7684\u5de5\u4f5c\u7a7a\u95f4":73,"\u5de5\u4f5c\u7a7a\u95f4\u914d\u7f6e":73,"\u5de5\u5177":93,"\u5de5\u5177\u4e2d\u7684\u811a\u672c":93,"\u5de5\u5177\u6765\u7ba1\u7406":66,"\u5de5\u5177\u94fe\u7684android":58,"\u5de5\u7a0b\u5e08":90,"\u5de6\u56fe\u6784\u9020\u7f51\u7edc\u6a21\u5757\u7684\u65b9\u5f0f\u88ab\u7528\u4e8e34\u5c42\u7684\u7f51\u7edc\u4e2d":87,"\u5de6\u8fb9\u662f":87,"\u5dee\u8bc4":89,"\u5df2\u6253\u5f00":66,"\u5df2\u7ecf\u5728\u96c6\u7fa4\u63d0\u4ea4\u73af\u5883\u4e2d\u5b8c\u6210\u8bbe\u7f6e":75,"\u5df2\u7ecf\u63d0\u4f9b\u4e86\u811a\u672c\u6765\u5e2e\u52a9\u60a8\u521b\u5efa\u8fd9\u4e24\u4e2a\u6587\u4ef6":73,"\u5e02\u573a":90,"\u5e02\u9762\u4e0a\u5df2\u7ecf\u6709nvidia\u6216\u7b2c\u4e09\u65b9\u63d0\u4f9b\u7684\u4f17\u591a\u5de5\u5177":72,"\u5e0c\u671b\u80fd\u8ba9\u6211\u4eec\u77e5\u6653":91,"\u5e26\u6709\u4e0b\u9762\u4e24\u4e2a\u6a21\u677f\u53c2\u6570":68,"\u5e2e\u52a9\u6211\u4eec\u5b8c\u6210\u5bf9\u8f93\u5165\u5e8f\u5217\u7684\u62c6\u5206":63,"\u5e2e\u52a9\u6211\u4eec\u66f4\u597d\u5730\u63cf\u8ff0\u6bb5\u843d":63,"\u5e2e\u52a9\u6211\u4eec\u6784\u9020\u4e00\u4e9b\u590d\u6742\u7684\u8f93\u5165\u4fe1\u606f":60,"\u5e38\u5e38\u51fa\u73b0":46,"\u5e38\u7528\u4f18\u5316\u7b97\u6cd5\u5305\u62ecmomentum":89,"\u5e38\u89c1\u7684\u53ef\u9009\u5b58\u50a8\u670d\u52a1\u5305\u62ec":78,"\u5e72\u51c0":61,"\u5e73\u53f0\u4e3a\u60f3\u89c2\u6d4b\u8bcd\u5411\u91cf\u7684\u7528\u6237\u63d0\u4f9b\u4e86\u5c06\u4e8c\u8fdb\u5236\u8bcd\u5411\u91cf\u6a21\u578b\u8f6c\u6362\u4e3a\u6587\u672c\u6a21\u578b\u7684\u529f\u80fd":85,"\u5e73\u5747\u6545\u969c\u4fee\u590d\u65f6\u95f4":20,"\u5e73\u5747\u6545\u969c\u7387":20,"\u5e73\u5747\u7279\u5f81\u56fe\u7684\u9ad8\u5ea6\u53ca\u5bbd\u5ea6":86,"\u5e74\u9f84":90,"\u5e74\u9f84\u4ece\u4e0b\u5217\u5217\u8868\u8303\u56f4\u4e2d\u9009\u53d6":90,"\u5e74\u9f84\u548c\u804c\u4e1a":91,"\u5e76\u4e0d\u4fdd\u8bc1":67,"\u5e76\u4e0d\u662f\u4f7f\u7528\u53cc\u5c42rnn\u89e3\u51b3\u5b9e\u9645\u7684\u95ee\u9898":61,"\u5e76\u4e0d\u662fkubernetes\u4e2d\u7684node\u6982\u5ff5":80,"\u5e76\u4e0d\u771f\u6b63\u7684\u548c":61,"\u5e76\u4e0d\u96be":65,"\u5e76\u4e14":[2,64],"\u5e76\u4e14\u4e66\u5199\u4e00\u4efd\u4ee3\u7801":69,"\u5e76\u4e14\u4f7f\u7528":34,"\u5e76\u4e14\u5185\u5c42\u7684\u5e8f\u5217\u64cd\u4f5c\u4e4b\u95f4\u72ec\u7acb\u65e0\u4f9d\u8d56":61,"\u5e76\u4e14\u5206\u522b\u91cd\u547d\u540d\u6587\u4ef6\u540e\u7f00":94,"\u5e76\u4e14\u52a0\u4e0a\u4e0b\u9762\u7684\u547d\u4ee4\u884c\u53c2\u6570":77,"\u5e76\u4e14\u5305\u62ecunit":66,"\u5e76\u4e14\u53ea\u6709\u4e00\u4e2a\u6743\u91cd":87,"\u5e76\u4e14\u53ef\u80fd\u4f1a\u52a0\u901f\u8bad\u7ec3\u8fc7\u7a0b":49,"\u5e76\u4e14\u540e\u7eed\u4ecd\u5728\u4e0d\u65ad\u6539\u8fdb":52,"\u5e76\u4e14\u542f\u52a8\u540e\u6267\u884c":54,"\u5e76\u4e14\u542f\u52a8\u8bad\u7ec3":80,"\u5e76\u4e14\u5728\u5185\u5b58\u8db3\u591f\u7684\u60c5\u51b5\u4e0b\u8d8a\u5927\u8d8a\u597d":2,"\u5e76\u4e14\u5728\u5e38\u89c1\u7684\u5e73\u53f0\u4e0a":33,"\u5e76\u4e14\u5728\u968f\u540e\u7684\u8bfb\u53d6\u6570\u636e\u8fc7\u7a0b\u4e2d\u586b\u5145\u8bcd\u8868":89,"\u5e76\u4e14\u5728dataprovider\u4e2d\u5b9e\u73b0\u5982\u4f55\u8bbf\u95ee\u8bad\u7ec3\u6587\u4ef6\u5217\u8868":1,"\u5e76\u4e14\u5b83\u4eec\u7684\u987a\u5e8f\u4e0e":87,"\u5e76\u4e14\u5bf9\u7528\u6237\u7684\u7279\u5f81\u505a\u540c\u6837\u7684\u64cd\u4f5c":91,"\u5e76\u4e14\u5c06\u9884\u5904\u7406\u597d\u7684\u6570\u636e\u96c6\u5b58\u653e\u5728":94,"\u5e76\u4e14\u5c55\u793a\u4e86\u5982\u4f55\u5229\u7528paddlepaddle\u6765\u89e3\u51b3\u4e00\u4e2a\u7ecf\u5178\u7684\u7ebf\u6027\u56de\u5f52\u95ee\u9898":56,"\u5e76\u4e14\u628a\u5404\u79cd\u5f00\u53d1\u5de5\u5177\u5b89\u88c5\u8fdb\u53bb":65,"\u5e76\u4e14\u628a\u7cfb\u7edf\u751f\u6210\u7684ca":28,"\u5e76\u4e14\u628a\u7ed3\u679c\u8fd4\u56depfsclient\u7aef":28,"\u5e76\u4e14\u67e5\u8be2paddlepaddle\u5355\u5143\u6d4b\u8bd5\u7684\u65e5\u5fd7":46,"\u5e76\u4e14\u7b2c\u4e8c\u4e2a\u662f\u53cd\u5411lstm":93,"\u5e76\u4e14\u7f16\u8bd1\u65f6\u9700\u8981\u6253\u5f00":68,"\u5e76\u4e14\u7f16\u8bd1\u80fd\u901a\u8fc7\u4ee3\u7801\u6837\u5f0f\u68c0\u67e5":66,"\u5e76\u4e14\u8ba9\u63a5\u53e3\u8131\u79bb\u5b9e\u73b0\u7ec6\u8282":33,"\u5e76\u4e14\u8bbe\u7f6e\u9ed8\u8ba4\u503c\u4e3a1":68,"\u5e76\u4e14\u8f93\u51fa\u4e00\u4e2a":66,"\u5e76\u4e14\u8fd0\u884c":65,"\u5e76\u4e14\u901a\u8fc7\u7ed9\u51fa\u5f53\u524d\u76ee\u6807\u5355\u8bcd\u6765\u9884\u6d4b\u4e0b\u4e00\u4e2a\u76ee\u6807\u5355\u8bcd":94,"\u5e76\u4e14\u96c6\u7fa4\u4f5c\u4e1a\u4e2d\u7684\u6240\u6709\u8282\u70b9\u5c06\u5728\u6b63\u5e38\u60c5\u51b5\u4e0b\u5904\u7406\u5177\u6709\u76f8\u540c\u903b\u8f91\u4ee3\u7801\u7684\u6587\u4ef6":73,"\u5e76\u4e14\u9700\u8981\u91cd\u5199\u57fa\u7c7b\u4e2d\u7684\u4ee5\u4e0b\u51e0\u4e2a\u865a\u51fd\u6570":67,"\u5e76\u4e14cpu":68,"\u5e76\u4e14softmax\u5c42\u7684\u4e24\u4e2a\u8f93\u5165\u4e5f\u4f7f\u7528\u4e86\u540c\u6837\u7684\u53c2\u6570":51,"\u5e76\u4f20\u5165\u76f8\u5e94\u7684\u547d\u4ee4\u884c\u53c2\u6570\u521d\u59cb\u5316paddlepaddl":4,"\u5e76\u4f7f\u7528":92,"\u5e76\u4f7f\u7528\u4e86dropout":89,"\u5e76\u4f7f\u7528\u8fd9\u4e2a\u795e\u7ecf\u7f51\u7edc\u6765\u5bf9\u56fe\u7247\u8fdb\u884c\u5206\u7c7b":86,"\u5e76\u521b\u5efa\u4e86\u4e00\u4e2a\u65b0\u6587\u4ef6":66,"\u5e76\u521b\u5efaoptim":56,"\u5e76\u521d\u59cb\u5316":68,"\u5e76\u5220\u9664":42,"\u5e76\u5220\u9664\u66f4\u65e9\u7684\u5feb\u7167":20,"\u5e76\u52a0\u8f7d\u5176\u4e2d\u7684\u53c2\u6570":20,"\u5e76\u53d1\u5e03\u5230pypi":42,"\u5e76\u5728\u4e58\u79ef\u7ed3\u679c\u4e0a\u518d\u52a0\u4e0a\u7ef4\u5ea6\u4e3a":67,"\u5e76\u5728\u6700\u5f00\u59cb\u521d\u59cb\u5316\u4e3a\u8d77\u59cb\u8bcd":64,"\u5e76\u5728\u7b14\u8bb0\u672c\u4e0a\u901a\u8fc7ssh\u4e0e\u5176\u8fde\u63a5":54,"\u5e76\u5728\u7c7b\u6784\u5efa\u51fd\u6570\u4e2d\u628a\u5b83\u653e\u5165\u4e00\u4e2a\u7c7b\u6210\u5458\u53d8\u91cf\u91cc":67,"\u5e76\u5728\u8be5layer\u91cc\u91c7\u7528\u7b2c\u4e00\u79cd\u65b9\u5f0f\u8bbe\u7f6e":50,"\u5e76\u5728\u96c6\u7fa4\u4e2d\u8fd0\u884c\u591a\u4e2a\u5206\u5e03\u5f0f\u6570\u636e\u5904\u7406\u4efb\u52a1":21,"\u5e76\u5728python\u811a\u672c\u4e2d\u5b8c\u6210\u4e0eoperator\u540c\u6837\u7684\u8ba1\u7b97\u903b\u8f91":68,"\u5e76\u5b89\u88c5\u4e86python":46,"\u5e76\u5b9a\u4e49\u5143\u7c7b":68,"\u5e76\u5bf9\u6bd4\u662f\u5426\u548c\u6b63\u5728\u5b89\u88c5\u7684\u540e\u7f00\u4e00\u81f4":46,"\u5e76\u5bf9\u76f8\u5e94\u7684\u53c2\u6570\u8c03\u7528":67,"\u5e76\u5c06\u5176\u6295\u5c04\u5230":64,"\u5e76\u5c06\u5b83\u4eec\u6309\u7167\u542f\u53d1\u4ee3\u4ef7":94,"\u5e76\u5c06\u5b83\u4eec\u653e\u5728":94,"\u5e76\u5c06\u6bcf\u8f6e\u7684\u6a21\u578b\u7ed3\u679c\u5b58\u653e\u5728":52,"\u5e76\u5c06\u8be5layer\u4e0a\u4e00\u65f6\u95f4\u6b65\u7684\u8f93\u51fa\u4f5c\u4e3a\u81ea\u8eab\u5f53\u524d\u65f6\u95f4\u6b65\u7684\u8f93\u51fa":50,"\u5e76\u5c06c":34,"\u5e76\u5c06develop\u548ctest\u6570\u636e\u5206\u522b\u653e\u5165\u4e0d\u540c\u7684\u6587\u4ef6\u5939":94,"\u5e76\u60f3\u4f7f\u7528gpu\u6765\u8bad\u7ec3\u8bbe\u7f6e\u4e3atru":93,"\u5e76\u628a\u5feb\u7167\u4fdd\u5b58\u5230\u8fd9\u4e2a\u76ee\u5f55\u4e0b":20,"\u5e76\u6307\u5b9abatch":94,"\u5e76\u63d0\u4f9b\u4e86\u7b80\u5355\u7684cache\u529f\u80fd":2,"\u5e76\u66f4\u6362job":47,"\u5e76\u6b22\u8fce\u60a8\u6765\u53c2\u4e0e\u8d21\u732e":93,"\u5e76\u6ca1\u6709paddle\u7279\u522b\u9700\u8981\u7684\u7279\u6027":33,"\u5e76\u6dfb\u52a0\u6ce8\u91ca":68,"\u5e76\u7279\u5316\u6a21\u677f\u53c2\u6570\u4e3a":68,"\u5e76\u7531":92,"\u5e76\u7ed9\u51fa\u5206\u7c7b\u7ed3\u679c":86,"\u5e76\u7ed9\u51fa\u7684\u76f8\u5173\u6a21\u578b\u683c\u5f0f\u7684\u5b9a\u4e49":85,"\u5e76\u81ea\u52a8\u7f16\u8bd1\u5bbf\u4e3b\u673a\u7248protoc\u53ef\u6267\u884c\u6587\u4ef6":[58,59],"\u5e76\u88ab\u53cd\u5411\u5904\u7406":92,"\u5e76\u88ab\u5b58\u50a8\u5728\u8bf8\u5982hadoop":21,"\u5e76\u89c2\u5bdf\u7ed3\u679c":72,"\u5e76\u89e3\u91ca\u4e86\u5404\u81ea\u542b\u4e49":68,"\u5e76\u8bb0\u5f55\u5b83\u7684\u7f16\u53f7":66,"\u5e76\u8bbe\u7f6e":73,"\u5e76\u8fdb\u884c\u521d\u59cb\u5316\u64cd\u4f5c":56,"\u5e76\u9002\u5e94github\u7684\u7279\u6027\u505a\u4e86\u4e00\u4e9b\u533a\u522b":42,"\u5e76\u9010\u6e10\u5c55\u793a\u66f4\u52a0\u6df1\u5165\u7684\u529f\u80fd":89,"\u5e76\u91cd\u65b0\u6253\u5305wheel\u5305":42,"\u5e76\u94fe\u63a5\u5230\u751f\u6210\u7684lib\u5e93\u4e2d":68,"\u5e8a\u4e0a\u7528\u54c1":61,"\u5e8a\u57ab":61,"\u5e8f\u5217\u4e2d\u542b\u6709\u5143\u7d20\u7684\u6570\u76ee\u540c":60,"\u5e8f\u5217\u5230\u5e8f\u5217":94,"\u5e8f\u5217\u6570\u636e\u662f\u81ea\u7136\u8bed\u8a00\u5904\u7406\u4efb\u52a1\u9762\u5bf9\u7684\u4e00\u79cd\u4e3b\u8981\u8f93\u5165\u6570\u636e\u7c7b\u578b":63,"\u5e8f\u5217\u662f\u4e00\u79cd\u5e38\u89c1\u7684\u6570\u636e\u7c7b\u578b":60,"\u5e8f\u5217\u751f\u6210\u4efb\u52a1\u5927\u591a\u9075\u5faaencod":63,"\u5e8f\u5217\u751f\u6210\u4efb\u52a1\u7684\u8f93\u5165":63,"\u5e8f\u5217\u7684\u5f00\u59cb":94,"\u5e8f\u5217\u7684\u6bcf\u4e2a\u5143\u7d20\u662f\u539f\u6765\u53cc\u5c42\u5e8f\u5217\u6bcf\u4e2asubseq\u5143\u7d20\u7684\u5e73\u5747\u503c":60,"\u5e8f\u5217\u7684\u7ed3\u5c3e":94,"\u5e8f\u5217\u7684\u7ed3\u675f":94,"\u5e8f\u5217\u8f93\u5165\u65f6\u7b49\u4e8e":49,"\u5e93":73,"\u5e93\u7684\u8def\u5f84":73,"\u5e94\u7528\u524d\u5411\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":64,"\u5e94\u7528\u53cd\u5411\u9012\u5f52\u795e\u7ecf\u7f51\u7edc":64,"\u5e94\u7528\u6a21\u578b":89,"\u5e94\u80fd\u53cd\u6620\u5f53\u524dcommit\u7684\u5185\u5bb9":66,"\u5e94\u8be5":61,"\u5e94\u8be5\u4e0e\u5b83\u7684memory\u540d\u5b57\u76f8\u540c":64,"\u5e94\u8be5\u8bf4\u8c22\u8c22":66,"\u5e94\u8be5\u8bfb\u53d6\u5f53\u524d\u76ee\u5f55\u4e0b\u7684":65,"\u5e94\u8be5\u964d\u4f4e\u5b66\u4e60\u7387":51,"\u5e95\u5c42\u8fdb\u7a0b":73,"\u5efa\u7acb\u4e00\u4e2a":66,"\u5efa\u8bae":[42,66],"\u5efa\u8bae\u5c06\u5176\u8bbe\u7f6e\u4e3a\u8f83\u5927":73,"\u5efa\u8bae\u5c06\u8be5\u53c2\u6570\u8bbe\u4e3atrue":75,"\u5f00\u53d1\u4e86\u6a21\u578b\u9884\u6d4b\u7684\u6837\u4f8b\u4ee3\u7801":34,"\u5f00\u53d1\u4eba\u5458\u4f7f\u7528":66,"\u5f00\u53d1\u4eba\u5458\u53ef\u4ee5\u5728docker\u5f00\u53d1\u955c\u50cf\u4e2d\u5f00\u53d1paddlepaddl":54,"\u5f00\u53d1\u8005\u4f7f\u7528":65,"\u5f00\u53d1\u8005\u4fee\u6539\u81ea\u5df1\u7684\u4ee3\u7801":42,"\u5f00\u53d1\u8005fork\u7684\u7248\u672c\u5e93\u4e2d":42,"\u5f00\u53d1\u8005fork\u7684\u7248\u672c\u5e93\u4f7f\u7528":42,"\u5f00\u53d1\u955c\u50cf":[54,66],"\u5f00\u53d1\u955c\u50cf\u5305\u542b\u4e86\u4ee5\u4e0b\u5de5\u5177":54,"\u5f00\u5934":68,"\u5f00\u59cb":52,"\u5f00\u59cb\u63d0\u4f9b\u670d\u52a1":20,"\u5f00\u59cb\u6807\u8bb0":64,"\u5f00\u59cb\u8bad\u7ec3\u6a21\u578b":89,"\u5f00\u59cb\u9636\u6bb5":72,"\u5f02\u6b65\u968f\u673a\u68af\u5ea6\u4e0b\u964d":74,"\u5f15\u5165\u4e86\u7c7b\u578b\u7684\u5934\u6587\u4ef6":34,"\u5f15\u5165lstm\u6a21\u578b\u4e3b\u8981\u662f\u4e3a\u4e86\u514b\u670d\u6d88\u5931\u68af\u5ea6\u7684\u95ee\u9898":93,"\u5f15\u5165paddlepaddle\u7684pydataprovider2\u5305":2,"\u5f15\u53d1":12,"\u5f15\u5bfc\u5c42":64,"\u5f15\u7528":73,"\u5f15\u7528memory\u5f97\u5230\u8fd9layer\u4e0a\u4e00\u65f6\u523b\u8f93\u51fa":63,"\u5f3a\u70c8\u63a8\u8350":61,"\u5f3a\u70c8\u63a8\u8350\u4f7f\u7528virtualenv\u6765\u521b\u9020\u4e00\u4e2a\u5e72\u51c0\u7684python\u73af\u5883":91,"\u5f52\u4e00\u5316\u6982\u7387\u5411\u91cf":64,"\u5f53":77,"\u5f53\u4f20\u9012\u76f8\u540c\u7684\u6837\u672c\u6570\u65f6":93,"\u5f53\u4f60\u6267\u884c\u547d\u4ee4":67,"\u5f53\u4fdd\u5b58\u7684\u7f51\u7edc\u53c2\u6570\u4e3afloat\u7c7b\u578b\u65f6\u4e3a4":51,"\u5f53\u51fd\u6570\u8fd4\u56de\u7684\u65f6\u5019":2,"\u5f53\u524d\u5355\u8bcd\u5728\u76f8\u6bd4\u4e4b\u4e0b\u603b\u662f\u88ab\u5f53\u4f5c\u771f\u503c":94,"\u5f53\u524d\u5355\u8bcd\u662f\u89e3\u7801\u5668\u6700\u540e\u4e00\u6b65\u7684\u8f93\u51fa":94,"\u5f53\u524d\u65f6\u95f4\u6b65\u5904\u7684memory\u7684\u8f93\u51fa\u4f5c\u4e3a\u4e0b\u4e00\u65f6\u95f4\u6b65memory\u7684\u8f93\u5165":64,"\u5f53\u524d\u7684\u5b66\u4e60\u7387\u4e3a\u6240\u8bbe\u7f6e":51,"\u5f53\u524d\u7684\u5b9e\u73b0\u65b9\u5f0f\u4e0b":67,"\u5f53\u524d\u7684\u8f93\u5165y\u548c\u4e0a\u4e00\u4e2a\u65f6\u95f4\u6b65\u7684\u8f93\u51farnn_state\u505a\u4e86\u4e00\u4e2a\u5168\u94fe\u63a5":61,"\u5f53\u524d\u8bc4\u4f30\u4e2d":94,"\u5f53\u524dbatch\u7684cost":94,"\u5f53\u524dlog_period\u4e2abatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747\u5206\u7c7b\u9519\u8bef\u7387":89,"\u5f53\u524dlog_period\u4e2abatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747cost":89,"\u5f53\u529f\u80fd\u5206\u652f\u5f00\u53d1\u5b8c\u6bd5\u540e":42,"\u5f53\u5728\u7f51\u7edc\u5c42\u914d\u7f6e\u4e2d\u8bbe\u7f6e":75,"\u5f53\u5728\u8bad\u7ec3\u914d\u7f6e\u4e2d\u8bbe\u7f6e":75,"\u5f53\u5bb9\u5668\u56e0\u4e3a\u5404\u79cd\u539f\u56e0\u88ab\u9500\u6bc1\u65f6":78,"\u5f53\u5df2\u8bad\u7ec3\u6837\u672c\u6570\u5927\u4e8e1000\u5c0f\u4e8e\u7b49\u4e8e2000\u65f6":51,"\u5f53\u5df2\u8bad\u7ec3\u6837\u672c\u6570\u5927\u4e8e2000\u65f6":51,"\u5f53\u5df2\u8bad\u7ec3\u6837\u672c\u6570\u5c0f\u4e8e\u7b49\u4e8e1000\u65f6":51,"\u5f53\u5df2\u8bad\u7ec3pass\u6570\u5927\u4e8e1\u5c0f\u4e8e\u7b49\u4e8e2\u65f6":51,"\u5f53\u5df2\u8bad\u7ec3pass\u6570\u5927\u4e8e2\u65f6":51,"\u5f53\u5df2\u8bad\u7ec3pass\u6570\u5c0f\u4e8e\u7b49\u4e8e1\u65f6":51,"\u5f53\u6240\u6709pod\u90fd\u5904\u4e8erunning\u72b6\u6001":80,"\u5f53\u6a21\u578b\u53c2\u6570\u4e0d\u5b58\u5728\u65f6":75,"\u5f53\u6a21\u578b\u8bad\u7ec3\u597d\u4e86\u4e4b\u540e":89,"\u5f53\u6a21\u5f0f\u4e3a":75,"\u5f53\u7136":72,"\u5f53\u7136\u53ef\u4ee5":65,"\u5f53\u7528\u6237\u4f7f\u7528\u5b8c\u8fd9\u4e2a\u53c2\u6570\u540e":34,"\u5f53\u7528\u6237\u6ca1\u6709\u663e\u5f0f\u8bbe\u5b9a\u65f6":50,"\u5f53\u7f51\u7edc\u5c42\u7528\u4e00\u4e2a\u6279\u6b21\u505a\u8bad\u7ec3\u65f6":67,"\u5f53\u89e3\u8bfb\u6bcf\u4e00\u4e2a":64,"\u5f53\u8bad\u7ec3\u6570\u636e\u975e\u5e38\u591a\u65f6":2,"\u5f53\u8d85\u8fc7\u8be5\u9608\u503c\u65f6":75,"\u5f53\u8f93\u5165\u662f\u7ef4\u5ea6\u5f88\u9ad8\u7684\u7a00\u758f\u6570\u636e\u65f6":77,"\u5f53\u9700\u8981\u5b8c\u6210\u8ba1\u7b97\u65f6":69,"\u5f53\u9700\u8981\u5feb\u901f\u6216\u8005\u9891\u7e41\u7684\u8bc4\u4f30\u65f6":94,"\u5f53classif":94,"\u5f53destination\u6587\u4ef6\u4e0d\u5b58\u5728\u6216\u8005\u5927\u5c0f\u548csource\u6587\u4ef6\u4e0d\u4e00\u81f4\u65f6":28,"\u5f53n1":49,"\u5f62\u6210recurr":63,"\u5f62\u6210recurrent\u8fde\u63a5":63,"\u5f62\u72b6":87,"\u5f88":[61,89],"\u5f88\u591a":[61,65],"\u5f88\u591a\u5f00\u53d1\u8005\u4f1a\u4f7f\u7528\u8fdc\u7a0b\u7684\u5b89\u88c5\u6709gpu\u7684\u670d\u52a1\u5668\u5de5\u4f5c":54,"\u5f88\u5b89\u9759":61,"\u5f88\u5e72\u51c0":61,"\u5f88\u65b9\u4fbf":61,"\u5f88\u6709\u53ef\u80fd\u5b9e\u9645\u5e94\u7528\u5c31\u662f\u6ca1\u6709\u6309\u7167\u60a8\u7684\u9884\u671f\u60c5\u51b5\u8fd0\u884c":72,"\u5f88\u6709\u53ef\u80fd\u662f\u975e\u72ec\u5360\u65b9\u5f0f\u6267\u884c\u5bfc\u81f4\u7684\u7aef\u53e3\u51b2\u7a81":47,"\u5f88\u9002\u5408\u6784\u5efa\u7528\u4e8e\u7406\u89e3\u56fe\u7247\u5185\u5bb9\u7684\u6a21\u578b":86,"\u5f88\u96be\u4fdd\u8bc1\u591a\u8bed\u8a00\u4ee3\u7801\u98ce\u683c\u7684\u4e00\u81f4\u6027":33,"\u5f88\u96be\u6574\u4f53\u4fee\u6b63":2,"\u5f8b\u5e08":90,"\u5f97":61,"\u5f97\u4f7f\u7528":33,"\u5f97\u5230\u53e5\u5b50\u7684\u8868\u793a":89,"\u5f97\u5230\u6700\u597d\u8f6e\u6b21\u4e0b\u7684\u6a21\u578b":91,"\u5f97\u5230\u8f93\u51fa\u503c":68,"\u5faa\u73af\u5c55\u5f00\u7684\u6bcf\u4e2a\u65f6\u95f4\u6b65\u603b\u662f\u80fd\u591f\u5f15\u7528\u6240\u6709\u8f93\u5165":63,"\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u4e2d":64,"\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u4f5c\u4e3a\u4f7f\u7528":64,"\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u548c":64,"\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u5728\u6bcf\u4e2a\u65f6\u95f4\u6b65\u9aa4\u987a\u5e8f\u5730\u5904\u7406\u5e8f\u5217":64,"\u5faa\u73af\u7f51\u7edc\u4ece":64,"\u5fc5\u8981":34,"\u5fc5\u987b":67,"\u5fc5\u987b\u4e00\u81f4":2,"\u5fc5\u987b\u4f7f\u7528python\u5173\u952e\u8bcd":2,"\u5fc5\u987b\u5c06\u524d\u4e00\u4e2a\u5b50\u53e5\u7684\u6700\u540e\u4e00\u4e2a\u5143\u7d20":61,"\u5fc5\u987b\u6307\u5411\u4e00\u4e2apaddlepaddle\u5b9a\u4e49\u7684lay":63,"\u5fc5\u987b\u662f\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":63,"\u5fc5\u987b\u662f\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":60,"\u5fc5\u987b\u7531\u53ea\u8bfbmemory\u7684":64,"\u5fc5\u987b\u8bbe\u7f6e\u4e3a":58,"\u5fc5\u987b\u914d\u7f6e\u4e3a":59,"\u5fd8\u8bb0\u95e8\u548c\u8f93\u51fa\u95e8":93,"\u5feb":[61,93],"\u5feb\u901f\u5165\u95e8":88,"\u5feb\u901f\u5728\u672c\u5730\u542f\u52a8\u4e00\u4e2a\u5355\u673a\u7684kubernetes\u670d\u52a1\u5668":78,"\u5feb\u901f\u90e8\u7f72\u96c6\u7fa4":78,"\u6027\u4ef7\u6bd4":61,"\u6027\u522b":[90,91],"\u6027\u80fd\u5206\u6790":72,"\u6027\u80fd\u5206\u6790\u5de5\u5177\u662f\u7528\u4e8e\u7ed9\u5e94\u7528\u7a0b\u5e8f\u7684\u6027\u80fd\u505a\u5b9a\u91cf\u5206\u6790\u7684":72,"\u6027\u80fd\u5206\u6790\u662f\u6027\u80fd\u4f18\u5316\u7684\u5173\u952e\u4e00\u6b65":72,"\u6027\u80fd\u548c\u628a\u7f16\u8bd1\u5de5\u5177\u5b89\u88c5\u5728\u672c\u673a\u8fd0\u884c\u4e00\u6837":65,"\u6027\u80fd\u8c03\u4f18":74,"\u603b\u4f53\u6765\u8bf4":61,"\u603b\u8ba1\u7684\u53c2\u6570\u4e2a\u6570":85,"\u603b\u8bc4\u520610\u5206":93,"\u6050\u6016\u7247":90,"\u60a8\u4f1a\u5728\u63a5\u4e0b\u6765\u7684\u90e8\u5206\u4e2d\u83b7\u5f97\u66f4\u591a\u7684\u7ec6\u8282\u4ecb\u7ecd":72,"\u60a8\u53ef\u4ee5\u4efb\u610f\u4f7f\u7528\u4e00\u4e2a\u6216\u4e24\u4e2a\u6765\u5bf9\u611f\u5174\u8da3\u7684\u4ee3\u7801\u6bb5\u505a\u6027\u80fd\u5206\u6790":72,"\u60a8\u53ef\u4ee5\u5bfc\u5165":72,"\u60a8\u53ef\u4ee5\u91c7\u7528\u4e0b\u9762\u4e94\u4e2a\u6b65\u9aa4":72,"\u60a8\u5c06\u4e86\u89e3\u5982\u4f55":64,"\u60a8\u5c31\u80fd\u83b7\u5f97\u5982\u4e0b\u7684\u5206\u6790\u7ed3\u679c":72,"\u60a8\u6309\u5982\u4e0b\u6b65\u9aa4\u64cd\u4f5c\u5373\u53ef":72,"\u60a8\u6700\u597d\u5148\u786e\u8ba4":72,"\u60a8\u9700\u8981\u66f4\u6539":54,"\u60a8\u9996\u5148\u9700\u8981\u5728\u76f8\u5173\u4ee3\u7801\u6bb5\u4e2d\u52a0\u5165":72,"\u60ac\u7591\u7247":90,"\u60c5\u6001\u52a8\u8bcd":92,"\u60c5\u611f\u5206\u6790":[42,88],"\u60c5\u611f\u5206\u6790\u4e5f\u5e38\u7528\u4e8e\u57fa\u4e8e\u5927\u91cf\u8bc4\u8bba\u548c\u4e2a\u4eba\u535a\u5ba2\u6765\u76d1\u63a7\u793e\u4f1a\u5a92\u4f53":93,"\u60c5\u611f\u5206\u6790\u662f\u81ea\u7136\u8bed\u8a00\u7406\u89e3\u4e2d\u6700\u5178\u578b\u7684\u95ee\u9898\u4e4b\u4e00":93,"\u60c5\u611f\u5206\u6790\u6709\u8bb8\u591a\u5e94\u7528\u573a\u666f":93,"\u60ca\u9669\u7535\u5f71":90,"\u60f3\u4e86\u89e3\u66f4\u591a\u7ec6\u8282\u53ef\u4ee5\u53c2\u8003pydataprovider\u90e8\u5206\u7684\u6587\u6863":93,"\u610f\u5473\u7740\u4e0d\u540c\u65f6\u95f4\u6b65\u7684\u8f93\u5165\u90fd\u662f\u76f8\u540c\u7684\u503c":64,"\u610f\u601d\u662f\u4e0d\u4f7f\u7528\u5e73\u5747\u53c2\u6570\u6267\u884c\u6d4b\u8bd5":75,"\u610f\u601d\u662f\u4e0d\u4fdd\u5b58\u7ed3\u679c":75,"\u610f\u601d\u662f\u4f7f\u7528\u7b2ctest":75,"\u610f\u601d\u662f\u5728gpu\u6a21\u5f0f\u4e0b\u4f7f\u75284\u4e2agpu":75,"\u611f\u89c9":61,"\u620f\u5267":90,"\u6210\u529f\u8bad\u7ec3\u4e14\u9000\u51fa\u7684pod\u6570\u76ee\u4e3a3\u65f6":80,"\u6210\u5458":68,"\u6211\u4eec\u4e0d\u80fd\u901a\u8fc7\u5e38\u89c4\u7684\u68af\u5ea6\u68c0\u67e5\u7684\u65b9\u5f0f\u6765\u8ba1\u7b97\u68af\u5ea6":67,"\u6211\u4eec\u4e3b\u8981\u4f1a\u4ecb\u7ecdnvprof\u548cnvvp":72,"\u6211\u4eec\u4e5f\u53ef\u4ee5\u786e\u5b9a\u6bcf\u4e00\u4e2a\u53c2\u6570\u7684\u7c7b\u578b":34,"\u6211\u4eec\u4ec5\u4ec5\u5bf9\u795e\u7ecf\u7f51\u7edc\u7684\u8f93\u5165\u8fdb\u884c\u4e86\u63cf\u8ff0":56,"\u6211\u4eec\u4ec5\u4ec5\u662f\u5c06\u6bcf\u4e2a\u7279\u5f81\u79cd\u7c7b\u6620\u5c04\u5230\u4e00\u4e2a\u7279\u5f81\u5411\u91cf\u4e2d":91,"\u6211\u4eec\u4ec5\u6709\u4e00\u4e2a\u8f93\u5165":67,"\u6211\u4eec\u4ec5\u7528":91,"\u6211\u4eec\u4ecb\u7ecd\u5982\u4f55\u5728":79,"\u6211\u4eec\u4ecb\u7ecd\u5982\u4f55\u5728kubernetes\u96c6\u7fa4\u4e0a\u8fdb\u884c\u5206\u5e03\u5f0fpaddlepaddle\u8bad\u7ec3\u4f5c\u4e1a":80,"\u6211\u4eec\u4ece\u63d0\u524d\u7ed9\u5b9a\u7684\u7c7b\u522b\u96c6\u5408\u4e2d\u9009\u62e9\u5176\u6240\u5c5e\u7c7b\u522b":89,"\u6211\u4eec\u4ee5mnist\u624b\u5199\u8bc6\u522b\u4e3a\u4f8b":2,"\u6211\u4eec\u4f1a\u53d1\u73b0\u6570\u636e\u96c6":94,"\u6211\u4eec\u4f1a\u5728":54,"\u6211\u4eec\u4f1a\u7ee7\u7eed\u4f7f\u7528\u73b0\u6709\u7684\u5185\u5b58\u5757":67,"\u6211\u4eec\u4f1a\u91cd\u65b0\u5206\u914d\u5185\u5b58":67,"\u6211\u4eec\u4f7f\u7528":[67,73,93],"\u6211\u4eec\u4f7f\u7528\u4e0d\u540c\u7684layer\u8fdb\u884c\u7ec4\u5408":56,"\u6211\u4eec\u4f7f\u7528\u4e86":61,"\u6211\u4eec\u4f7f\u7528\u4e86\u4e00\u4e2a\u7f16\u89e3\u7801\u6a21\u578b\u7684\u6269\u5c55":94,"\u6211\u4eec\u4f7f\u7528\u4e86\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":93,"\u6211\u4eec\u4f7f\u7528\u5176\u4e2d\u7684":94,"\u6211\u4eec\u4f7f\u7528\u52a8\u6001\u5e93\u6765\u5206\u53d1paddl":33,"\u6211\u4eec\u4f7f\u7528\u6700\u5927\u6982\u7387\u7684\u6807\u7b7e\u4f5c\u4e3a\u7ed3\u679c":92,"\u6211\u4eec\u4f7f\u7528\u96c6\u675f\u641c\u7d22":94,"\u6211\u4eec\u4f7f\u7528paddlepaddle\u5728ilsvrc\u7684\u9a8c\u8bc1\u96c6\u517150":87,"\u6211\u4eec\u5047\u8bbe\u4e00\u53f0\u673a\u5668\u4e0a\u67094\u4e2agpu":77,"\u6211\u4eec\u5047\u8bbe\u623f\u4ea7\u7684\u4ef7\u683c":52,"\u6211\u4eec\u5148\u4ece\u4e00\u6761\u968f\u673a\u7684\u76f4\u7ebf":52,"\u6211\u4eec\u5148\u8c03\u7528\u6bcf\u4e2a":69,"\u6211\u4eec\u51c6\u5907\u4e86\u4e00\u4e2a\u811a\u672c":86,"\u6211\u4eec\u5373\u53ef\u5b8c\u6210\u795e\u7ecf\u7f51\u7edc\u7684\u642d\u5efa":56,"\u6211\u4eec\u53ea\u4f7f\u7528\u5df2\u7ecf\u6807\u6ce8\u8fc7\u7684\u8bad\u7ec3\u96c6\u548c\u6d4b\u8bd5\u96c6":93,"\u6211\u4eec\u53ea\u6240\u4ee5\u4f7f\u7528lstm\u6765\u6267\u884c\u8fd9\u4e2a\u4efb\u52a1\u662f\u56e0\u4e3a\u5176\u6539\u8fdb\u7684\u8bbe\u8ba1\u5e76\u4e14\u5177\u6709\u95e8\u673a\u5236":93,"\u6211\u4eec\u53ea\u6f14\u793a\u4e00\u4e2a\u5355\u673a\u4f5c\u4e1a":79,"\u6211\u4eec\u53ea\u9700\u8981":65,"\u6211\u4eec\u53ea\u9700\u8981\u4f7f\u7528lstm":61,"\u6211\u4eec\u53ea\u9700\u8981\u8fd0\u884c":89,"\u6211\u4eec\u53ea\u9700\u8981\u8fd0\u884c\u4e0b\u9762\u547d\u4ee4\u628a\u7f16\u8bd1\u597d\u7684paddlepaddle\u6253\u5305\u6210\u4e00\u4e2a":66,"\u6211\u4eec\u53ea\u9700\u8981\u914d\u7f6e":65,"\u6211\u4eec\u53ef\u4ee5":65,"\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528":86,"\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528\u5176\u4ed6layer\u8fdb\u884c\u7ec4\u5408":56,"\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528\u5b83\u6765\u751f\u6210\u5e8f\u5217":64,"\u6211\u4eec\u53ef\u4ee5\u521b\u5efatrainer\u6765\u5bf9\u7f51\u7edc\u8fdb\u884c\u8bad\u7ec3":56,"\u6211\u4eec\u53ef\u4ee5\u53c2\u8003tensorflow\u7684":69,"\u6211\u4eec\u53ef\u4ee5\u5728":66,"\u6211\u4eec\u53ef\u4ee5\u5b9a\u4e49\u5982\u4e0b\u7684layer\u7ec4\u5408":56,"\u6211\u4eec\u53ef\u4ee5\u5b9a\u4e49\u5982\u4e0blayer\u6765\u63cf\u8ff0\u795e\u7ecf\u7f51\u7edc\u7684\u8f93\u5165":56,"\u6211\u4eec\u53ef\u4ee5\u5c06":73,"\u6211\u4eec\u53ef\u4ee5\u6309\u7167\u5982\u4e0b\u5c42\u6b21\u5b9a\u4e49\u975e\u5e8f\u5217":60,"\u6211\u4eec\u53ef\u4ee5\u751f\u6210":91,"\u6211\u4eec\u53ef\u4ee5\u76f4\u63a5\u89c2\u5bdf\u6a21\u578b\u7684\u53c2\u6570\u662f\u5426\u7b26\u5408\u9884\u671f\u6765\u8fdb\u884c\u68c0\u9a8c":52,"\u6211\u4eec\u53ef\u4ee5\u770b\u5230\u5728\u76ee\u5f55":93,"\u6211\u4eec\u53ef\u4ee5\u8bbe\u8ba1\u642d\u5efa\u4e00\u4e2a\u7075\u6d3b\u7684":63,"\u6211\u4eec\u53ef\u4ee5\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4\u6765\u505ableu\u8bc4\u4f30":94,"\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u4ee5\u4e0b\u547d\u4ee4\u6765\u8bad\u7ec3\u6a21\u578b":94,"\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u4ee5\u4e0b\u547d\u4ee4\u6765\u8fdb\u884c\u4ece\u6cd5\u8bed\u5230\u82f1\u8bed\u7684\u6587\u672c\u7ffb\u8bd1":94,"\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u5982\u4e0b\u547d\u4ee4\u8fdb\u884c\u9884\u5904\u7406\u5de5\u4f5c":86,"\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u65e5\u5fd7\u67e5\u770b\u5bb9\u5668\u8bad\u7ec3\u7684\u60c5\u51b5":80,"\u6211\u4eec\u5728":69,"\u6211\u4eec\u57285\u5929\u91cc\u8bad\u7ec3\u4e8616\u4e2apass":94,"\u6211\u4eec\u5728\u51fd\u6570\u7684\u7ed3\u5c3e\u8fd4\u56de":64,"\u6211\u4eec\u5728\u62e5\u670950\u4e2a\u8282\u70b9\u7684\u96c6\u7fa4\u4e2d\u8bad\u7ec3\u6a21\u578b":94,"\u6211\u4eec\u5728\u8bad\u7ec3\u4e4b\u524d\u9700\u8981\u5e38\u89c1\u4e00\u4e2a\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":94,"\u6211\u4eec\u5728initialzier\u51fd\u6570\u91cc\u521d\u59cb\u5316\u8bcd\u8868":89,"\u6211\u4eec\u5bf9\u6a21\u578b\u8fdb\u884c\u4e86\u4ee5\u4e0b\u66f4\u6539":64,"\u6211\u4eec\u5c06":[80,91],"\u6211\u4eec\u5c06\u4e00\u6bb5\u8bdd\u770b\u6210\u53e5\u5b50\u7684\u6570\u7ec4":61,"\u6211\u4eec\u5c06\u4ecb\u7ecd\u5982\u4f55\u542f\u52a8\u5206\u5e03\u5f0f\u8bad\u7ec3\u4f5c\u4e1a":79,"\u6211\u4eec\u5c06\u4ee5":[73,89],"\u6211\u4eec\u5c06\u4ee5\u6700\u57fa\u672c\u7684\u903b\u8f91\u56de\u5f52\u7f51\u7edc\u4f5c\u4e3a\u8d77\u70b9":89,"\u6211\u4eec\u5c06\u4f7f\u7528":64,"\u6211\u4eec\u5c06\u4f7f\u7528\u7b80\u5355\u7684":64,"\u6211\u4eec\u5c06\u4f7f\u7528cifar":86,"\u6211\u4eec\u5c06\u539f\u59cb\u6570\u636e\u7684\u6bcf\u4e00\u7ec4":61,"\u6211\u4eec\u5c06\u5728\u540e\u9762\u4ecb\u7ecd\u8bad\u7ec3\u548c\u9884\u6d4b\u6d41\u7a0b\u7684\u811a\u672c":89,"\u6211\u4eec\u5c06\u5b83\u4eec\u5212\u5206\u4e3a\u4e0d\u540c\u7684\u7c7b\u522b":74,"\u6211\u4eec\u5c06\u5bf9\u8fd9\u4e24\u4e2a\u6b65\u9aa4\u7ed9\u51fa\u4e86\u8be6\u7ec6\u7684\u89e3\u91ca":89,"\u6211\u4eec\u5c06\u653e\u7f6e\u4f9d\u8d56\u5e93":73,"\u6211\u4eec\u5c06\u8bc4\u5206\u5206\u6210\u4e24\u90e8\u5206":91,"\u6211\u4eec\u5c06\u9610\u91ca\u5982\u4f55\u5728\u96c6\u7fa4\u4e0a\u8fd0\u884c\u5206\u5e03\u5f0f":73,"\u6211\u4eec\u5c31\u53ef\u4ee5\u7740\u624b\u5bf9\u5206\u7c7b\u6a21\u578b\u8fdb\u884c\u8bad\u7ec3\u4e86":86,"\u6211\u4eec\u5c31\u53ef\u4ee5\u8bad\u7ec3\u6a21\u578b\u4e86":89,"\u6211\u4eec\u5c31\u53ef\u4ee5\u8fdb\u884c\u9884\u6d4b\u4e86":89,"\u6211\u4eec\u5c31\u5b8c\u6210\u4e86\u4e00\u6b21\u4ee3\u7801\u8d21\u732e\u7684\u8fc7\u7a0b":66,"\u6211\u4eec\u5c55\u793a\u5982\u4f55\u7528paddlepaddle\u89e3\u51b3":52,"\u6211\u4eec\u5df2\u7ecf\u5b9e\u73b0\u4e86\u5927\u591a\u6570\u5e38\u7528\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u67b6\u6784":64,"\u6211\u4eec\u5e0c\u671b\u80fd\u591f\u68c0\u9a8c\u6a21\u578b\u7684\u597d\u574f":52,"\u6211\u4eec\u5e94\u5f53\u4f1a\u5f97\u5230\u4e00\u4e2a\u540d\u4e3acifar":86,"\u6211\u4eec\u5efa\u8bae\u4f60\u4e3a\u4f60\u7684python\u5c01\u88c5\u5b9e\u73b0\u4e00\u4e2a":67,"\u6211\u4eec\u5efa\u8bae\u4f60\u5728\u5199\u65b0\u7f51\u7edc\u5c42\u65f6\u628a\u6d4b\u8bd5\u4ee3\u7801\u653e\u5165\u65b0\u7684\u6587\u4ef6\u4e2d":67,"\u6211\u4eec\u5efa\u8bae\u4f7f\u7528\u7b2c\u4e8c\u7c7b\u5b9e\u73b0":50,"\u6211\u4eec\u603b\u7ed3\u4e86\u5404\u4e2a\u7f51\u7edc\u7684\u590d\u6742\u5ea6\u548c\u6548\u679c":89,"\u6211\u4eec\u611f\u8c22":94,"\u6211\u4eec\u628apaddlepaddle\u7684\u7f16\u8bd1\u73af\u5883\u6253\u5305\u6210\u4e00\u4e2a\u955c\u50cf":54,"\u6211\u4eec\u63a8\u8350\u4f7f\u7528":54,"\u6211\u4eec\u63a8\u8350\u4f7f\u7528\u6700\u65b0\u7248\u672c\u7684cudnn":53,"\u6211\u4eec\u63d0\u4f9b\u4e24\u4e2a\u7f51\u7edc":93,"\u6211\u4eec\u63d0\u4f9b\u4e24\u4e2a\u8f6c\u6362\u65b9\u5f0f":21,"\u6211\u4eec\u63d0\u4f9b\u4e86\u4e00\u4e2a\u6570\u636e\u9884\u5904\u7406\u811a\u672c":93,"\u6211\u4eec\u63d0\u4f9b\u4e86\u4e00\u4e2a\u793a\u4f8b\u811a\u672c":87,"\u6211\u4eec\u63d0\u4f9b\u4e86\u56fd\u5185\u7684\u955c\u50cf\u670d\u52a1\u5668\u4f9b\u5927\u5bb6\u4f7f\u7528":54,"\u6211\u4eec\u63d0\u4f9b\u4e86\u811a\u672c\u6765\u6784\u5efa\u5b57\u5178\u548c\u9884\u5904\u7406\u6570\u6910":93,"\u6211\u4eec\u63d0\u4f9b\u4e86c":87,"\u6211\u4eec\u63d0\u4f9b\u53ef\u4ee5\u76f4\u63a5\u8fd0\u884cpaddlepaddl":54,"\u6211\u4eec\u63d0\u51fa\u4e86chunk\u7684\u6982\u5ff5":28,"\u6211\u4eec\u65e0\u6cd5\u5728docker\u5bb9\u5668\u4e2d":54,"\u6211\u4eec\u662f\u5bf9\u6bcf\u4e00\u4e2a\u5b50\u5e8f\u5217\u53d6\u6700\u540e\u4e00\u4e2a\u5143\u7d20":61,"\u6211\u4eec\u6700\u7ec8\u7684\u52a8\u6001\u5e93\u4e2d\u4e0d\u5d4c\u5165python\u6216\u8005\u5176\u4ed6\u4efb\u4f55\u8bed\u8a00\u7684\u89e3\u91ca\u5668":33,"\u6211\u4eec\u6709\u4e00\u4e2a\u5e8f\u5217\u4f5c\u4e3a\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u72b6\u6001":64,"\u6211\u4eec\u7528":90,"\u6211\u4eec\u7528\u4ee5\u4e0b\u7684\u4e00\u4e9b":91,"\u6211\u4eec\u7528\u7f16\u53f7\u4f5c\u4e3akei":91,"\u6211\u4eec\u7528paddlepaddle\u89e3\u51b3\u4e86\u5355\u53d8\u91cf\u7ebf\u6027\u56de\u5f52\u95ee\u9898":52,"\u6211\u4eec\u7684":65,"\u6211\u4eec\u7684\u5b57\u5178\u4f7f\u7528\u5185\u90e8\u7684\u5206\u8bcd\u5de5\u5177\u5bf9\u767e\u5ea6\u77e5\u9053\u548c\u767e\u5ea6\u767e\u79d1\u7684\u8bed\u6599\u8fdb\u884c\u5206\u8bcd\u540e\u4ea7\u751f":85,"\u6211\u4eec\u7684\u6807\u51c6\u5f00\u53d1\u6d41\u7a0b\u662f\u628a\u8fd9\u4e9b\u5de5\u5177\u90fd\u88c5\u8fdb\u4e00\u4e2adocker":66,"\u6211\u4eec\u7684\u8bad\u7ec3\u66f2\u7ebf\u5982\u4e0b":92,"\u6211\u4eec\u770b\u4e00\u4e0b\u5355\u5c42rnn\u7684\u914d\u7f6e":61,"\u6211\u4eec\u770b\u4e00\u4e0b\u8bed\u4e49\u76f8\u540c\u7684\u53cc\u5c42rnn\u7684\u7f51\u7edc\u914d\u7f6e":61,"\u6211\u4eec\u771f\u8bda\u5730\u611f\u8c22\u60a8\u7684\u5173\u6ce8":93,"\u6211\u4eec\u771f\u8bda\u5730\u611f\u8c22\u60a8\u7684\u8d21\u732e":66,"\u6211\u4eec\u79f0\u4e4b\u4e3a\u4e00\u4e2a0\u5c42\u7684\u5e8f\u5217":60,"\u6211\u4eec\u8bbe\u8ba1\u8bf4\u660e\u4e86\u540d\u4e3afilemanager\u7cfb\u7edf":28,"\u6211\u4eec\u8c03\u7528\u4e86eigenvector\u7684flatten\u63a5\u53e3":69,"\u6211\u4eec\u8fd8\u53ef\u4ee5\u767b\u5f55\u5230\u5bbf\u4e3b\u673a\u4e0a\u67e5\u770b\u8bad\u7ec3\u7ed3\u679c":79,"\u6211\u4eec\u8fd8\u5c06\u7f16\u7801\u5411\u91cf\u6295\u5c04\u5230":64,"\u6211\u4eec\u9009\u53d6\u5355\u53cc\u5c42\u5e8f\u5217\u914d\u7f6e\u4e2d\u7684\u4e0d\u540c\u90e8\u5206":61,"\u6211\u4eec\u9009\u62e9":21,"\u6211\u4eec\u901a\u5e38\u501f\u52a9":68,"\u6211\u4eec\u901a\u5e38\u5728\u6240\u6709\u8282\u70b9\u4e0a\u521b\u5efa\u4e00\u4e2a":73,"\u6211\u4eec\u901a\u5e38\u5c06\u4e00\u53e5\u8bdd\u7406\u89e3\u6210\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217":61,"\u6211\u4eec\u901a\u8fc7\u8bfb\u53d6":80,"\u6211\u4eec\u9075\u5faa":94,"\u6211\u4eec\u90fd\u63d0\u4f9bpython\u7684\u8f6c\u6362\u5e93":21,"\u6211\u4eec\u91c7\u7528\u4e0a\u9762\u7684\u7279\u5f81\u6a21\u677f":92,"\u6211\u4eec\u91c7\u7528\u5355\u5c42lstm\u6a21\u578b":89,"\u6211\u4eec\u91c7\u7528\u7b2c\u4e8c\u4e2a\u65b9\u6cd5":54,"\u6211\u4eec\u9700\u8981":65,"\u6211\u4eec\u9700\u8981\u5148\u521b\u5efa\u4e00\u4e2a\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":86,"\u6211\u4eec\u9700\u8981\u5148\u628a\u8f93\u5165tensor\u548c\u8f93\u51fatensor\u8f6c\u6362\u4e3aeigen\u652f\u6301\u7684\u683c\u5f0f":69,"\u6211\u4eec\u9700\u8981\u521b\u5efa\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":94,"\u6211\u4eec\u9700\u8981\u5236\u4f5c\u4e00\u4e2a\u5305\u542b\u8bad\u7ec3\u6570\u636e\u7684paddle\u955c\u50cf":79,"\u6211\u4eec\u9700\u8981\u5728\u96c6\u7fa4\u7684\u6240\u6709\u8282\u70b9\u4e0a\u5b89\u88c5":73,"\u6211\u4eec\u9700\u8981\u7b49\u5f0f\u5de6\u8fb9\u7684eigentensor\u8c03\u7528device\u63a5\u53e3":69,"\u6211\u4eec\u9700\u8981\u8ba1\u7b97":67,"\u6211\u4eec\u9700\u8981\u8bbe\u7f6e":91,"\u6211\u4eec\u9700\u8981\u9884\u5904\u7406\u6570\u6910\u5e76\u6784\u5efa\u4e00\u4e2a\u5b57\u5178":93,"\u6211\u4eec\u975e\u5e38\u6b22\u8fce\u60a8\u7528paddlepaddle\u6784\u5efa\u66f4\u597d\u7684\u793a\u4f8b":91,"\u6211\u4eec\u9884\u8bad\u7ec3\u5f97\u52304\u79cd\u4e0d\u540c\u7ef4\u5ea6\u7684\u8bcd\u5411\u91cf":85,"\u6211\u4eec\u9996\u5148\u5904\u7406\u7535\u5f71\u6216\u7528\u6237\u7684\u7279\u5f81\u6587\u4ef6":91,"\u6211\u4eec\u9996\u5148\u9700\u8981\u6839\u636e\u795e\u7ecf\u7f51\u7edc\u7ed3\u6784\u6765\u521b\u5efa\u6240\u9700\u8981\u4f18\u5316\u7684paramet":56,"\u6211\u4eec\u9ed8\u8ba4\u4f7f\u7528imdb\u7684\u6d4b\u8bd5\u6570\u636e\u96c6\u4f5c\u4e3a\u9a8c\u8bc1":93,"\u6211\u5220\u9664\u4e86":66,"\u6211\u53ef\u4ee5\u7528":65,"\u6211\u53ef\u4ee5\u9009\u62e9\u4e0d\u7528docker\u5417":65,"\u6216":[2,72,86,92],"\u6216\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":60,"\u6216\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u53d8\u6210\u4e00\u4e2a0\u5c42\u5e8f\u5217":60,"\u6216\u4e00\u4e2a\u5411\u91cf":63,"\u6216\u4e0d\u786e\u5b9a":90,"\u6216\u5355\u5c42\u5e8f\u5217\u7ecf\u8fc7\u8fd0\u7b97\u53d8\u6210\u4e00\u4e2a0\u5c42\u5e8f\u5217":60,"\u6216\u662f\u624b\u52a8\u7f16\u8f91\u751f\u6210":91,"\u6216\u6700\u5927\u503c":60,"\u6216\u6d4b\u8bd5\u6587\u4ef6\u5217\u8868":1,"\u6216\u7b2c\u4e00\u4e2a":60,"\u6216\u7b2c\u4e00\u4e2a\u5143\u7d20":60,"\u6216\u8005":[33,34,49,54,60,61,65,68,72],"\u6216\u8005\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":63,"\u6216\u8005\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":[60,63],"\u6216\u8005\u4ece\u5de5\u5177\u7684\u754c\u9762\u91cc\u8fd0\u884c\u60a8\u7684\u5e94\u7528":72,"\u6216\u8005\u53cd\u5411\u5730\u4ece":64,"\u6216\u8005\u5728cpu\u6a21\u5f0f\u4e0b\u4f7f\u75284\u4e2a\u7ebf\u7a0b":75,"\u6216\u8005\u5c06\u8fd9\u53f0\u8282\u70b9\u8fc1\u79fb\u5230\u53e6\u4e00\u4e2a\u8282\u70b9\u5e76\u542f\u52a8\u5373\u53ef\u6062\u590d\u8bad\u7ec3\u4efb\u52a1":20,"\u6216\u8005\u5df2\u7ecf\u5728\u96c6\u7fa4\u63d0\u4ea4\u73af\u5883\u4e2d\u81ea\u52a8\u8bbe\u7f6e":74,"\u6216\u8005\u6570\u636e\u5e93\u8fde\u63a5\u8def\u5f84\u7b49":1,"\u6216\u8005\u6570\u7ec4\u7684\u6570\u7ec4\u8fd9\u4e2a\u6982\u5ff5":61,"\u6216\u8005\u662f\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":60,"\u6216\u8005\u662f\u4e00\u4e2a\u5c0f\u7684\u6587\u672c\u7247\u6bb5":93,"\u6216\u8005\u662f\u51fd\u6570\u8c03\u7528\u7684\u9891\u7387\u548c\u8017\u65f6\u7b49":72,"\u6216\u8005\u66f4\u65e9":51,"\u6216\u8005\u6bcf\u4e00\u4e2a\u7cfb\u5217\u91cc\u7684\u7279\u5f81\u6570\u636e":61,"\u6216\u8005\u7528tuple\u8868\u793a\u7684\u591a\u4e2a\u503c":21,"\u6216\u8005\u7531\u5b83\u4eec\u7ec4\u6210\u7684list":21,"\u6216\u8005\u76f4\u63a5\u4f7f\u7528\u4e0b\u9762\u7684shell\u547d\u4ee4":87,"\u6216\u8005\u76f4\u63a5\u6254\u6389\u975e\u5e38\u957f\u7684\u5e8f\u5217":49,"\u6216\u8005\u76f8\u5bf9\u4e8e\u6784\u5efa\u76ee\u5f55\u7684\u76f8\u5bf9\u8def\u5f84":[58,59],"\u6216\u8005\u8f93\u5165\u6570\u636e\u5c3a\u5ea6\u8fc7\u5927":49,"\u6216\u8005\u8fd0\u884c":46,"\u6216\u8005\u91c7\u7528\u5e76\u884c\u8ba1\u7b97\u6765\u52a0\u901f\u67d0\u4e9b\u5c42\u7684\u66f4\u65b0":77,"\u6216\u8005\u9700\u8981\u53d1\u5e03\u60a8\u7684\u5e94\u7528\u7684\u955c\u50cf":54,"\u6216\u8005\u9700\u8981\u66f4\u9ad8\u7684\u6548\u7387":1,"\u6216\u8bbe\u7f6e\u4e3anone":1,"\u6216gpu":75,"\u6216gpu\u4e2a\u6570":93,"\u6218\u4e89\u7247":90,"\u622a\u65ad\u5bf9\u8c61\u4e0d\u540c":49,"\u623f":61,"\u623f\u95f4":61,"\u6240\u4ee5":[2,49],"\u6240\u4ee5\u4e00\u822c\u9700\u8981\u5bf9\u8bad\u7ec3\u7528\u7684\u6a21\u578b\u914d\u7f6e\u6587\u4ef6\u7a0d\u4f5c\u76f8\u5e94\u4fee\u6539\u624d\u80fd\u5728\u9884\u6d4b\u65f6\u4f7f\u7528":4,"\u6240\u4ee5\u4e0d\u80fd\u91c7\u7528\u7b2c\u4e00\u79cd\u65b9\u5f0f\u5728\u8fd9\u51e0\u4e2alayer\u91cc\u8bbe\u7f6e":50,"\u6240\u4ee5\u4f60\u53ea\u7528\u6309\u4e0b\u9762\u7684\u7ed3\u6784\u6765\u7ec4\u7ec7\u6570\u6910\u5c31\u884c\u4e86":93,"\u6240\u4ee5\u505a\u6cd5\u53ef\u4ee5\u6709\u4e24\u79cd":49,"\u6240\u4ee5\u53ef\u4ee5\u5229\u7528\u5982\u4e0b\u65b9\u6cd5\u8bfb\u53d6\u6a21\u578b\u7684\u53c2\u6570":52,"\u6240\u4ee5\u53ef\u4ee5\u7b80\u5316\u5bf9\u73af\u5883\u7684\u8981\u6c42":79,"\u6240\u4ee5\u5728\u5199\u5165\u5feb\u7167\u7684\u8fc7\u7a0b\u4e2d":20,"\u6240\u4ee5\u5916\u5c42\u8f93\u51fa\u7684\u5e8f\u5217\u5f62\u72b6":61,"\u6240\u4ee5\u5982\u679c\u9700\u8981\u81ea\u884c\u914d\u7f6e\u5f00\u53d1\u73af\u5883\u9700\u8981\u8003\u8651\u7248\u672c\u7684\u56e0\u7d20":54,"\u6240\u4ee5\u5b83\u4eec\u4f7f\u7528\u540c\u4e00\u4e2aip\u5730\u5740":78,"\u6240\u4ee5\u5bf9":61,"\u6240\u4ee5\u5f88\u591a\u65f6\u5019\u4f60\u9700\u8981\u505a\u7684\u53ea\u662f\u5b9a\u4e49\u6b63\u786e\u7684\u7f51\u7edc\u5c42\u5e76\u628a\u5b83\u4eec\u8fde\u63a5\u8d77\u6765":52,"\u6240\u4ee5\u6027\u80fd\u4e5f\u5c31\u9010\u6b65\u53d8\u6210\u4e86\u6df1\u5ea6\u5b66\u4e60\u9886\u57df\u6700\u91cd\u8981\u7684\u6307\u6807":72,"\u6240\u4ee5\u6211\u4eec\u4f7f\u7528\u8fd9\u4e2a\u955c\u50cf\u6765\u4e0b\u8f7d\u8bad\u7ec3\u6570\u636e\u5230docker":79,"\u6240\u4ee5\u6211\u4eec\u53ef\u4ee5\u5728\u8fd9\u4e2a\u57fa\u7840\u4e0a":80,"\u6240\u4ee5\u6211\u4eec\u9700\u8981\u5c06\u8f93\u5165\u6570\u636e\u6807\u8bb0\u6210":61,"\u6240\u4ee5\u63a8\u8350\u4f7f\u7528\u663e\u5f0f\u6307\u5b9a\u7684\u65b9\u5f0f\u6765\u8bbe\u7f6einput_typ":2,"\u6240\u4ee5\u653e\u4e00\u4e2a\u7a7a\u5217\u8868":52,"\u6240\u4ee5\u65e0\u9700\u5224\u65adavx\u5373\u53ef\u4f7f\u7528":54,"\u6240\u4ee5\u7528\u6237\u9700\u8981\u9996\u5148\u5728":28,"\u6240\u4ee5\u76f8\u6bd4\u4e8erecurr":50,"\u6240\u4ee5\u8bad\u7ec3":73,"\u6240\u4ee5\u8f93\u51fa\u7684value\u5305\u542b\u4e24\u4e2a\u5411\u91cf":4,"\u6240\u4ee5\u8fd9\u4e00\u6b65\u662f\u5fc5\u8981\u7684":67,"\u6240\u4f7f\u7528\u7684\u7f16\u8bd1\u5668\u4e3aarm":59,"\u6240\u5bf9\u5e94\u7684\u8bcd\u8868index\u6570\u7ec4":61,"\u6240\u6709\u4e0e\u7c7b\u578b\u76f8\u5173\u7684\u51fd\u6570":34,"\u6240\u6709\u4ee3\u7801\u5fc5\u987b\u5177\u6709\u5355\u5143\u6d4b\u8bd5":66,"\u6240\u6709\u53c2\u6570\u7f6e\u4e3a\u96f6":75,"\u6240\u6709\u540c\u76ee\u5f55\u4e0b\u7684\u6587\u672c\u5b9e\u4f8b\u6587\u4ef6\u90fd\u662f\u540c\u7ea7\u522b\u7684":93,"\u6240\u6709\u547d\u4ee4\u884c\u9009\u9879\u53ef\u4ee5\u8bbe\u7f6e\u4e3a":73,"\u6240\u6709\u6587\u4ef6\u5217\u8868":2,"\u6240\u6709\u672c\u5730\u8bad\u7ec3":73,"\u6240\u6709\u67b6\u6784":58,"\u6240\u6709\u6807\u8bb0\u7684\u6d4b\u8bd5\u96c6\u548c\u8bad\u7ec3\u96c6":93,"\u6240\u6709\u751f\u6210\u7684\u7684\u6587\u4ef6\u90fd\u5728build\u76ee\u5f55\u4e0b":54,"\u6240\u6709\u7684":[66,67],"\u6240\u6709\u7684\u4eba\u53e3\u7edf\u8ba1\u5b66\u4fe1\u606f\u7531\u7528\u6237\u81ea\u613f\u63d0\u4f9b":90,"\u6240\u6709\u7684\u5355\u6d4b\u90fd\u4f1a\u88ab\u6267\u884c\u4e00\u6b21":67,"\u6240\u6709\u7684\u63a5\u53e3\u5747\u4e3ac\u63a5\u53e3":34,"\u6240\u6709\u7684\u64cd\u4f5c\u90fd\u662f\u9488\u5bf9\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u6765\u8fdb\u884c\u7684":61,"\u6240\u6709\u7684\u7528\u6237\u4fe1\u606f\u90fd\u5305\u542b\u5728":90,"\u6240\u6709\u7684\u7535\u5f71\u4fe1\u606f\u90fd\u5305\u542b\u5728":90,"\u6240\u6709\u7684\u8bc4\u5206\u6570\u636e\u90fd\u5305\u542b\u5728":90,"\u6240\u6709\u7684python\u5c01\u88c5\u90fd\u4f7f\u7528":67,"\u6240\u6709\u7684python\u5c01\u88c5\u90fd\u5728":67,"\u6240\u6709\u7c7b\u578b\u540d\u4e3a":34,"\u6240\u6709\u7f51\u7edc\u5c42\u7684\u68af\u5ea6\u68c0\u67e5\u5355\u6d4b\u90fd\u4f4d\u4e8e":67,"\u6240\u6709\u8282\u70b9\u8fd0\u884c\u96c6\u7fa4\u4f5c\u4e1a\u7684\u4e3b\u673a\u540d\u6216":73,"\u6240\u6709\u8f93\u5165\u5e8f\u5217\u5e94\u8be5\u6709\u76f8\u540c\u7684\u957f\u5ea6":64,"\u6240\u6709\u914d\u7f6e\u90fd\u80fd\u5728":89,"\u6240\u6784\u5efa\u7f51\u7edc\u7ed3\u6784\u7684\u7684\u6df1\u5ea6\u6bd4\u4e4b\u524d\u4f7f\u7528\u7684\u7f51\u7edc\u6709\u5927\u5e45\u5ea6\u7684\u63d0\u9ad8":87,"\u6240\u793a":92,"\u6240\u8c13\u65f6\u95f4\u6b65\u4fe1\u606f":2,"\u6240\u9700\u652f\u6301\u7684\u6700\u4f4eandroid":58,"\u624b\u5199\u591a\u8bed\u8a00\u7ed1\u5b9a":33,"\u624b\u5de5\u827a\u8005":90,"\u624d\u4f1a\u91ca\u653e\u8be5\u6bb5\u5185\u5b58":2,"\u624d\u4f1astop":2,"\u624d\u80fd\u4fdd\u8bc1\u548c\u5355\u5c42\u5e8f\u5217\u7684\u914d\u7f6e\u4e2d":61,"\u624d\u80fd\u5145\u5206\u5229\u7528mac":54,"\u624d\u80fd\u53d1\u6325\u5176\u5168\u90e8\u80fd\u529b":72,"\u6253\u5370\u5728\u5c4f\u5e55\u4e0a":91,"\u6253\u5370\u7684\u65e5\u5fd7\u53d8\u591a":53,"\u6253\u5f00":72,"\u6253\u5f00\u6587\u672c\u6587\u4ef6":2,"\u6253\u5f00\u6d4f\u89c8\u5668\u8bbf\u95ee\u5bf9\u5e94\u76ee\u5f55\u4e0b\u7684index":70,"\u6253\u5f00\u8fd9\u4e2a\u7f16\u8bd1\u9009\u9879":34,"\u6267\u884c":[92,93],"\u6267\u884c\u4e0b\u9762\u7684\u547d\u4ee4\u5c31\u53ef\u4ee5\u9884\u5904\u7406\u6570\u6910":93,"\u6267\u884c\u4e0b\u9762\u7684\u547d\u4ee4\u6765\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5":68,"\u6267\u884c\u4ee5\u4e0b\u547d\u4ee4":[58,59],"\u6267\u884c\u4ee5\u4e0b\u64cd\u4f5c":64,"\u6267\u884c\u5b8c\u5b89\u88c5\u547d\u4ee4\u540e":[58,59],"\u6267\u884c\u60a8\u7684\u4ee3\u7801":72,"\u6267\u884c\u7684\u547d\u4ee4\u5982\u4e0b":87,"\u6269\u5c55\u673a\u5236\u7b49\u529f\u80fd":78,"\u6279\u6b21\u540e\u6253\u5370\u65e5\u5fd7":91,"\u6279\u6b21\u5bf9\u5e73\u5747\u53c2\u6570\u8fdb\u884c\u6d4b\u8bd5":92,"\u6279\u6b21\u7684\u6570\u636e":91,"\u627e\u5230":64,"\u627e\u5230\u6700\u65e9\u62a5\u9519\u7684\u5730\u65b9":47,"\u627e\u5230\u8fd0\u884c\u6162\u7684\u539f\u56e0":72,"\u627e\u5230\u8fd0\u884c\u6162\u7684\u90e8\u5206":72,"\u6280\u672f\u5458":90,"\u628a":[21,67],"\u628a\u4e4b\u524d\u793a\u4f8b\u4e2d\u8f6c\u6362\u5b8c\u6bd5\u7684random":21,"\u628a\u4efb\u610f\u7ef4\u5ea6\u7684tensor\u8f6c\u4e3a\u4e86\u4e00\u7ef4\u7684eigenvector":69,"\u628a\u5de5\u5177\u548c\u914d\u7f6e\u90fd\u5b89\u88c5\u5728\u4e00\u4e2a":65,"\u628a\u7528\u6237\u5728\u8d2d\u7269\u7f51\u7ad9":93,"\u628a\u7f16\u8bd1\u51fa\u6765\u7684paddlepaddle\u4e5f\u6253\u5305\u6210\u4e00\u4e2a\u955c":54,"\u628a\u8bad\u7ec3\u6570\u636e\u76f4\u63a5\u653e\u5728":79,"\u628a\u8fd9\u4e9b\u5de5\u5177\u5b89\u88c5\u5230\u672c\u673a":65,"\u6293\u53d6\u4ea7\u54c1\u7684\u7528\u6237\u8bc4\u8bba\u5e76\u5206\u6790\u4ed6\u4eec\u7684\u60c5\u611f":93,"\u6295\u5c04\u53cd\u5411rnn\u7684\u7b2c\u4e00\u4e2a\u5b9e\u4f8b\u5230":64,"\u6295\u5c04\u7f16\u7801\u5411\u91cf\u5230":64,"\u62bd\u53d6\u51fa\u7684\u65b0\u8bcd\u8868\u7684\u4fdd\u5b58\u8def\u5f84":85,"\u62bd\u53d6\u5bf9\u5e94\u7684\u8bcd\u5411\u91cf\u6784\u6210\u65b0\u7684\u8bcd\u8868":85,"\u62c6\u5206\u5230\u4e0d\u540c\u6587\u4ef6\u5939":94,"\u62c6\u89e3":63,"\u62c6\u89e3\u6210\u7684\u6bcf\u4e00\u53e5\u8bdd\u518d\u901a\u8fc7\u4e00\u4e2alstm\u7f51\u7edc":61,"\u62f7\u8d1d\u8bad\u7ec3\u6587\u4ef6\u5230\u5bb9\u5668\u5185":80,"\u62fc\u63a5":49,"\u62fc\u63a5\u6210\u4e00\u4e2a\u65b0\u7684\u5411\u91cf":89,"\u6301\u7eed\u96c6\u6210\u7cfb\u7edf\u4e5f\u53ef\u4ee5\u7b80\u5355\u5730\u590d\u73b0\u8fd9\u4e2a\u8fc7\u7a0b":54,"\u6302\u8f7dvolum":54,"\u6307\u5411\u4e00\u4e2alayer":63,"\u6307\u5b9a":[49,50,63,64],"\u6307\u5b9a\u4e00\u53f0\u673a\u5668\u4e0a\u4f7f\u7528\u7684\u7ebf\u7a0b\u6570":75,"\u6307\u5b9a\u4f7f\u75282":49,"\u6307\u5b9a\u521d\u59cb\u5316\u6a21\u578b\u8def\u5f84":89,"\u6307\u5b9a\u524d\u5411\u7f51\u7edc\u6700\u7ec8\u7684\u8f93\u51fa\u76ee\u6807\u53d8\u91cf":68,"\u6307\u5b9a\u52a0\u8f7d\u7684\u65b9\u5f0f":75,"\u6307\u5b9a\u5bf9\u8f93\u5165\u53d8\u91cf":68,"\u6307\u5b9a\u5de5\u4f5c\u6a21\u578b\u8fdb\u884c\u9884\u6d4b":87,"\u6307\u5b9a\u5de5\u4f5c\u6a21\u5f0f\u6765\u63d0\u53d6\u7279\u5f81":87,"\u6307\u5b9a\u63d0\u53d6\u7279\u5f81\u7684\u5c42":87,"\u6307\u5b9a\u662f\u5426\u4f7f\u7528gpu":87,"\u6307\u5b9a\u751f\u6210\u6570\u636e\u7684\u51fd\u6570":89,"\u6307\u5b9a\u7684\u5b57\u5178\u5355\u8bcd\u6570":94,"\u6307\u5b9a\u7684\u6570\u636e\u5c06\u4f1a\u88ab\u6d4b\u8bd5":89,"\u6307\u5b9a\u7684\u8f93\u5165\u4e0d\u4f1a\u88ab":63,"\u6307\u5b9a\u8981\u8f93\u51fa\u7684\u5b57\u6bb5\u8fdb\u884c\u8f93\u51fa":49,"\u6307\u5b9a\u8bad\u7ec3\u6570\u636e\u548c\u6d4b\u8bd5\u6570\u636e":89,"\u6307\u5b9abatch":94,"\u6307\u5b9acudnn\u7684\u6700\u5927\u5de5\u4f5c\u7a7a\u95f4\u5bb9\u9650":75,"\u6307\u5bf9\u4e8e\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217\u8f93\u5165\u6570\u636e":61,"\u6307\u6df1\u5ea6\u5b66\u4e60\u8bad\u7ec3\u4e4b\u540e\u5f97\u5230\u7684\u6240\u6709\u53c2\u6570":20,"\u6307\u793a\u4f7f\u7528\u54ea\u4e2agpu\u6838":75,"\u6307\u793a\u5728\u7b80\u5355\u7684recurrentlayer\u5c42\u7684\u8ba1\u7b97\u4e2d\u662f\u5426\u4f7f\u7528\u6279\u5904\u7406\u65b9\u6cd5":75,"\u6307\u793a\u5f53\u6307\u5b9a\u8f6e\u7684\u6d4b\u8bd5\u6a21\u578b\u4e0d\u5b58\u5728\u65f6":75,"\u6307\u793a\u662f\u5426\u4f7f\u7528\u591a\u7ebf\u7a0b\u6765\u8ba1\u7b97\u4e00\u4e2a\u795e\u7ecf\u7f51\u7edc":75,"\u6307\u793a\u662f\u5426\u5f00\u542f\u53c2\u6570\u670d\u52a1\u5668":75,"\u6307\u793a\u662f\u5426\u663e\u793a\u53c2\u6570\u670d\u52a1\u5668\u4e0a\u7684\u7a00\u758f\u53c2\u6570\u5206\u5e03\u7684\u65e5\u5fd7\u7ec6\u8282":75,"\u6307\u793a\u662f\u5426\u68c0\u67e5\u6240\u6709\u53c2\u6570\u670d\u52a1\u5668\u4e0a\u7684\u7a00\u758f\u53c2\u6570\u7684\u5206\u5e03\u662f\u5747\u5300\u7684":75,"\u6307\u793a\u6d4b\u8bd5\u4efb\u52a1":92,"\u6307\u793a\u6d4b\u8bd5\u4efb\u52a1\u7684\u6807\u8bb0":92,"\u6309\u542f\u53d1\u5f0f\u635f\u5931\u7684\u5927\u5c0f\u9012\u589e\u6392\u5e8f":75,"\u6309\u7167":59,"\u6309\u7167\u4e0b\u9762\u6b65\u9aa4\u5373\u53ef":80,"\u6309\u7167\u5176\u5185\u5bb9\u521b\u5efa\u4e00\u4e2a\u540d\u4e3a":65,"\u6309\u7167\u5177\u4f53\u5b9e\u73b0\u65b9\u5f0f\u53ef\u4ee5\u5f52\u7eb3\u4e3a2\u7c7b":50,"\u6309\u94ae":66,"\u633a":61,"\u633a\u597d":61,"\u6355\u83b7":89,"\u6362":61,"\u6392\u6210\u4e00\u5217\u7684\u591a\u4e2a\u5143\u7d20":60,"\u63a5\u4e0a\u4e00\u4e2a\u5168\u8fde\u63a5\u5c42":56,"\u63a5\u4e0a\u5e73\u65b9\u8bef\u5dee\u5c42":56,"\u63a5\u4e0b\u6765":[68,89,93],"\u63a5\u4e0b\u6765\u53ef\u4ee5\u8003\u8651\u4e0b\u65f6\u95f4\u7ebf\u7684\u5206\u6790":72,"\u63a5\u4e0b\u6765\u5c31\u53ef\u4ee5\u4f7f\u7528":72,"\u63a5\u4e0b\u6765\u6211\u4eec\u521b\u5efa\u4e00\u4e2a\u539f\u59cb":66,"\u63a5\u4e0b\u6765\u6211\u4eec\u53d6\u6d88\u5bf9":66,"\u63a5\u4e0b\u6765\u6211\u4eec\u5c06\u5c55\u793a\u5982\u4f55\u7528paddlepaddle\u8bad\u7ec3\u4e00\u4e2a\u6587\u672c\u5206\u7c7b\u6a21\u578b":89,"\u63a5\u4e0b\u6765\u7b49\u5f85":66,"\u63a5\u53d7":92,"\u63a5\u53d7\u4e00\u4e2a\u8f93\u5165\u53c2\u6570":68,"\u63a5\u53d7\u7684\u4e1c\u897f":92,"\u63a5\u53d7\u8005":92,"\u63a5\u53e3":[33,34,68,69],"\u63a5\u53e3\u4f1a\u88ab\u8c03\u7528":69,"\u63a5\u53e3\u540d\u79f0":73,"\u63a5\u53e3\u5c42\u505a\u8fc7\u591a\u5c01\u88c5":34,"\u63a5\u53e3\u63d0\u53d6\u7684\u7ed3\u679c\u662f\u4e00\u81f4\u7684":87,"\u63a5\u53e3\u662f":21,"\u63a5\u53e3\u6700\u7ec8\u4f1a\u8c03\u7528\u5bf9\u5e94":69,"\u63a5\u53e3\u6709\u4e00\u4e2a":49,"\u63a5\u53e3\u6765\u52a0\u8f7d\u6570\u636e":89,"\u63a5\u53e3\u6765\u52a0\u8f7d\u8be5\u6587\u4ef6":87,"\u63a5\u53e3\u6765\u6253\u5f00\u8be5\u6587\u4ef6":87,"\u63a5\u53e3\u7684":49,"\u63a5\u6536\u5904\u7406pfsclient\u7aef\u7684\u6587\u4ef6\u7ba1\u7406\u8bf7\u6c42":28,"\u63a5\u7740\u6211\u4eec\u5c31\u80fd\u591f\u6253\u5f00\u6d4f\u89c8\u5668\u5728":54,"\u63a7\u5236":75,"\u63a7\u5236\u5982\u4f55\u6539\u53d8\u6a21\u578b\u53c2\u6570":52,"\u63a7\u5236\u662f\u5426\u751f\u6210\u5355\u5143\u6d4b\u8bd5":54,"\u63a7\u5236\u751f\u6210\u7684\u751f\u4ea7\u955c\u50cf\u662f\u5426\u652f\u6301avx":54,"\u63a7\u5236\u751f\u6210\u7684\u751f\u4ea7\u955c\u50cf\u662f\u5426\u652f\u6301gpu":54,"\u63a7\u5236\u7528\u6237\u6743\u9650":21,"\u63a8\u5bfc\u8be5\u5c42\u524d\u5411\u548c\u540e\u5411\u4f20\u9012\u7684\u65b9\u7a0b":67,"\u63a8\u8350":61,"\u63a8\u8350\u4f7f\u7528":2,"\u63a8\u8350\u56fd\u5185\u7528\u6237\u4f7f\u7528docker":54,"\u63a8\u8350\u6e05\u7406\u6574\u4e2a\u7f16\u8bd1\u76ee\u5f55":53,"\u63a8\u8350\u76f4\u63a5\u5b58\u653e\u5230\u8bad\u7ec3\u76ee\u5f55":1,"\u63a8\u8350\u7cfb\u7edf":73,"\u63a8\u9001\u5230\u8fdc\u7a0b\u4ed3\u5e93":66,"\u63a8\u9500\u5458":90,"\u63cf\u8ff0":53,"\u63cf\u8ff0\u7684\u9ed8\u8ba4\u5165\u53e3\u7a0b\u5e8f":65,"\u63cf\u8ff0\u7f51\u7edc\u7ed3\u6784\u548c\u4f18\u5316\u7b97\u6cd5":89,"\u63cf\u8ff0\u8be5op\u7684\u8f93\u5165":68,"\u63cf\u8ff0\u95ee\u9898":66,"\u63cf\u8ff0kubernetes\u4e0a\u8fd0\u884c\u7684\u4f5c\u4e1a":78,"\u63d0\u4ea4\u65b9\u5f0f\u53c2\u89c1":70,"\u63d0\u4ea4pull":66,"\u63d0\u4f9b":73,"\u63d0\u4f9b\u4e03\u5c42\u534f\u8bae\u7684\u53cd\u5411\u4ee3\u7406":28,"\u63d0\u4f9b\u4e86\u4e00\u4e2a\u542f\u52a8\u811a\u672c":80,"\u63d0\u4f9b\u4e86\u547d\u4ee4\u6837\u4f8b\u6765\u8fd0\u884c":73,"\u63d0\u4f9b\u4e86\u81ea\u52a8\u5316\u811a\u672c\u6765\u542f\u52a8\u4e0d\u540c\u8282\u70b9\u4e2d\u7684\u6240\u6709":73,"\u63d0\u4f9b\u51e0\u4e4e\u6240\u6709\u8bad\u7ec3\u7684\u5185\u90e8\u8f93\u51fa\u65e5\u5fd7":73,"\u63d0\u4f9b\u5e38\u7528\u7684\u547d\u4ee4\u884c\u7ba1\u7406\u547d\u4ee4\u7ba1\u7406\u6587\u4ef6\u548c\u76ee\u5f55":28,"\u63d0\u4f9b\u6269\u5c55\u7684\u957f\u5ea6\u4fe1\u606f":60,"\u63d0\u4f9b\u6700\u65b0":54,"\u63d0\u4f9b\u7528\u6237\u7ba1\u7406\u6587\u4ef6\u7684\u547d\u4ee4":28,"\u63d0\u4f9b\u7ed9paddle\u4f5c\u4e3a\u8bad\u7ec3\u6570\u636e":21,"\u63d0\u4f9b\u8bad\u7ec3\u8fc7\u7a0b\u7684":73,"\u63d0\u51fa\u7684\u4ee3\u7801\u9700\u6c42":85,"\u63d0\u793a":46,"\u642d\u5efa\u795e\u7ecf\u7f51\u7edc\u5c31\u50cf\u4f7f\u7528\u79ef\u6728\u642d\u5efa\u5b9d\u5854\u4e00\u6837":56,"\u64cd\u4f5c":61,"\u64cd\u4f5c\u7cfb\u7edf":65,"\u64cd\u6301\u5bb6\u52a1\u8005":90,"\u652f\u6301\u53cc\u5c42\u5e8f\u5217\u4f5c\u4e3a\u8f93\u5165\u7684layer":[62,63],"\u652f\u6301\u5927\u6587\u4ef6\u7684\u65ad\u70b9\u4e0a\u4f20":28,"\u652f\u6301\u7684\u6700\u5c0f\u7684android":58,"\u652f\u6301rbd":78,"\u653e\u5728\u8fd9\u4e2a\u76ee\u5f55\u91cc\u7684\u6587\u4ef6\u5176\u5b9e\u662f\u4fdd\u5b58\u5230\u4e86mfs\u4e0a":80,"\u653e\u5fc3":61,"\u6545\u800c\u662f\u4e00\u4e2a\u5355\u5c42\u65f6\u95f4\u5e8f\u5217":61,"\u6548\u679c\u603b\u7ed3":89,"\u6559\u7a0b\u6587\u6863\u5230":73,"\u6559\u80b2\u5de5\u4f5c\u8005":90,"\u6570":[63,92],"\u6570\u5fc5\u987b\u4e25\u683c\u76f8\u7b49":63,"\u6570\u636e":[28,94],"\u6570\u636e\u4e0b\u8f7d\u4e4b\u540e":86,"\u6570\u636e\u4e2d0":51,"\u6570\u636e\u5217\u8868":87,"\u6570\u636e\u5c06\u4fdd\u5b58\u5728":85,"\u6570\u636e\u5c42":[52,91],"\u6570\u636e\u5e94\u8be5\u5728\u542f\u52a8\u96c6\u7fa4\u4f5c\u4e1a\u4e4b\u524d\u51c6\u5907\u597d":73,"\u6570\u636e\u63d0\u4f9b\u5668":74,"\u6570\u636e\u63d0\u4f9b\u811a\u672c\u4ec5\u4ec5\u662f\u8bfb\u53d6meta":91,"\u6570\u636e\u63d0\u4f9b\u811a\u672c\u7684\u7ec6\u8282\u6587\u6863\u53ef\u4ee5\u53c2\u8003":91,"\u6570\u636e\u76ee\u5f55\u4e2d\u7684\u6240\u6709\u6587\u4ef6\u88ab":73,"\u6570\u636e\u7c7b\u578b":4,"\u6570\u636e\u7f13\u5b58\u7684\u7b56\u7565":2,"\u6570\u636e\u8bbf\u95ee":0,"\u6570\u636e\u8bfb\u53d6\u5747\u4ea4\u7531\u5176\u4ed6\u8bed\u8a00\u5b8c\u6210":33,"\u6570\u636e\u8f93\u5165":63,"\u6570\u636e\u8f93\u5165\u683c\u5f0f":2,"\u6570\u636e\u957f\u5ea6\u53ca\u6821\u9a8c\u503c\u7ec4\u6210":28,"\u6570\u636e\u96c6":90,"\u6570\u636e\u96c6\u63cf\u8ff0":91,"\u6570\u636e\u96c6\u6587\u4ef6\u5939\u540d\u79f0":94,"\u6570\u636e\u96c6\u9700\u8981\u9884\u5148\u88ab\u8f6c\u6362\u6210paddlepaddle\u5206\u5e03\u5f0f\u8bad\u7ec3\u4f7f\u7528\u7684\u5b58\u50a8\u683c":21,"\u6570\u636e\u9884\u5904\u7406\u4efb\u52a1":21,"\u6570\u636e\u9884\u5904\u7406\u5b8c\u6210\u4e4b\u540e":89,"\u6570\u636e\u9884\u6d4b":92,"\u6570\u6910\u5b9a\u4e49":93,"\u6570\u6910\u8bf4\u660e\u6587\u6863":93,"\u6570\u6910\u96c6\u548c":93,"\u6570\u76ee":77,"\u6574\u4f53":61,"\u6574\u4f53\u6570\u636e\u548c\u539f\u59cb\u6570\u636e\u5b8c\u5168\u4e00\u6837":61,"\u6574\u4f53\u7684\u7ed3\u6784\u56fe\u5982\u4e0b":80,"\u6574\u6570":67,"\u6574\u6570\u6807\u7b7e":[2,56],"\u6574\u6d01":61,"\u6587\u4e66\u5de5\u4f5c":90,"\u6587\u4ef6":[33,65,66,68,79,92],"\u6587\u4ef6\u4e2d":[68,80,87,90,92],"\u6587\u4ef6\u4e2d\u6307\u5b9a\u6a21\u578b\u8def\u5f84\u548c\u8f93\u51fa\u7684\u76ee\u5f55":87,"\u6587\u4ef6\u4e2d\u6307\u5b9a\u8981\u63d0\u53d6\u7279\u5f81\u7684\u7f51\u7edc\u5c42\u7684\u540d\u5b57":87,"\u6587\u4ef6\u4e2d\u6ce8\u518c\u524d\u5411":68,"\u6587\u4ef6\u4e2d\u6ce8\u518c\u8be5op\u548ckernel":68,"\u6587\u4ef6\u4e2d\u6ce8\u518cgpu":68,"\u6587\u4ef6\u4e2d\u7684":87,"\u6587\u4ef6\u4e2d\u7684\u6bcf\u884c\u90fd\u5fc5\u987b\u662f\u4e00\u4e2a\u53e5\u5b50":94,"\u6587\u4ef6\u4e3a":[49,94],"\u6587\u4ef6\u4e4b\u5916":66,"\u6587\u4ef6\u4e5f\u53ef\u4ee5\u7528\u4e8e\u5bf9\u6837\u672c\u8fdb\u884c\u9884\u6d4b":87,"\u6587\u4ef6\u4f20\u8f93\u7684\u7684\u5173\u952e\u5728\u4e8e\u9700\u8981pfsclient\u7aef\u5bf9\u6bd4source\u548cdestination\u7684\u6587\u4ef6chunks\u7684checksum\u662f\u5426\u4fdd\u6301\u4e00\u81f4":28,"\u6587\u4ef6\u5185\u5bb9\u4e3a":33,"\u6587\u4ef6\u5206\u5272\u4e3a\u8bad\u7ec3\u548c\u6d4b\u8bd5\u6587\u4ef6":91,"\u6587\u4ef6\u540d\u4e3a\u6b64uuid":20,"\u6587\u4ef6\u540d\u79f0\u4e3a":91,"\u6587\u4ef6\u547d\u540d\u4ee5":68,"\u6587\u4ef6\u59390":80,"\u6587\u4ef6\u5939\u4e0b":70,"\u6587\u4ef6\u5939\u4e2d\u7684\u6bcf\u4e2a\u6587\u4ef6\u7684\u6bcf\u4e00\u884c\u5305\u542b\u4e24\u90e8\u5206":94,"\u6587\u4ef6\u5bf9\u5e94\u7684data":21,"\u6587\u4ef6\u7684\u4e0a\u4f20\u548c\u4e0b\u8f7d\u90fd\u662f\u901a\u8fc7\u5bf9chunk\u7684\u64cd\u4f5c\u6765\u5b9e\u73b0\u7684":28,"\u6587\u4ef6\u7684\u5206\u9694\u7b26\u4e3a":91,"\u6587\u4ef6\u7684\u6539\u53d8":66,"\u6587\u4ef6\u7684\u683c\u5f0f\u53ef\u4ee5":91,"\u6587\u4ef6\u7a0d\u6709\u5dee\u522b":86,"\u6587\u4ef6\u7cfb\u7edf\u4ee5\u53ca\u7f51\u7edc\u4e4b\u4e0a":54,"\u6587\u4ef6\u7d22\u5f15":73,"\u6587\u4ef6\u7ed9\u51fa\u4e86\u5b8c\u6574\u4f8b\u5b50":89,"\u6587\u4ef6model":77,"\u6587\u5b57\u7684\u4ea4\u4e92\u5f0f\u6587\u6863":54,"\u6587\u672c\u4e2d\u7684\u5355\u8bcd\u7528\u7a7a\u683c\u5206\u9694":89,"\u6587\u672c\u4fe1\u606f\u5c31\u662f\u4e00\u4e2a\u5e8f\u5217\u6570\u636e":2,"\u6587\u672c\u5206\u7c7b\u95ee\u9898":89,"\u6587\u672c\u5377\u79ef\u5206\u53ef\u4e3a\u4e09\u4e2a\u6b65\u9aa4":89,"\u6587\u672c\u5377\u79ef\u91c7\u6837\u5c42":91,"\u6587\u672c\u6295\u5f71\u5c42":91,"\u6587\u672c\u683c\u5f0f\u7684\u5b9e\u4f8b\u6587\u4ef6":93,"\u6587\u6863":46,"\u6587\u68631":69,"\u6587\u68632":69,"\u6587\u6863\u7f16\u5199\u7b49\u5de5\u4f5c":54,"\u6587\u6863\u81ea\u52a8\u5206\u7c7b\u548c\u95ee\u7b54":92,"\u6587\u6863\u8f83\u5c11":69,"\u6587\u6863\u90fd\u662f\u901a\u8fc7":70,"\u6587\u7ae0":80,"\u65b0":61,"\u65b0\u5efa\u4e00\u4e2a\u6743\u91cd":67,"\u65b0\u624b\u5165\u95e8":83,"\u65b0\u624b\u5165\u95e8\u7ae0\u8282":42,"\u65b9\u4fbf":61,"\u65b9\u4fbf\u4eca\u540e\u7684\u5d4c\u5165\u5f0f\u79fb\u690d\u5de5\u4f5c":53,"\u65b9\u4fbf\u5f00\u53d1\u8005\u76f4\u63a5\u767b\u5f55\u5230\u955c\u50cf\u4e2d\u8fdb\u884c\u5f00\u53d1":54,"\u65b9\u4fbf\u6d4b\u8bd5\u4eba\u5458\u6d4b\u8bd5paddlepaddle\u7684\u884c\u4e3a":42,"\u65b9\u4fbf\u7528\u6237\u4e0a\u4f20\u81ea\u5df1\u7684\u8bad\u7ec3\u6570\u636e\u4ee5\u8fdb\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3":28,"\u65b9\u5f0f1":49,"\u65b9\u5f0f2":49,"\u65b9\u6848\u6765\u6807\u8bb0\u6bcf\u4e2a\u53c2\u6570":92,"\u65b9\u6cd5\u4e00":77,"\u65b9\u6cd5\u4e09":77,"\u65b9\u6cd5\u4e8c":77,"\u65c1\u8fb9":61,"\u65c5\u6e38\u7f51\u7ad9":93,"\u65e0":61,"\u65e0\u4e1a\u4eba\u58eb":90,"\u65e0\u5ef6\u8fdf":75,"\u65e0\u6cd5\u505a\u5230\u5bf9\u4e8e\u5404\u79cd\u8bed\u8a00\u9519\u8bef\u5904\u7406\u65b9\u5f0f\u7684\u9002\u914d":33,"\u65e0\u8bba\u5728\u672c\u5730\u8fd8\u662f\u5728\u4e91\u7aef":21,"\u65e0\u8bba\u662f\u4ece":21,"\u65e0\u8bba\u662f\u5728\u672c\u5730\u6216\u662f\u4e91\u7aef\u8f6c\u6362":21,"\u65e0\u9ed8\u8ba4\u503c":[58,59],"\u65e5\u5fd7\u5c06\u4fdd\u5b58\u5728":93,"\u65e5\u5fd7\u62a5\u9519\u4e3a\u7f51\u7edc\u901a\u4fe1\u7c7b\u9519\u8bef":47,"\u65e8\u5728\u5efa\u7acb\u4e00\u4e2a\u53ef\u4ee5\u88ab\u534f\u540c\u8c03\u81f3\u6700\u4f18\u7ffb\u8bd1\u6548\u679c\u7684\u5355\u795e\u7ecf\u5143\u7f51\u7edc":94,"\u65e9\u9910":61,"\u65f6":[20,49,51,60,64,67,75,80],"\u65f6\u5019":61,"\u65f6\u52a0\u4e0a":93,"\u65f6\u5e8f\u6a21\u578b\u5747\u4f7f\u7528\u8be5\u811a\u672c":89,"\u65f6\u5e8f\u6a21\u578b\u662f\u6307\u6570\u636e\u7684\u67d0\u4e00\u7ef4\u5ea6\u662f\u4e00\u4e2a\u5e8f\u5217\u5f62\u5f0f":2,"\u65f6\u76ee\u6807\u8bed\u8a00\u7684\u6587\u4ef6":94,"\u65f6\u88ab\u8bad\u7ec3\u7684":67,"\u65f6\u8bbe\u5907id\u53f7\u7684\u5206\u914d":77,"\u65f6\u95f4":61,"\u65f6\u95f4\u6233":90,"\u65f6\u95f4\u6233\u8868\u793a\u4e3a\u4ece1970":90,"\u65f6\u95f4\u6b65\u7684\u6982\u5ff5":61,"\u65f6\u987b\u4ece\u7b2c17\u5b57\u8282\u5f00\u59cb":51,"\u6620\u5c04\u4e3a":65,"\u6620\u5c04\u5230\u4e00\u4e2a\u7ef4\u5ea6\u4e3a":67,"\u662f":[28,46,53,61],"\u662f\u4e00\u4e2a\u51681\u7684\u5411\u91cf":67,"\u662f\u4e00\u4e2a\u5185\u7f6e\u7684\u5b9a\u65f6\u5668\u5c01\u88c5":72,"\u662f\u4e00\u4e2a\u52a8\u6001\u7a0b\u5e8f\u5206\u6790\u7684\u672f\u8bed":72,"\u662f\u4e00\u4e2a\u5305\u88c5\u6570\u636e\u7684":92,"\u662f\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":60,"\u662f\u4e00\u4e2a\u53cc\u5c42\u7684\u5e8f\u5217":60,"\u662f\u4e00\u4e2a\u591a\u8bed\u8a00\u63a5\u53e3\u7684\u4ee3\u7801\u751f\u6210\u5668":33,"\u662f\u4e00\u4e2a\u5c01\u88c5\u5bf9\u8c61":72,"\u662f\u4e00\u4e2a\u5f88\u6709\u7528\u7684\u53c2\u6570":77,"\u662f\u4e00\u4e2a\u7b26\u5408\u9ad8\u65af\u5206\u5e03\u7684\u968f\u673a\u53d8\u91cf":52,"\u662f\u4e00\u4e2a\u7c7b\u578b\u7684\u6807\u5fd7":34,"\u662f\u4e00\u4e2a\u7edf\u8ba1\u5b66\u7684\u673a\u5668\u7ffb\u8bd1\u7cfb\u7edf":94,"\u662f\u4e00\u4e2a\u914d\u7f6e\u6587\u4ef6\u7684\u4f8b\u5b50":93,"\u662f\u4e00\u4e2a\u975e\u7ebf\u6027\u7684":67,"\u662f\u4e00\u4e2apython\u7684":2,"\u662f\u4e00\u4e2aunbound":63,"\u662f\u4e00\u6761\u65f6\u95f4\u5e8f\u5217":[2,56],"\u662f\u4e00\u79cd\u4efb\u610f\u590d\u6742\u7684rnn\u5355\u5143":63,"\u662f\u4e00\u7ec4":78,"\u662f\u4e0d\u5305\u62ec\u6e90\u7801\u7684":79,"\u662f\u4e0d\u5e38\u89c1\u7684\u505a\u6cd5":33,"\u662f\u4e0d\u662f\u5f88\u7b80\u5355\u5462":2,"\u662f\u4e0d\u662f\u8981\u5bf9\u6570\u636e\u505ashuffl":2,"\u662f\u4e3b\u8981\u7684\u53ef\u6267\u884cpython\u811a\u672c":92,"\u662f\u4ec0\u4e48\u4e5f\u6ca1\u5173\u7cfb":2,"\u662f\u4f17\u591a\u8bef\u5dee\u4ee3\u4ef7\u51fd\u6570\u5c42\u7684\u4e00\u79cd":52,"\u662f\u4f7f\u5f97\u8981\u5171\u4eab\u7684\u53c2\u6570\u4f7f\u7528\u540c\u6837\u7684":51,"\u662f\u504f\u5dee":64,"\u662f\u5176\u5927\u5c0f":52,"\u662f\u51e0\u4e4e\u4e0d\u5360\u5185\u5b58\u7684":2,"\u662f\u539f\u59cb\u6cd5\u8bed\u6587\u4ef6":94,"\u662f\u5404\u4e2a\u5b9e\u73b0\u4e2d\u5171\u4eab\u7684\u5934\u6587\u4ef6":34,"\u662f\u5411\u91cf":67,"\u662f\u5426\u4ee5\u9006\u5e8f\u5904\u7406\u8f93\u5165\u5e8f\u5217":63,"\u662f\u5426\u4f7f\u7528\u53cc\u7cbe\u5ea6\u6d6e\u70b9\u6570":53,"\u662f\u5426\u4f7f\u7528\u65e7\u7684remoteparameterupdat":75,"\u662f\u5426\u4f7f\u7528\u6743\u91cd":67,"\u662f\u5426\u4f7f\u7528arm\u6a21\u5f0f":58,"\u662f\u5426\u4f7f\u7528gpu":91,"\u662f\u5426\u4f7f\u7528gpu\u8bad\u7ec3":94,"\u662f\u5426\u4f7f\u7528neon\u6307\u4ee4":[58,59],"\u662f\u5426\u5141\u8bb8\u6682\u5b58\u7565\u5fae\u591a\u4f59pool_size\u7684\u6570\u636e":2,"\u662f\u5426\u5185\u5d4cpython\u89e3\u91ca\u5668":53,"\u662f\u5426\u5219\u5171\u4eab\u540c\u4e00\u4e2a":68,"\u662f\u5426\u5c06\u5168\u5c40\u79cd\u5b50\u5e94\u7528\u4e8e\u672c\u5730\u7ebf\u7a0b\u7684\u968f\u673a\u6570":75,"\u662f\u5426\u5f00\u542f\u5355\u5143\u6d4b\u8bd5":53,"\u662f\u5426\u5f00\u542f\u8ba1\u65f6\u529f\u80fd":53,"\u662f\u5426\u5f00\u542frdma":53,"\u662f\u5426\u6253\u5370\u7248\u672c\u4fe1\u606f":75,"\u662f\u5426\u652f\u6301gpu":53,"\u662f\u5426\u663e\u793a":75,"\u662f\u5426\u7a00\u758f":67,"\u662f\u5426\u7f16\u8bd1\u4e2d\u82f1\u6587\u6587\u6863":53,"\u662f\u5426\u7f16\u8bd1\u542b\u6709avx\u6307\u4ee4\u96c6\u7684paddlepaddle\u4e8c\u8fdb\u5236\u6587\u4ef6":53,"\u662f\u5426\u7f16\u8bd1\u65f6\u8fdb\u884c\u4ee3\u7801\u98ce\u683c\u68c0\u67e5":53,"\u662f\u5426\u7f16\u8bd1python\u7684swig\u63a5\u53e3":53,"\u662f\u5426\u8fd0\u884c\u65f6\u52a8\u6001\u52a0\u8f7dcuda\u52a8\u6001\u5e93":53,"\u662f\u5426\u9700\u8981\u7b49\u5f85\u8be5\u8f6e\u6a21\u578b\u53c2\u6570":75,"\u662f\u56e0\u4e3a\u8fd9\u4e2a\u6d41\u7a0b\u6bd4\u5176\u4ed6\u65b9\u6cd5\u90fd\u66f4\u7b80\u4fbf":65,"\u662f\u56e0\u4e3ac99\u652f\u6301":33,"\u662f\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u52a0\u8f7d\u5b57\u5178\u5e76\u5b9a\u4e49\u6570\u636e\u63d0\u4f9b\u7a0b\u5e8f\u6a21\u5757\u548c\u7f51\u7edc\u67b6\u6784\u7684\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6":92,"\u662f\u5728paddlepaddle\u4e2d\u6784\u9020\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u65f6\u6700\u91cd\u8981\u7684\u6982\u5ff5":64,"\u662f\u57fa\u7840\u7684\u8ba1\u7b97\u5355\u5143":52,"\u662f\u5b58\u6709\u4e00\u7cfb\u5217\u53d8\u6362\u77e9\u9635\u7684\u6743\u91cd":67,"\u662f\u5b58\u6709\u504f\u7f6e\u5411\u91cf\u7684\u6743\u91cd":67,"\u662f\u5bf9\u7528\u6237\u6587\u4ef6\u5b58\u50a8\u7a7a\u95f4\u7684\u62bd\u8c61":28,"\u662f\u5e8f\u5217":91,"\u662f\u5f85\u6269\u5c55\u7684\u6570\u636e":60,"\u662f\u6211\u4eec":66,"\u662f\u6307":34,"\u662f\u6307\u4e00\u7cfb\u5217\u7684\u7279\u5f81\u6570\u636e":61,"\u662f\u6307recurrent_group\u7684\u591a\u4e2a\u8f93\u5165\u5e8f\u5217":61,"\u662f\u6570\u636e\u8f93\u5165":64,"\u662f\u6574\u4e2a\u7684\u8bcd\u5d4c\u5165":91,"\u662f\u6709\u610f\u4e49\u7684":61,"\u662f\u6784\u5efa\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u6700\u91cd\u8981\u7684\u5de5\u5177":64,"\u662f\u6a21\u578b\u53c2\u6570\u4f18\u5316\u7684\u76ee\u6807\u51fd\u6570":52,"\u662f\u6ca1\u6709\u540d\u5b57\u7684":65,"\u662f\u6d45\u5c42\u8bed\u4e49\u89e3\u6790\u7684\u4e00\u79cd\u5f62\u5f0f":92,"\u662f\u6e90\u8bed\u8a00\u7684\u6587\u4ef6":94,"\u662f\u7528\u6237\u4f7f\u7528c":34,"\u662f\u7684":65,"\u662f\u76ee\u6807\u82f1\u8bed\u6587\u4ef6":94,"\u662f\u77e9\u9635":67,"\u662f\u795e\u7ecf\u7f51\u7edc\u7684\u5165\u53e3":52,"\u662f\u7f51\u7edc\u548c\u6570\u636e\u914d\u7f6e\u6587\u4ef6":86,"\u662f\u7f51\u7edc\u5c42\u5b9e\u4f8b\u7684\u540d\u5b57\u6807\u8bc6\u7b26":67,"\u662f\u7f51\u7edc\u5c42\u7684\u6807\u8bc6\u7b26":67,"\u662f\u7f51\u7edc\u5c42\u7684\u7c7b\u578b":67,"\u662f\u7f51\u7edc\u5c42\u8f93\u51fa\u7684\u5927\u5c0f":67,"\u662f\u8be5\u5c42\u7684\u6807\u8bc6\u7b26":67,"\u662f\u8be5\u5c42\u7684\u7c7b\u540d":67,"\u662f\u8be5\u7f51\u7edc\u5c42\u7684":67,"\u662f\u8f93\u5165":64,"\u662f\u8fd9\u4e00\u7c7b\u7684":50,"\u662f\u901a\u7528\u7269\u4f53\u5206\u7c7b\u9886\u57df\u4e00\u4e2a\u4f17\u6240\u5468\u77e5\u7684\u6570\u636e\u5e93":87,"\u662f\u9700\u8981\u4e86\u89e3\u54ea\u4e9b\u6b65\u9aa4\u62d6\u6162\u4e86\u6574\u4f53":72,"\u662fc":34,"\u662fdecoder\u7684\u6570\u636e\u8f93\u5165":63,"\u662fgoogle\u5f00\u6e90\u7684\u5bb9\u5668\u96c6\u7fa4\u7ba1\u7406\u7cfb\u7edf":78,"\u662fnvidia\u6027\u80fd\u5206\u6790\u5de5\u5177":72,"\u662fpaddlepaddle\u652f\u6301\u7684\u4e00\u79cd\u4efb\u610f\u590d\u6742\u7684rnn\u5355\u5143":63,"\u662fpaddlepaddle\u8d1f\u8d23\u63d0\u4f9b\u6570\u636e\u7684\u6a21\u5757":89,"\u662fpod\u5185\u7684\u5bb9\u5668\u90fd\u53ef\u4ee5\u8bbf\u95ee\u7684\u5171\u4eab\u76ee\u5f55":78,"\u662fpython\u5c01\u88c5\u7684\u7c7b\u540d":67,"\u662frnn\u72b6\u6001":64,"\u663e":89,"\u663e\u5f0f\u6307\u5b9a\u8fd4\u56de\u7684\u662f\u4e00\u4e2a28":2,"\u665a":61,"\u666e\u901a\u7528\u6237\u8bf7\u8d70\u5b89\u88c5\u6d41\u7a0b":55,"\u6682\u4e0d\u8003\u8651\u5728\u5185":49,"\u6682\u65f6\u4e0d\u8003\u8651\u591a\u4e2aparamet":20,"\u66b4\u9732\u8fd9\u4e2a\u6982\u5ff5\u5fc5\u8981\u51fd\u6570":34,"\u66f4\u522b\u63d0\u7b80\u5316\u95ee\u9898\u590d\u73b0\u5e26\u6765\u7684\u597d\u5904\u4e86":65,"\u66f4\u591a\u5173\u4e8edocker\u7684\u5b89\u88c5\u4e0e\u4f7f\u7528":46,"\u66f4\u591a\u5185\u5bb9\u53ef\u67e5\u770b\u53c2\u8003\u6587\u732e":93,"\u66f4\u591a\u7684\u7ec6\u8282\u53ef\u4ee5\u5728\u6587\u732e\u4e2d\u627e\u5230":93,"\u66f4\u591a\u7684\u7f16\u8bd1\u914d\u7f6e\u89c1":59,"\u66f4\u591a\u7684\u8f6c\u6362\u65b9\u6cd5\u8bf7\u53c2\u8003eigen":69,"\u66f4\u597d\u5730\u5b8c\u6210\u4e00\u4e9b\u590d\u6742\u7684\u8bed\u8a00\u7406\u89e3\u4efb\u52a1":63,"\u66f4\u5feb":64,"\u66f4\u65b0":46,"\u66f4\u65b0\u53ef\u80fd\u5bfc\u81f4\u9700\u8981\u65b0\u7684\u5f00\u53d1\u5de5\u5177":65,"\u66f4\u65b0\u6a21\u5f0f":49,"\u66f4\u65b0\u7684\u6587\u6863\u4ee5pr\u7684\u5f62\u5f0f\u63d0\u4ea4\u5230github\u4e2d":70,"\u66f4\u65b0\u7f51\u7edc\u53c2\u6570\u65f6\u5e94\u7528":49,"\u66f4\u65b9\u4fbf\u7684\u8bbe\u7f6e\u65b9\u5f0f":51,"\u66f4\u8be6\u7ec6\u6570\u636e\u683c\u5f0f\u548c\u7528\u4f8b\u8bf7\u53c2\u8003":89,"\u66f4\u8be6\u7ec6\u7684\u7f51\u7edc\u914d\u7f6e\u8fde\u63a5\u8bf7\u53c2\u8003":89,"\u66f4\u8be6\u7ec6\u7684\u8bf4\u660e":89,"\u66f4\u8fdb\u4e00\u6b65":63,"\u66f4\u9ad8":64,"\u66ff\u6211\u4eec\u5b8c\u6210\u4e86\u539f\u59cb\u8f93\u5165\u6570\u636e\u7684\u62c6\u5206":63,"\u6700":61,"\u6700\u4e0d\u540c\u7684\u7279\u8272\u662f\u5b83\u5e76\u6ca1\u6709\u5c06\u8f93\u5165\u8bed\u53e5\u7f16\u7801\u4e3a\u4e00\u4e2a\u5355\u72ec\u7684\u5b9a\u957f\u5411\u91cf":94,"\u6700\u4e3b\u8981\u7684\u5de5\u4f5c\u5c31\u662f\u89e3\u6790\u51fa":80,"\u6700\u4f73\u63a8\u8350":2,"\u6700\u540e":[2,66,67,73,89,93],"\u6700\u540e\u4e00\u4e2a":60,"\u6700\u540e\u4e00\u5c42cost\u4e2d\u8bb0\u5f55\u4e86\u795e\u7ecf\u7f51\u7edc\u7684\u6240\u6709\u62d3\u6251\u7ed3\u6784":56,"\u6700\u540e\u4e00\u90e8\u5206\u662f\u795e\u7ecf\u7f51\u7edc\u7684\u914d\u7f6e":52,"\u6700\u540e\u518d\u8c03\u7528mutabl":69,"\u6700\u540e\u5220\u9664":42,"\u6700\u540e\u6211\u4eec\u4f7f\u7528\u94fe\u5f0f\u6cd5\u5219\u8ba1\u7b97":67,"\u6700\u597d\u7684\u6a21\u578b\u662f":93,"\u6700\u5c11\u663e\u793a\u591a\u5c11\u4e2a\u8282\u70b9":75,"\u6700\u5e38\u89c1\u7684\u9519\u8bef\u5904\u7406\u65b9\u5f0f\u662fexcept":33,"\u6700\u65b0\u7684\u4ee3\u7801":66,"\u6700\u65b0\u7684paddlepaddl":46,"\u6700\u65b0log":93,"\u6700\u7ec8":67,"\u6700\u7ec8\u5b9e\u73b0\u4e00\u4e2a\u5c42\u6b21\u5316\u7684\u590d\u6742rnn":63,"\u6700\u7ec8\u6211\u4eec\u53ef\u4ee5\u8c03\u7528trainer\u7684train\u65b9\u6cd5\u542f\u52a8\u8bad\u7ec3":56,"\u6700\u7ec8\u751f\u6210\u540d\u4e3apaddl":54,"\u6700\u7ec8\u7684\u8f93\u51fa\u7ed3\u679c":63,"\u6700\u7ec8\u8d8b\u4e8e\u63a5\u8fd1":52,"\u6708\u6e56":61,"\u6709":61,"\u6709\u4e00\u4e2a\u57fa\u672c\u7684\u8ba4\u8bc6":78,"\u6709\u4e00\u4e9b\u5fc5\u987b\u914d\u7f6e\u7684\u53c2\u6570":[58,59],"\u6709\u4e24\u4e2a\u7279\u6b8a\u6807\u8bc6":94,"\u6709\u4e86meta\u914d\u7f6e\u6587\u4ef6\u4e4b\u540e":91,"\u6709\u4e9b\u5c42\u53ef\u80fd\u9700\u8981\u9ad8\u7cbe\u5ea6\u6765\u4fdd\u8bc1\u68af\u5ea6\u68c0\u67e5\u5355\u6d4b\u6b63\u786e\u6267\u884c":67,"\u6709\u4e9b\u5c42\u6216\u8005\u6fc0\u6d3b\u9700\u8981\u505a\u5f52\u4e00\u5316\u4ee5\u4fdd\u8bc1\u5b83\u4eec\u7684\u8f93\u51fa\u7684\u548c\u662f\u4e00\u4e2a\u5e38\u6570":67,"\u6709\u4e9b\u7279\u5f81\u7684\u53d6\u503c\u8fbe\u5230\u6570\u767e\u4e07":49,"\u6709\u4e9b\u7535\u5f71id\u53ef\u80fd\u4e0e\u5b9e\u9645\u7535\u5f71\u4e0d\u76f8\u7b26\u5408":90,"\u6709\u4eba\u7528\u865a\u62df\u673a\u6765\u7c7b\u6bd4":65,"\u6709\u5173":61,"\u6709\u5173\u53c2\u6570\u914d\u7f6e\u7684\u8be6\u7ec6\u8bf4\u660e\u89c1":58,"\u6709\u5173\u7ebf\u6027\u56de\u5f52\u7684\u5b9e\u9645\u5e94\u7528":56,"\u6709\u5173kubernetes\u76f8\u5173\u6982\u5ff5\u4ee5\u53ca\u5982\u4f55\u642d\u5efa\u548c\u914d\u7f6ekubernetes\u96c6\u7fa4":80,"\u6709\u52a9\u4e8e\u7406\u89e3\u7528\u6237\u5bf9\u4e0d\u540c\u516c\u53f8":93,"\u6709\u52a9\u4e8e\u8bca\u65ad\u5206\u5e03\u5f0f\u9519\u8bef":73,"\u6709\u591a\u96be":65,"\u6709\u65f6\u5019\u6211\u4eec\u4f1a\u5e0c\u671b\u6e05\u7406\u6389\u5df2\u7ecf\u4e0b\u8f7d\u7684\u7b2c\u4e09\u65b9\u4f9d\u8d56\u4ee5\u53ca\u5df2\u7ecf\u7f16\u8bd1\u7684\u4e8c\u8fdb\u5236\u6587\u4ef6":65,"\u6709\u65f6\u5019\u6211\u4eec\u53ea\u60f3\u8fd0\u884c\u4e00\u4e2a\u7279\u5b9a\u7684\u5355\u5143\u6d4b\u8bd5":65,"\u6709\u65f6\u79f0\u4e3a":93,"\u6709\u6807\u51c6\u7684":33,"\u6709\u7684\u65f6\u5019":33,"\u6709\u7684\u65f6\u5019\u7b80\u7b80\u5355\u5355\u7684\u6539\u53d8\u5c31\u80fd\u5728\u6027\u80fd\u4e0a\u4ea7\u751f\u660e\u663e\u7684\u4f18\u5316\u6548\u679c":72,"\u6709\u7684\u8bdd\u9700\u8981\u5148\u5378\u8f7d":46,"\u670d\u52a1":61,"\u670d\u52a1\u5458":61,"\u672a\u5305\u542b\u5728\u5b57\u5178\u4e2d\u7684\u5355\u8bcd":94,"\u672a\u6307\u5b9a\u6309\u7167double\u7cbe\u5ea6\u7f16\u8bd1":51,"\u672a\u6807\u8bb0\u7684\u8bc4\u4ef7\u6837\u672c":93,"\u672a\u77e5\u8bcd":85,"\u672c\u4f8b\u4e2d\u4e3a0":85,"\u672c\u4f8b\u4e2d\u4e3a32":85,"\u672c\u4f8b\u4e2d\u4e3a4":85,"\u672c\u4f8b\u4e2d\u4f7f\u7528for\u5faa\u73af\u8fdb\u884c\u591a\u6b21\u8c03\u7528":2,"\u672c\u4f8b\u4e2d\u7684\u539f\u59cb\u6570\u636e\u4e00\u5171\u670910\u4e2a\u6837\u672c":61,"\u672c\u4f8b\u4e2d\u7684\u8f93\u5165\u7279\u5f81\u662f\u8bcdid\u7684\u5e8f\u5217":2,"\u672c\u4f8b\u6839\u636e\u7f51\u7edc\u914d\u7f6e\u4e2d":2,"\u672c\u4f8b\u6bcf\u884c\u4fdd\u5b58\u4e00\u6761\u6837\u672c":89,"\u672c\u4f8b\u7531\u6613\u5230\u96be\u5c55\u793a4\u79cd\u4e0d\u540c\u7684\u6587\u672c\u5206\u7c7b\u7f51\u7edc\u914d\u7f6e":89,"\u672c\u4f8b\u7684":2,"\u672c\u4f8b\u7684\u6240\u6709\u5b57\u7b26\u90fd\u5c06\u8f6c\u6362\u4e3a\u8fde\u7eed\u6574\u6570\u8868\u793a\u7684id\u4f20\u7ed9\u6a21\u578b":89,"\u672c\u4f8b\u91c7\u7528\u82f1\u6587\u60c5\u611f\u5206\u7c7b\u7684\u6570\u636e":2,"\u672c\u4f8b\u91c7\u7528adam\u4f18\u5316\u65b9\u6cd5":89,"\u672c\u5217\u8868\u8bf4\u660epaddlepaddle\u53d1\u7248\u4e4b\u524d\u9700\u8981\u6d4b\u8bd5\u7684\u529f\u80fd\u70b9":42,"\u672c\u5730":46,"\u672c\u5730\u6d4b\u8bd5":74,"\u672c\u5730\u8bad\u7ec3":74,"\u672c\u5730\u8bad\u7ec3\u4e0e\u9884\u6d4b":48,"\u672c\u5730\u8bad\u7ec3\u7684\u5b9e\u9a8c":77,"\u672c\u5b9e\u4f8b\u4e2d":85,"\u672c\u5c0f\u8282\u6211\u4eec\u5c06\u4ecb\u7ecd\u6a21\u578b\u7f51\u7edc\u7ed3\u6784":89,"\u672c\u5c42\u5c3a\u5bf8":87,"\u672c\u5c42\u6709\u56db\u4e2a\u53c2\u6570":87,"\u672c\u6559\u7a0b\u4e2d\u6211\u4eec\u7ed9\u51fa\u4e86\u4e09\u4e2aresnet\u6a21\u578b":87,"\u672c\u6559\u7a0b\u4e3b\u8981\u4ecb\u7ecd\u5e26kernel\u7684op\u5982\u4f55\u5199":68,"\u672c\u6559\u7a0b\u5c06\u4ecb\u7ecd\u4f7f\u7528\u6df1\u5ea6\u53cc\u5411\u957f\u77ed\u671f\u8bb0\u5fc6":92,"\u672c\u6559\u7a0b\u5c06\u6307\u5bfc\u4f60\u5982\u4f55\u5728":64,"\u672c\u6559\u7a0b\u5c06\u6307\u5bfc\u60a8\u5b8c\u6210\u957f\u671f\u77ed\u671f\u8bb0\u5fc6":93,"\u672c\u6559\u7a0b\u63d0\u4f9b\u4e86\u4e00\u4e2a\u7528\u4e8eimagenet\u4e0a\u7684\u5377\u79ef\u5206\u7c7b\u7f51\u7edc\u6a21\u578b":87,"\u672c\u6587\u4e2d\u6240\u6709\u7684\u4f8b\u5b50":61,"\u672c\u6587\u4e2d\u7531\u4e8e\u8f93\u5165\u6570\u636e\u662f\u968f\u673a\u751f\u6210\u7684\u4e0d\u9700\u8981\u8bfb\u8f93\u5165\u6587\u4ef6":52,"\u672c\u6587\u4e2d\u7684\u4f8b\u5b50\u91cc":65,"\u672c\u6587\u4e2d\u793a\u4f8b\u6240\u4f7f\u7528\u7684\u5355\u5143\u6d4b\u8bd5\u6587\u4ef6\u662f":61,"\u672c\u6587\u4ee5paddlepaddle\u7684\u53cc\u5c42rnn\u5355\u5143\u6d4b\u8bd5\u4e3a\u793a\u4f8b":61,"\u672c\u6587\u53ea\u4f7f\u7528\u4e86\u9ed8\u8ba4\u547d\u540d\u7a7a\u95f4":78,"\u672c\u6587\u5c06\u4ecb\u7ecd\u5728kubernetes\u5bb9\u5668\u7ba1\u7406\u5e73\u53f0\u4e0a\u5feb\u901f\u6784\u5efapaddlepaddle\u5bb9\u5668\u96c6\u7fa4":80,"\u672c\u6587\u6863\u4ecb\u7ecd\u5982\u4f55\u5728paddlepaddle\u5e73\u53f0\u4e0a":85,"\u672c\u6587\u6863\u5185\u4e0d\u91cd\u590d\u4ecb\u7ecd":78,"\u672c\u6587\u6863\u5c06\u4ee5linux":58,"\u672c\u6587\u6863\u63cf\u8ff0paddl":34,"\u672c\u6765":61,"\u672c\u6b21\u8bad\u7ec3\u6587\u4ef6\u6240\u5728\u76ee\u5f55":80,"\u672c\u6b21\u8bad\u7ec3\u7684yaml\u6587\u4ef6\u53ef\u4ee5\u5199\u6210":80,"\u672c\u6b21\u8bad\u7ec3\u8981\u6c42\u67093\u4e2apaddlepaddle\u8282\u70b9":80,"\u672c\u6b21\u8bd5\u9a8c":89,"\u672c\u793a\u4f8b\u4e2d\u4f7f\u7528\u7684\u539f\u59cb\u6570\u636e\u5982\u4e0b":61,"\u672c\u793a\u4f8b\u610f\u56fe\u4f7f\u7528\u5355\u5c42rnn\u548c\u53cc\u5c42rnn\u5b9e\u73b0\u4e24\u4e2a\u5b8c\u5168\u7b49\u4ef7\u7684\u5168\u8fde\u63a5rnn":61,"\u672c\u793a\u4f8b\u7684\u9884\u6d4b\u7ed3\u679c":93,"\u672c\u7bc7\u6559\u7a0b\u5728paddlepaddle\u4e2d\u91cd\u73b0\u4e86\u8fd9\u4e00\u826f\u597d\u7684\u8bad\u7ec3\u7ed3\u679c":94,"\u672c\u7bc7\u6559\u7a0b\u5c06\u4f1a\u6307\u5bfc\u4f60\u901a\u8fc7\u8bad\u7ec3\u4e00\u4e2a":94,"\u672c\u8d28\u4e0a\u4e0e\u8bad\u7ec3\u6a21\u578b\u4e00\u6837":94,"\u673a\u5668\u7684\u8bbe\u5907":77,"\u673a\u5668\u7ffb\u8bd1":[42,88],"\u6743\u91cd\u66f4\u65b0\u7684\u68af\u5ea6":75,"\u6761\u4ef6\u4e0b":78,"\u6765":61,"\u6765\u4e0b\u8f7ddocker\u955c\u50cf":54,"\u6765\u4ee3\u66ff":66,"\u6765\u4f7f\u7528\u4e00\u4e2a\u955c\u50cf\u542f\u52a8\u4e00\u4e2a\u5bb9\u5668":54,"\u6765\u4f7f\u7528dropout":50,"\u6765\u4f7f\u7528dropout\u7684":50,"\u6765\u4fdd\u8bc1\u8bad\u7ec3\u8fc7\u7a0b\u53ef\u4ee5\u4ece\u4e2d\u95f4\u72b6\u6001\u91cd\u65b0\u542f\u52a8":20,"\u6765\u505a\u68af\u5ea6\u68c0\u67e5":67,"\u6765\u505ableu\u8bc4\u4f30":94,"\u6765\u505c\u6b62\u8bad\u7ec3":91,"\u6765\u5206\u6790\u6267\u884c\u6587\u4ef6":72,"\u6765\u5206\u79bb\u8bad\u7ec3\u548c\u6d4b\u8bd5\u6587\u4ef6":91,"\u6765\u5206\u9694\u6bcf\u4e00\u884c":91,"\u6765\u5217\u51fa\u5f53\u524d\u7cfb\u7edf\u4e2d\u7684\u6240\u6709\u955c\u50cf":54,"\u6765\u521d\u59cb\u5316\u53c2\u6570":51,"\u6765\u542f\u52a8\u548c":65,"\u6765\u5b89\u88c5":73,"\u6765\u5b8c\u6210\u7f51\u7edc\u7684\u8bad\u7ec3":56,"\u6765\u5b9a\u4e49\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":64,"\u6765\u5bf9\u6bd4\u5206\u6790\u4e24\u8005\u8bed\u4e49\u76f8\u540c\u7684\u539f\u56e0":61,"\u6765\u5e2e\u52a9\u4f60\u7406\u89e3paddlepaddle\u7684\u5185\u90e8\u8fd0\u884c\u673a\u5236":89,"\u6765\u5f00\u542f\u672c\u5730\u7684\u8bad\u7ec3":93,"\u6765\u5f71\u54cdpaddlepaddle\u7684\u7f16\u8bd1\u8fc7\u7a0b":[58,59],"\u6765\u5f97\u5230\u67d0\u4e2a\u7279\u5b9a\u53c2\u6570\u7684\u68af\u5ea6\u77e9\u9635":67,"\u6765\u6267\u884c":65,"\u6765\u6307\u5b9a\u7f51\u7edc\u5c42\u7684\u6570\u76ee":87,"\u6765\u63a5\u53d7\u4e0d\u4f7f\u7528\u7684\u51fd\u6570\u4ee5\u4fdd\u8bc1\u517c\u5bb9\u6027":2,"\u6765\u63cf\u8ff0\u7684":69,"\u6765\u63cf\u8ff0\u8be5op\u7684\u8f93\u5165":68,"\u6765\u642d\u5efa\u795e\u7ecf\u7f51\u7edc":56,"\u6765\u6ce8\u518c\u8be5\u5c42":67,"\u6765\u6df7\u5408\u4f7f\u7528gpu\u548ccpu\u8ba1\u7b97\u7f51\u7edc\u5c42\u7684\u53c2\u6570":77,"\u6765\u6e05\u7406\u8fd9\u4e9b\u5185\u5bb9":65,"\u6765\u751f\u6210\u5e8f\u5217":94,"\u6765\u7684\u79d2\u6570":90,"\u6765\u786e\u4fdd\u628a":33,"\u6765\u786e\u5b9a\u5bf9\u5e94\u5173\u7cfb":2,"\u6765\u7f16\u8bd1":65,"\u6765\u81ea\u5b9a\u4e49\u4f20\u6570\u636e\u7684\u8fc7\u7a0b":1,"\u6765\u83b7\u5f97\u8f93\u51fa\u7684\u68af\u5ea6":67,"\u6765\u8868\u793a":64,"\u6765\u8868\u793a\u53c2\u6570\u4f4d\u7f6e":92,"\u6765\u8868\u793a\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":64,"\u6765\u8868\u793apaddle\u5185\u90e8\u7c7b":33,"\u6765\u89e3\u51b3\u4e0a\u9762\u7684\u95ee\u9898":49,"\u6765\u8ba1\u7b97\u68af\u5ea6":67,"\u6765\u8bb2\u89e3\u5982\u4f55\u4f7f\u7528\u53cc\u5c42rnn":61,"\u6765\u8bbe\u7f6e":51,"\u6765\u8bbf\u95ee\u7528\u6237\u81ea\u5df1\u7684\u6570\u636e":21,"\u6765\u8bf4\u660epydataprovider2\u7684\u7b80\u5355\u4f7f\u7528\u573a\u666f":2,"\u6765\u8c03\u6574":66,"\u6765\u8c03\u7528":65,"\u6765\u8fd0\u884c":73,"\u6765\u8fd0\u884c\u5305\u62ec":65,"\u6765\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5\u4e86":65,"\u6765\u8fd0\u884c\u6027\u80fd\u5206\u6790\u548c\u8c03\u4f18":72,"\u6765\u8fd0\u884c\u955c\u50cf":54,"\u6765\u8fdb\u884c\u8ba8\u8bba":34,"\u6765\u9884\u6d4b\u8fd9\u4e2a\u4e2d\u95f4\u7684\u8bcd":49,"\u676f\u5b50":61,"\u6784\u5efa":58,"\u6784\u6210\u4e86\u8f93\u51fa\u53cc\u5c42\u5e8f\u5217\u7684\u7b2ci\u4e2a":60,"\u6784\u9020":80,"\u6784\u9020\u51fd\u6570\u542b\u67092\u4e2a\u53c2\u6570":68,"\u6784\u9020\u51fd\u6570\u91cc\u901a\u8fc7":68,"\u6784\u9020paddl":4,"\u67b6\u6784\u6765\u8bad\u7ec3\u60c5\u611f\u5206\u6790\u6a21\u578b":93,"\u67d0\u4e00\u4e2a\u795e\u7ecf\u5143\u7684\u4e00\u4e2a\u8f93\u5165\u4e3a\u4e0a\u4e00\u4e2a\u65f6\u95f4\u6b65\u7f51\u7edc\u4e2d\u67d0\u4e00\u4e2a\u795e\u7ecf\u5143\u7684\u8f93\u51fa":61,"\u67d0\u4e9b\u53c2\u6570\u53ea\u53ef\u7528\u4e8e\u7279\u5b9a\u7684\u5c42\u4e2d":74,"\u67e5\u770b":[66,89],"\u67e5\u770b\u5f53\u524d\u72b6\u6001":66,"\u67e5\u770b\u5f53\u524d\u8fdc\u7a0b\u4ed3\u5e93\u7684\u540d\u5b57":66,"\u67e5\u770b\u6587\u4ef6\u5177\u4f53\u88ab\u4fee\u6539\u7684\u5185\u5bb9":66,"\u67e5\u770b\u662f\u5426\u662f\u5176\u4ed6\u9519\u8bef\u5f15\u53d1\u7684\u62a5\u9519":47,"\u67e5\u770bjob\u7684\u8be6\u7ec6\u60c5\u51b5":79,"\u6807\u51c6\u5dee\u4e3a":51,"\u6807\u51c6\u8868\u793apaddlepaddle\u7248\u672c\u53f7":42,"\u6807\u51c6lstm\u4ee5\u6b63\u5411\u5904\u7406\u8be5\u5e8f\u5217":92,"\u6807\u793a\u56fe\u7247\u662f\u5f69\u8272\u56fe\u6216\u7070\u5ea6\u56fe":86,"\u6807\u793a\u662f\u5426\u4e3a\u5f69\u8272\u56fe\u7247":86,"\u6807\u7b7e0\u8868\u793a\u8d1f\u9762\u7684\u8bc4\u8bba":93,"\u6807\u7b7e1\u8868\u793a\u6b63\u9762\u7684\u8bc4\u8bba":93,"\u6807\u7b7e\u4e0b\u627e\u5230\u6700\u65b0\u7684paddle\u955c\u50cf\u7248\u672c":54,"\u6807\u7b7e\u6587\u4ef6":92,"\u6807\u7b7e\u65b9\u6848\u6765\u81ea":92,"\u6807\u8bc6\u662f\u5426\u4e3a\u8fde\u7eed\u7684batch\u8ba1\u7b97":75,"\u6838\u4e00\u6837\u591a\u7684\u8fdb\u7a0b\u6765\u5e76\u884c\u7f16\u8bd1":65,"\u6839\u636e\u4f60\u7684\u4efb\u52a1":77,"\u6839\u636e\u524d\u6587\u7684\u63cf\u8ff0":80,"\u6839\u636e\u5728\u6a21\u578b\u914d\u7f6e\u6587\u4ef6\u4e2d\u4f7f\u7528\u7684\u540d\u4e3a":73,"\u6839\u636e\u6570\u636e\u91cf\u89c4\u6a21":90,"\u6839\u636e\u7528\u6237\u6307\u5b9a\u7684\u5b57\u5178":85,"\u6839\u636e\u7f51\u7edc\u914d\u7f6e\u4e2d\u7684":75,"\u6839\u636e\u8f93\u5165tensor\u7684\u5927\u5c0f\u6765\u8bbe\u7f6e\u8f93\u51fatensor\u7684\u5927\u5c0f":69,"\u6839\u636e\u8fd9\u4e9b\u53c2\u6570\u7684\u4f7f\u7528\u573a\u5408":74,"\u6839\u636e\u9ed8\u8ba4\u503c\u9012\u589e":75,"\u6839\u636e\u9ed8\u8ba4\u7aef\u53e3\u53f7\u9012\u589e":75,"\u6839\u636ecpu":54,"\u6839\u636ejob\u5bf9\u5e94\u7684pod\u4fe1\u606f":79,"\u683c\u5f0f":75,"\u683c\u5f0f\u5982\u4e0b":89,"\u683c\u5f0f\u7684\u6587\u4ef6\u6765\u5b58\u653e":68,"\u683c\u5f0f\u8bf4\u660e":85,"\u68af\u5ea6\u4f1a\u5c31\u5730":67,"\u68af\u5ea6\u53c2\u6570\u7684\u5206\u5757\u6570\u76ee":75,"\u68af\u5ea6\u5c31\u53ef\u4ee5\u901a\u8fc7\u8fd9\u4e2a\u65b9\u7a0b\u8ba1\u7b97\u5f97\u5230":67,"\u68af\u5ea6\u670d\u52a1\u5668\u7684\u6570\u91cf":75,"\u68af\u5ea6\u68c0\u67e5\u5355\u5143\u6d4b\u8bd5\u901a\u8fc7\u6709\u9650\u5dee\u5206\u6cd5\u6765\u9a8c\u8bc1\u4e00\u4e2a\u5c42\u7684\u68af\u5ea6":67,"\u68af\u5ea6\u68c0\u67e5\u7684\u8f93\u5165\u6570\u636e\u7684\u6279\u6b21\u5927\u5c0f":67,"\u68c0\u67e5\u70b9\u4fdd\u5b58\u7a0b\u5e8f\u6d41\u7a0b":20,"\u68c0\u67e5\u8f93\u5165\u6570\u636e\u7ef4\u5ea6":68,"\u68d2":89,"\u697c\u5c42":61,"\u6a21\u5757":86,"\u6a21\u5757\u4e0b\u7684\u76f8\u5173":69,"\u6a21\u5757\u4e2d\u7684":2,"\u6a21\u578b":92,"\u6a21\u578b\u4e00\u76f4\u4e0d\u6536\u655b":49,"\u6a21\u578b\u4fdd\u5b58\u5728\u76ee\u5f55":93,"\u6a21\u578b\u5171\u5305\u542b1":85,"\u6a21\u578b\u5217\u8868\u6587\u4ef6":92,"\u6a21\u578b\u53c2\u6570\u68c0\u67e5\u70b9\u901a\u8fc7\u5b9a\u671f\u5411\u78c1\u76d8\u4e0a\u4fdd\u5b58\u4e00\u4efd\u5b58\u50a8\u5728paramet":20,"\u6a21\u578b\u53ca\u53c2\u6570\u4f1a\u88ab\u4fdd\u5b58\u5728\u8def\u5f84":86,"\u6a21\u578b\u5b58\u50a8\u8def\u5f84":89,"\u6a21\u578b\u5c06\u4fdd\u5b58\u5728\u76ee\u5f55":92,"\u6a21\u578b\u5c31\u8bad\u7ec3\u6210\u529f\u4e86":94,"\u6a21\u578b\u6570\u636e\u68c0\u67e5\u70b9\u7684\u5b9e\u73b0":20,"\u6a21\u578b\u6587\u4ef6\u5c06\u88ab\u5199\u5165\u8282\u70b9":73,"\u6a21\u578b\u6765\u5c06\u6cd5\u8bed\u7ffb\u8bd1\u6210\u82f1\u8bed":94,"\u6a21\u578b\u6765\u6307\u5bfc\u4f60\u5b8c\u6210\u8fd9\u4e9b\u6b65\u9aa4":64,"\u6a21\u578b\u6f14\u793a\u5982\u4f55\u914d\u7f6e\u590d\u6742\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u6a21\u578b":64,"\u6a21\u578b\u7684\u4ee3\u7801\u53ef\u4ee5\u5728":64,"\u6a21\u578b\u7684\u7ed3\u6784\u548c\u8bad\u7ec3\u8fc7\u7a0b":85,"\u6a21\u578b\u7684\u7f16\u7801\u5668\u90e8\u5206\u5982\u4e0b\u6240\u793a":64,"\u6a21\u578b\u88ab\u4fdd\u5b58\u5728":91,"\u6a21\u578b\u8bad\u7ec3\u4f1a\u770b\u5230\u7c7b\u4f3c\u4e0a\u9762\u8fd9\u6837\u7684\u65e5\u5fd7\u4fe1\u606f":89,"\u6a21\u578b\u8bad\u7ec3\u548c\u6700\u540e\u7684\u7ed3\u679c\u9a8c\u8bc1":52,"\u6a21\u578b\u8bad\u7ec3\u7b49\u4efb\u52a1":56,"\u6a21\u578b\u8def\u5f84":[87,92],"\u6a21\u578b\u8f93\u51fa\u8def\u5f84":92,"\u6a21\u578b\u914d\u7f6e":[0,48],"\u6a21\u578b\u914d\u7f6e\u89e3\u6790":33,"\u6a21\u578b\u91c7\u7528":85,"\u6a21\u578b\u9884\u6d4b":4,"\u6a2a\u5411\u62fc\u63a5":49,"\u6b21":61,"\u6b22\u8fce\u901a\u8fc7":66,"\u6b63\u6837\u672c":89,"\u6b63\u786e\u7684\u89e3\u51b3\u65b9\u6cd5\u662f":46,"\u6b63\u8d1f\u5bf9\u9a8c\u8bc1":74,"\u6b63\u9762\u7684\u8bc4\u8bba\u7684\u5f97\u5927\u4e8e\u7b49\u4e8e7":93,"\u6b63\u9762\u8bc4\u4ef7\u6837\u672c":93,"\u6b64\u547d\u4ee4\u5c06\u5728your":58,"\u6b64\u5904":85,"\u6b64\u5904\u90fd\u4e3a2":61,"\u6b64\u5916":[50,58,65,66],"\u6b64\u6559\u7a0b\u5c06\u5411\u60a8\u5206\u6b65\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528\u5185\u7f6e\u7684\u5b9a\u65f6\u5de5\u5177":72,"\u6b64\u6570\u636e\u96c6\u5305\u542b\u7535\u5f71\u8bc4\u8bba\u53ca\u5176\u76f8\u5173\u8054\u7684\u7c7b\u522b\u6807\u7b7e":93,"\u6b64\u65f6\u53ea\u9700\u8981":65,"\u6b64\u65f6\u53ef\u4ee5\u5728\u8c03\u7528infer\u63a5\u53e3\u65f6\u901a\u8fc7\u8bbe\u7f6e":49,"\u6b64\u65f6\u53ef\u4ee5\u8df3\u8fc7paddlepaddle\u6a21\u578b\u53c2\u6570\u6587\u4ef6\u7684\u5934\u4fe1\u606f":51,"\u6b64\u65f6\u60a8\u53ef\u4ee5\u8fd0\u884c\u8fd9\u4e2a\u547d\u4ee4\u5728\u5f00\u53d1\u673a\u4e0a\u8fdb\u884c\u6d4b\u8bd5\u8fd0\u884c":54,"\u6b64\u65f6master\u5c06\u8d1f\u8d23\u542f\u52a8\u4e00\u4e2a\u65b0\u7684train":20,"\u6b64\u7c7b\u62a5\u9519\u901a\u5e38\u662f\u7531\u4e8e\u67d0\u4e00\u4e2a\u8282\u70b9\u7684\u9519\u8bef\u5bfc\u81f4\u8fd9\u4e2a\u8282\u70b9\u7684\u8bad\u7ec3\u8fdb\u7a0b\u9000\u51fa":47,"\u6b65\u9aa4":49,"\u6bb5\u843d\u53ef\u4ee5\u770b\u4f5c\u662f\u4e00\u4e2a\u5d4c\u5957\u7684\u53cc\u5c42\u7684\u5e8f\u5217":63,"\u6bcf100\u4e2abatch\u6253\u5370\u4e00\u6b21\u7edf\u8ba1\u4fe1\u606f":93,"\u6bcf100\u4e2abatch\u663e\u793a\u53c2\u6570\u7edf\u8ba1":92,"\u6bcf20\u4e2abatch\u6253\u5370\u4e00\u6b21\u65e5\u5fd7":93,"\u6bcf20\u4e2abatch\u8f93\u51fa\u65e5\u5fd7":92,"\u6bcf\u4e00\u4e2a":42,"\u6bcf\u4e00\u4e2a\u4efb\u52a1\u6d41\u7a0b\u90fd\u53ef\u4ee5\u88ab\u5212\u5206\u4e3a\u5982\u4e0b\u4e94\u4e2a\u6b65\u9aa4":89,"\u6bcf\u4e00\u4e2a\u6587\u4ef6\u662f\u6570\u636e\u96c6\u7684\u4e00\u4e2ashard":21,"\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65":61,"\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u4e4b\u95f4\u7684\u795e\u7ecf\u7f51\u7edc\u5177\u6709\u4e00\u5b9a\u7684\u76f8\u5173\u6027":61,"\u6bcf\u4e00\u4e2a\u6d4b\u8bd5\u5468\u671f\u6d4b\u8bd5\u4e00\u6b21\u6240\u6709\u6570\u636e":91,"\u6bcf\u4e00\u4e2a\u8282\u70b9\u90fd\u6709\u76f8\u540c\u7684\u65e5\u5fd7\u7ed3\u6784":73,"\u6bcf\u4e00\u4e2akey\u7531":91,"\u6bcf\u4e00\u4e2alayer\u8f93\u51fa\u77e9\u9635\u7684\u9ad8\u5ea6":49,"\u6bcf\u4e00\u7ec4\u5185\u7684\u6240\u6709\u53e5\u5b50\u548clabel":61,"\u6bcf\u4e00\u884c\u8868\u793a\u4e00\u4e2a\u5b9e\u4f8b":93,"\u6bcf\u4e2a":[73,92],"\u6bcf\u4e2a\u503c\u7684\u7c7b\u578b\u53ef\u4ee5\u662f\u6574\u5f62":21,"\u6bcf\u4e2a\u5143\u7d20\u662f\u4e00\u4e2a0\u5c42\u5e8f\u5217":60,"\u6bcf\u4e2a\u5143\u7d20\u662f\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":60,"\u6bcf\u4e2a\u5355\u5c42rnn":63,"\u6bcf\u4e2a\u5355\u8bcd\u7684\u9884\u6d4b\u9519\u8bef\u7387":94,"\u6bcf\u4e2a\u53e5\u5b50\u53c8\u662f\u5355\u8bcd\u7684\u6570\u7ec4":61,"\u6bcf\u4e2a\u53e5\u5b50\u90fd\u4ee5\u5f00\u59cb\u6807\u8bb0\u5f00\u5934":64,"\u6bcf\u4e2a\u53e5\u5b50\u90fd\u4ee5\u7ed3\u675f\u6807\u8bb0\u7ed3\u5c3e":64,"\u6bcf\u4e2a\u5b50\u5e8f\u5217\u957f\u5ea6\u53ef\u4ee5\u4e0d\u4e00\u81f4":61,"\u6bcf\u4e2a\u5b50\u6587\u4ef6\u5939\u4e0b\u5b58\u50a8\u76f8\u5e94\u5206\u7c7b\u7684\u56fe\u7247":86,"\u6bcf\u4e2a\u5b57\u5178\u5305\u542b\u603b\u517130000\u4e2a\u5355\u8bcd":94,"\u6bcf\u4e2a\u5b57\u5178\u90fd\u6709dictsize\u4e2a\u5355\u8bcd":94,"\u6bcf\u4e2a\u5c42\u5728\u5176":67,"\u6bcf\u4e2a\u5c42\u90fd\u6709\u4e00\u4e2a\u6216\u591a\u4e2ainput":89,"\u6bcf\u4e2a\u6279\u6b21\u6570\u636e":75,"\u6bcf\u4e2a\u6587\u4ef6\u5939\u90fd\u5305\u542b\u6cd5\u8bed\u5230\u82f1\u8bed\u7684\u5e73\u884c\u8bed\u6599\u5e93":94,"\u6bcf\u4e2a\u6587\u4ef6\u662f\u4e00\u4e2a\u7535\u5f71\u8bc4\u8bba":93,"\u6bcf\u4e2a\u6587\u672c\u6587\u4ef6\u5305\u542b\u4e00\u4e2a\u6216\u8005\u591a\u4e2a\u5b9e\u4f8b":93,"\u6bcf\u4e2a\u65f6\u95f4\u6b65\u4e4b\u5185\u7684\u8fd0\u7b97\u662f\u72ec\u7acb\u7684":63,"\u6bcf\u4e2a\u65f6\u95f4\u6b65\u90fd\u7528\u4e86\u4e0a\u4e00\u4e2a\u65f6\u95f4\u6b65\u7684\u8f93\u51fa\u7ed3\u679c":61,"\u6bcf\u4e2a\u6743\u91cd\u5bf9\u5e94\u4e00\u4e2a\u8f93\u5165":67,"\u6bcf\u4e2a\u6837\u672c\u7531\u4e24\u90e8\u5206\u7ec4\u6210":61,"\u6bcf\u4e2a\u6837\u672c\u95f4\u7528\u7a7a\u884c\u5206\u5f00":61,"\u6bcf\u4e2a\u6d4b\u8bd5\u5468\u671f\u6d4b\u8bd5":91,"\u6bcf\u4e2a\u7279\u5f81\u7684meta\u914d\u7f6e":91,"\u6bcf\u4e2a\u72b6\u6001":63,"\u6bcf\u4e2a\u7c7b\u522b\u4e2d\u968f\u673a\u62bd\u53d6\u4e8610\u5f20\u56fe\u7247":86,"\u6bcf\u4e2a\u7c7b\u5305\u542b6000\u5f20":86,"\u6bcf\u4e2a\u7ebf\u7a0b":75,"\u6bcf\u4e2a\u7ebf\u7a0b\u5206\u914d\u5230128\u4e2a\u6837\u672c\u7528\u4e8e\u8bad\u7ec3":75,"\u6bcf\u4e2a\u8282\u70b9\u6709\u4e24\u4e2a6\u6838cpu":94,"\u6bcf\u4e2a\u8bad\u7ec3\u8282\u70b9\u5fc5\u987b\u6307\u5b9a\u4e00\u4e2a\u552f\u4e00\u7684id\u53f7":75,"\u6bcf\u4e2a\u8bb0\u5fc6\u5355\u5143\u5305\u542b\u56db\u4e2a\u4e3b\u8981\u7684\u5143\u7d20":93,"\u6bcf\u4e2a\u8bc4\u8bba\u7684\u7f51\u5740":93,"\u6bcf\u4e2a\u8f93\u5165\u90fd\u662f\u4e00\u4e2a":67,"\u6bcf\u4e2a\u8f93\u51fa\u8282\u70b9\u90fd\u8fde\u63a5\u5230\u6240\u6709\u7684\u8f93\u5165\u8282\u70b9\u4e0a":67,"\u6bcf\u4e2a\u91cc\u9762\u90fd\u5305\u542b202mb\u7684\u5168\u90e8\u7684\u6a21\u578b\u53c2\u6570":94,"\u6bcf\u4e2acommit\u53ea\u505a\u4e86\u5c11\u91cf\u7684\u4fee\u6539":66,"\u6bcf\u4e2adata":21,"\u6bcf\u4e2aparamet":20,"\u6bcf\u4e2apass\u7684\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747\u5206\u7c7b\u9519\u8bef\u7387":89,"\u6bcf\u4e2apass\u7684\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u6240\u6709\u6837\u672c\u7684\u5e73\u5747cost":89,"\u6bcf\u4e2apod\u5305\u542b\u4e00\u4e2apaddlepaddle\u5bb9\u5668":80,"\u6bcf\u4e2ashard\u5206\u522b\u5b58\u50a8\u5728\u5176\u4e2d\u4e00\u53f0paramet":20,"\u6bcf\u4f4d\u7528\u6237\u81f3\u5c11\u670920\u6761\u8bc4\u5206":90,"\u6bcf\u5c42\u4e0a\u53ea\u80fd\u4fdd\u5b58\u56fa\u5b9a\u6570\u76ee\u4e2a\u6700\u597d\u7684\u72b6\u6001":75,"\u6bcf\u5c42\u4f7f\u7528\u7684gpu\u53f7\u4f9d\u8d56\u4e8e\u53c2\u6570train":77,"\u6bcf\u5f53\u6a21\u578b\u5728\u7ffb\u8bd1\u8fc7\u7a0b\u4e2d\u751f\u6210\u4e86\u4e00\u4e2a\u5355\u8bcd":94,"\u6bcf\u6279\u6b21":75,"\u6bcf\u6b21":54,"\u6bcf\u6b21\u63d0\u4ea4\u4ee3\u7801":66,"\u6bcf\u6b21\u63d0\u4ea4\u65f6":66,"\u6bcf\u6b21\u6d4b\u8bd5\u90fd\u6d4b\u8bd5\u6240\u6709\u6570\u636e":93,"\u6bcf\u6b21\u751f\u62101\u4e2a\u5e8f\u5217":94,"\u6bcf\u6b21\u8bfb\u53d6\u4e00\u6761\u6570\u636e\u540e":89,"\u6bcf\u6b21\u8f93\u51fa\u4e00\u4e2adata":21,"\u6bcf\u6b21\u90fd\u4f1a\u4ecepython\u7aef\u8bfb\u53d6\u6570\u636e":2,"\u6bcf\u884c\u5b58\u50a8\u4e00\u4e2a\u8bcd":85,"\u6bcf\u884c\u5b58\u50a8\u7684\u662f\u4e00\u4e2a\u6837\u672c\u7684\u7279\u5f81":87,"\u6bcf\u884c\u6253\u537032\u4e2a\u53c2\u6570\u4ee5":85,"\u6bcf\u884c\u8868\u793a\u4e00\u4e2a\u6279\u6b21\u4e2d\u7684\u5355\u4e2a\u8f93\u5165":67,"\u6bcf\u884c\u90fd\u662f\u4e00\u4e2a\u6cd5\u8bed\u6216\u8005\u82f1\u8bed\u7684\u53e5\u5b50":94,"\u6bcf\u8f6e\u4f1a\u5c06\u6570\u636e\u96c6\u4e2d\u7684\u6240\u6709\u8bad\u7ec3\u6837\u672c\u4f7f\u7528\u4e00\u6b21":75,"\u6bcf\u8f6e\u7ed3\u675f\u65f6\u5bf9\u6240\u6709\u6d4b\u8bd5\u6570\u636e\u8fdb\u884c\u6d4b\u8bd5":75,"\u6bcf\u8f6e\u90fd\u4f1a\u4fdd\u5b58\u9884\u6d4b\u7ed3\u679c":75,"\u6bcf\u8fd0\u884c\u591a\u5c11\u4e2a\u6279\u6b21\u6267\u884c\u4e00\u6b21\u7a00\u758f\u53c2\u6570\u5206\u5e03\u7684\u68c0\u67e5":75,"\u6bcf\u969410\u5206\u949f":20,"\u6bcf\u9694\u591a\u5c11batch\u6253\u5370\u4e00\u6b21\u65e5\u5fd7":89,"\u6bcfdot":75,"\u6bcflog":75,"\u6bcfsave":75,"\u6bcftest":75,"\u6bd4\u5982":[21,47,49,54,58,65,66,89],"\u6bd4\u5982\u4e00\u53e5\u8bdd\u4e2d\u7684\u6bcf\u4e00\u4e2a\u5355\u8bcd":61,"\u6bd4\u5982\u5982\u679c\u8981build\u4e00\u4e2a\u4e0d\u4f9d\u8d56gpu":66,"\u6bd4\u5982\u5c06":42,"\u6bd4\u5982\u5e0c\u671b\u6700\u5c0f\u5316\u751f\u6210\u7684\u5e93\u7684\u5927\u5c0f":[58,59],"\u6bd4\u5982\u6bcf\u969410\u5206\u949f\u6700\u65b0\u7684\u5feb\u7167":20,"\u6bd4\u5982\u6d41\u5f0f\u6570\u636e\u5904\u7406":21,"\u6bd4\u5982\u8bbe\u7f6e\u4e00\u4e2a\u5168\u8fde\u63a5\u5c42\u7684\u53c2\u6570\u521d\u59cb\u5316\u65b9\u5f0f\u548cbias\u521d\u59cb\u5316\u65b9\u5f0f":51,"\u6bd4\u5982\u901a\u8fc78080\u7aef\u53e3":78,"\u6bd4\u5982fpe":47,"\u6bd4\u5982ide\u914d\u7f6e\u91cc":66,"\u6bd4\u5982imagenet\u8fd9\u4e2a\u6570\u636e\u96c6\u53ef\u80fd\u88ab\u5206\u62101000\u4e2ashard":21,"\u6c34\u6e29":61,"\u6c49\u5ead":61,"\u6c60\u5316\u5c42":86,"\u6ca1":61,"\u6ca1\u6709\u4f5c\u7528":2,"\u6ca1\u6709\u5b9e\u9645\u610f\u4e49":85,"\u6ca1\u6709\u627e\u5230\u548c\u5f53\u524d\u7cfb\u7edf\u5339\u914d\u7684paddlepaddle\u5b89\u88c5\u5305":46,"\u6ca1\u6709\u6d4b\u8bd5\u6570\u636e":2,"\u6ca1\u6709\u8bbe\u7f6e":[58,59],"\u6ca1\u6709\u8fdb\u884c\u6b63\u786e\u6027\u7684\u68c0\u67e5":90,"\u6ca1\u6709\u8fdb\u884c\u7ed3\u6784\u7684\u5fae\u8c03":91,"\u6cd5\u8bed":94,"\u6ce8":[20,54],"\u6ce8\u518c":68,"\u6ce8\u518ccpu":68,"\u6ce8\u518cop":68,"\u6ce8\u518cop\u65f6\u7684\u7c7b\u578b\u540d":68,"\u6ce8\u610f":[2,53,54,56,58,59,64,67,80,86],"\u6ce8\u610f\u4e0a\u8ff0\u547d\u4ee4\u4e2d":80,"\u6ce8\u610f\u5230\u6211\u4eec\u5df2\u7ecf\u5047\u8bbe\u673a\u5668\u4e0a\u67094\u4e2agpu":77,"\u6ce8\u610f\u5e94\u8be5\u786e\u4fdd\u9ed8\u8ba4\u6a21\u578b\u8def\u5f84":93,"\u6ce8\u610f\u9884\u6d4b\u6570\u636e\u901a\u5e38\u4e0d\u5305\u542blabel":4,"\u6ce8\u610fnode":80,"\u6ce8\u91ca":68,"\u6ce8\u91ca\u6389":93,"\u6cf3\u6c60":61,"\u6d41":61,"\u6d41\u7a0b\u6765\u63d0\u4ea4\u4ee3\u7801":66,"\u6d44":61,"\u6d4b\u8bd5":66,"\u6d4b\u8bd5\u524d\u5411oper":68,"\u6d4b\u8bd5\u6570\u636e":73,"\u6d4b\u8bd5\u6570\u636e\u4e5f\u5305\u542b":73,"\u6d4b\u8bd5\u6570\u636e\u548c\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":73,"\u6d4b\u8bd5\u6570\u636e\u548c\u751f\u6210\u6570\u636e":94,"\u6d4b\u8bd5\u6570\u636e\u653e\u7f6e\u5728\u5de5\u4f5c\u7a7a\u95f4\u4e2d\u4e0d\u540c\u76ee\u5f55\u7684\u8981\u6c42":73,"\u6d4b\u8bd5\u6570\u636e\u7684\u6240\u6709\u76f8\u5bf9\u6216\u7edd\u5bf9\u6587\u4ef6\u8def\u5f84":73,"\u6d4b\u8bd5\u6570\u6910\u96c6":93,"\u6d4b\u8bd5\u65f6\u6307\u5b9a\u7684\u5b58\u50a8\u6a21\u578b\u5217\u8868\u7684\u6587\u4ef6":75,"\u6d4b\u8bd5\u65f6\u9ed8\u8ba4\u4e0dshuffl":2,"\u6d4b\u8bd5\u662f":66,"\u6d4b\u8bd5\u6837\u672c":73,"\u6d4b\u8bd5\u6a21\u578b\u662f\u6307\u4f7f\u7528\u8bad\u7ec3\u51fa\u7684\u6a21\u578b\u8bc4\u4f30\u5df2\u6807\u8bb0\u7684\u9a8c\u8bc1\u96c6":93,"\u6d4b\u8bd5\u7684\u6a21\u578b\u5305\u62ec\u4ece\u7b2cm\u8f6e\u5230\u7b2cn":77,"\u6d4b\u8bd5\u811a\u672c\u662f":92,"\u6d4b\u8bd5\u96c6\u548c\u8bad\u7ec3\u96c6\u76ee\u5f55\u5305\u542b\u4e0b\u9762\u7684\u6587\u4ef6":93,"\u6d4b\u8bd5docker\u955c\u50cf":42,"\u6d4b\u8bd5model_list":74,"\u6d4b\u8bd5save_dir":74,"\u6d4f\u89c8\u4ee3\u7801":54,"\u6d6a\u6f2b\u7247":90,"\u6d6e\u70b9\u578b\u6570\u636e":21,"\u6d6e\u70b9\u5f02\u5e38\u901a\u5e38\u7684\u539f\u56e0\u662f\u6d6e\u70b9\u6570\u6ea2\u51fa":49,"\u6d6e\u70b9\u6570\u5360\u7528\u7684\u5b57\u8282\u6570":85,"\u6d6e\u70b9\u7a00\u758f\u6570\u636e":67,"\u6dd8\u5b9d\u7b49":93,"\u6df1\u5ea6\u53cc\u5411lstm\u5c42\u63d0\u53d6softmax\u5c42\u7684\u7279\u5f81":92,"\u6df1\u5ea6\u5b66\u4e60\u5165\u95e8\u8bfe\u7a0b":57,"\u6df7\u5408":92,"\u6df7\u5408\u5f53\u524d\u8bcd\u5411\u91cf\u548cattention\u52a0\u6743\u7f16\u7801\u5411\u91cf":64,"\u6dfb\u52a0\u4e86\u4e00\u4e2a\u8f93\u51fa":68,"\u6dfb\u52a0\u542f\u52a8\u811a\u672c":80,"\u6dfb\u52a0\u8f93\u5165\u53c2\u6570":68,"\u6dfb\u52a0\u8f93\u51fa\u53c2\u6570":68,"\u6dfb\u52a0op\u7684\u6ce8\u91ca":68,"\u6e05\u7406":65,"\u6e05\u7406\u6389\u8001\u65e7\u7684paddlepaddle\u5b89\u88c5\u5305":46,"\u6e29\u99a8":61,"\u6e90":94,"\u6e90\u4ee3\u7801":[54,89],"\u6e90\u4ee3\u7801\u683c\u5f0f":66,"\u6e90\u5b57\u5178":94,"\u6e90\u5e8f\u5217":64,"\u6e90\u7801":54,"\u6e90\u7801\u4e0edemo":79,"\u6e90\u7801\u6811\u6839\u76ee\u5f55":65,"\u6e90\u7801\u7f16\u8bd1paddlepaddl":59,"\u6e90\u8bed\u8a00\u5230\u76ee\u6807\u8bed\u8a00\u7684\u5e73\u884c\u8bed\u6599\u5e93\u6587\u4ef6":94,"\u6e90\u8bed\u8a00\u548c\u76ee\u6807\u8bed\u8a00\u5171\u4eab\u76f8\u540c\u7684\u7f16\u7801\u5b57\u5178":85,"\u6e90\u8bed\u8a00\u548c\u76ee\u6807\u8bed\u8a00\u90fd\u662f\u76f8\u540c\u7684\u8bed\u8a00":85,"\u6e90\u8bed\u8a00\u77ed\u8bed\u548c\u76ee\u6807\u8bed\u8a00\u77ed\u8bed\u7684\u5b57\u5178\u5c06\u88ab\u5408\u5e76":85,"\u6ee4\u6ce2\u5668\u6838\u5728\u5782\u76f4\u65b9\u5411\u4e0a\u7684\u5c3a\u5bf8":87,"\u6ee4\u6ce2\u5668\u6838\u5728\u6c34\u5e73\u65b9\u5411\u4e0a\u7684\u5c3a\u5bf8":87,"\u6f14\u793a\u4e2d\u4f7f\u7528\u7684":92,"\u6f14\u793a\u91c7\u7528":92,"\u6fc0\u6d3b":67,"\u6fc0\u6d3b\u51fd\u6570\u7c7b\u578b":89,"\u6fc0\u6d3b\u65b9\u7a0b":67,"\u6fc0\u6d3b\u7684\u7c7b\u578b":67,"\u70ed\u60c5":61,"\u7136\u540e":[72,73,85,91],"\u7136\u540e\u4ea4\u7ed9\u7528\u6237\u81ea\u5b9a\u4e49\u7684\u51fd\u6570":52,"\u7136\u540e\u4ea4\u7ed9step\u51fd\u6570":63,"\u7136\u540e\u4f60\u53ea\u9700\u8981\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4":94,"\u7136\u540e\u4f7f\u7528\u4e0b\u9762\u7684\u811a\u672c":93,"\u7136\u540e\u4f7f\u7528resize\u63a5\u53e3\u8bbe\u7f6etensor\u7684\u5927\u5c0f":69,"\u7136\u540e\u518d\u505a\u4e00\u6b21\u6587\u672c\u5377\u79ef\u7f51\u7edc\u64cd\u4f5c":91,"\u7136\u540e\u5229\u7528\u89c2\u6d4b\u6570\u636e\u8c03\u6574":52,"\u7136\u540e\u5355\u51fb":66,"\u7136\u540e\u53ef\u4ee5\u4f7f\u7528\u547d\u4ee4\u884c\u5de5\u5177\u521b\u5efajob":80,"\u7136\u540e\u53ef\u4ee5\u8f6c\u6362\u4e3a\u56fe\u7247":87,"\u7136\u540e\u5728":94,"\u7136\u540e\u5728\u4e0b\u4e00\u4e2a\u65f6\u95f4\u6b65\u8f93\u5165\u7ed9\u53e6\u4e00\u4e2a\u795e\u7ecf\u5143":61,"\u7136\u540e\u5728\u6d4f\u89c8\u5668\u4e2d\u8f93\u5165\u4ee5\u4e0b\u7f51\u5740":54,"\u7136\u540e\u5728\u89e3\u7801\u88ab\u7ffb\u8bd1\u7684\u8bed\u53e5\u65f6":94,"\u7136\u540e\u5728dataprovider\u91cc\u9762\u6839\u636e\u8be5\u5730\u5740\u52a0\u8f7d\u5b57\u5178":51,"\u7136\u540e\u5728etcd\u7684":20,"\u7136\u540e\u5b89\u88c5paddle\u7684python\u73af\u5883":46,"\u7136\u540e\u5b9a\u4e49":64,"\u7136\u540e\u5c06\u6784\u5efa\u6210\u529f\u7684\u955c\u50cf\u4e0a\u4f20\u5230\u955c\u50cf\u4ed3\u5e93":80,"\u7136\u540e\u5c06\u8fd9\u4e9blayer\u7684\u53c2\u6570":50,"\u7136\u540e\u5c31\u53ef\u4ee5\u5e76\u53d1\u5199\u5165\u591a\u4e2achunk":28,"\u7136\u540e\u5f97\u5230\u5e73\u5747\u91c7\u6837\u7684\u7ed3\u679c":91,"\u7136\u540e\u6211\u4eec\u5229\u7528\u591a\u8f93\u5165\u7684":91,"\u7136\u540e\u6211\u4eec\u53d1\u73b0pass":94,"\u7136\u540e\u6211\u4eec\u5806\u53e0\u4e00\u5bf9\u5bf9\u7684":92,"\u7136\u540e\u6211\u4eec\u6c42\u8fd9\u4e24\u4e2a\u7279\u5f81\u7684\u4f59\u5f26\u76f8\u4f3c\u5ea6":91,"\u7136\u540e\u6240\u6709\u7528":66,"\u7136\u540e\u624d\u80fd\u4f7f\u7528pfsclient":28,"\u7136\u540e\u6267\u884c\u4e0b\u9762\u7684\u547d\u4ee4":87,"\u7136\u540e\u628a\u8fd9\u4e2a\u5305\u542b\u4e86\u8bad\u7ec3\u6570\u636e\u7684container\u4fdd\u5b58\u4e3a\u4e00\u4e2a\u65b0\u7684\u955c\u50cf":79,"\u7136\u540e\u63d0\u4ea4\u65b0\u6dfb\u52a0\u7684":66,"\u7136\u540e\u63d0\u53d6\u9690\u85cflstm\u5c42\u7684\u6240\u6709\u65f6\u95f4\u6b65\u957f\u7684\u6700\u5927\u8bcd\u5411\u91cf\u4f5c\u4e3a\u6574\u4e2a\u5e8f\u5217\u7684\u8868\u793a":93,"\u7136\u540e\u662f\u5bf9\u5e94\u7684\u82f1\u8bed\u5e8f\u5217":94,"\u7136\u540e\u6dfb\u52a0\u4e00\u4e2a\u5168\u8fde\u63a5\u5c42":91,"\u7136\u540e\u70b9\u51fb":66,"\u7136\u540e\u7528\u5bc6\u7801":54,"\u7136\u540e\u7528pickle\u547d\u4ee4\u5c06\u7279\u5f81":91,"\u7136\u540e\u7533\u660e\u4e00\u4e2a\u5b58\u50a8\u5377":80,"\u7136\u540e\u89c2\u5bdf\u5230\u8f93\u51fa\u7684\u53d8\u5316\u4e3a":67,"\u7136\u540e\u89e3\u538b":94,"\u7136\u540e\u89e3\u7801\u5668\u901a\u8fc7\u8fd9\u4e2a\u5411\u91cf\u751f\u6210\u4e00\u4e2a\u76ee\u6807\u8bed\u53e5":94,"\u7136\u540e\u8f93\u51fa\u9884\u6d4b\u5206\u6570":91,"\u7136\u540e\u8fd4\u56de\u7ed9paddlepaddle\u8fdb\u7a0b":2,"\u7136\u540e\u8fdb\u884c\u968f\u673a\u6253\u4e71":91,"\u7136\u540e\u901a\u8fc7\u51fd\u6570":80,"\u7136\u540e\u901a\u8fc7\u81ea\u8eab\u7684ip\u5730\u5740\u5728":80,"\u7136\u540e\u91cd\u65b0cmake\u5373\u53ef":46,"\u7136\u800c":[64,75],"\u7136\u800c\u6709\u4e9b\u8bc4\u8bba\u4e0a\u4e0b\u6587\u975e\u5e38\u957f":93,"\u7248\u672c":65,"\u7248\u672c\u5206\u652f":42,"\u7248\u672c\u53f7":42,"\u7248\u672c\u53f7rc":42,"\u7248\u672c\u5728":66,"\u7248\u672cfork\u51fa\u81ea\u5df1\u7684\u529f\u80fd\u5206\u652f":42,"\u7279\u522b\u662f\u5728lstm\u7b49rnn\u4e2d":49,"\u7279\u522b\u662f\u5f53\u76f8\u540c\u7684\u8bcd\u5728\u53e5\u5b50\u4e2d\u51fa\u73b0\u591a\u4e8e\u4e00\u6b21\u65f6":92,"\u7279\u5f81":91,"\u7279\u5f81\u56fe\u5747\u503c":87,"\u7279\u5f81\u56fe\u65b9\u5dee":87,"\u7279\u5f81\u5c06\u4f1a\u5b58\u5230":87,"\u7279\u5f81\u6587\u4ef6":92,"\u7279\u5f81\u7684\u7c7b\u578b\u548c\u7ef4\u5ea6":91,"\u72af\u7f6a\u7247":90,"\u72ec\u7acb\u5de5\u5177\u94fe":58,"\u72ec\u7acb\u5de5\u5177\u94fe\u6240\u5728\u7684\u7edd\u5bf9\u8def\u5f84":58,"\u73af\u5883\u53d8\u91cf":80,"\u73af\u5883\u53d8\u91cf\u6765\u6307\u5b9a\u7279\u5b9a\u7684gpu":49,"\u73b0\u5728\u6211\u4eec\u53ef\u4ee5\u5f00\u59cbpaddle\u8bad\u7ec3\u4e86":91,"\u73b0\u9636\u6bb5paddle\u6709\u4e00\u4e2a\u95ee\u9898\u662f":33,"\u7406\u89e3":65,"\u751a\u81f3\u4e0d\u540c\u7ade\u4e89\u5bf9\u624b\u4ea7\u54c1\u7684\u504f\u597d":93,"\u751a\u81f3\u80fd\u89e3\u91ca\u4e3a\u4ec0\u4e48\u67d0\u4e2a\u64cd\u4f5c\u82b1\u4e86\u5f88\u957f\u65f6\u95f4":72,"\u751f\u4ea7\u73af\u5883\u4e2d\u7684\u8bad\u7ec3\u6570\u636e\u96c6\u901a\u5e38\u4f53\u79ef\u5f88\u5927":21,"\u751f\u4ea7\u73af\u5883\u7684\u65e5\u5fd7\u6570\u636e\u4f1a\u901a\u8fc7\u5b9e\u65f6\u6d41\u7684\u65b9\u5f0f":21,"\u751f\u4ea7\u955c\u50cf":[54,66],"\u751f\u4ea7\u955c\u50cf\u7684\u751f\u6210\u5206\u4e3a\u4e24\u6b65":54,"\u751f\u6210":80,"\u751f\u6210\u5404\u79cd\u8bed\u8a00\u7684\u7ed1\u5b9a\u4ee3\u7801":33,"\u751f\u6210\u540e\u7684\u6587\u6863\u5206\u522b\u5b58\u50a8\u5728\u7f16\u8bd1\u76ee\u5f55\u7684":70,"\u751f\u6210\u5e8f\u5217\u7684\u6700\u5927\u957f\u5ea6":64,"\u751f\u6210\u5f53\u524d\u5c42\u7684\u6240\u6709\u540e\u7ee7\u72b6\u6001":94,"\u751f\u6210\u6570\u636e\u7684\u76ee\u5f55":94,"\u751f\u6210\u6587\u6863":33,"\u751f\u6210\u751f\u4ea7\u955c\u50cf\u53ea\u9700\u8981\u8fd0\u884c":54,"\u751f\u6210\u7684":21,"\u751f\u6210\u7684\u6570\u636e\u5c06\u4f1a\u5b58\u50a8\u5728\u8fd9\u4e2avolume\u4e0b":80,"\u751f\u6210\u7684\u6570\u636e\u7f13\u5b58\u5728\u5185\u5b58\u91cc":49,"\u751f\u6210\u7684\u7ed3\u679c\u6587\u4ef6":94,"\u751f\u6210\u7684html\u7248\u672c\u7684c":54,"\u751f\u6210\u7684meta\u914d\u7f6e\u6587\u4ef6\u5982\u4e0b\u6240\u793a":91,"\u751f\u6210\u7ed9\u5b9a":21,"\u751f\u6210\u7f51\u7edc\u5c42\u914d\u7f6e":67,"\u751f\u6210\u81ea\u5df1\u76ee\u5f55\u4e0b\u7684\u4ed3\u5e93":66,"\u751f\u6210\u8bad\u7ec3\u9700\u8981\u7684\u6837\u672c":91,"\u751f\u6210\u8fd0\u884c\u7a0b\u5e8f":54,"\u751f\u6210\u968f\u673a\u7684\u8f93\u5165\u6570\u636e":68,"\u751f\u6210api\u6587\u6863":33,"\u751f\u6210docker\u955c\u50cf\u7684\u65b9\u5f0f\u6709\u4e24\u4e2a":54,"\u751f\u6210pfsclient\u548cpfsserver\u7684\u6846\u67b6\u90e8\u5206":28,"\u7528":[28,90,91,92],"\u75280\u548c1\u8868\u793a":2,"\u7528\u4e86\u4e24\u4e2a\u6708\u4e4b\u540e\u8fd9\u4e2a\u663e\u793a\u5668\u5c4f\u5e55\u788e\u4e86":89,"\u7528\u4e8e":73,"\u7528\u4e8e\u5207\u5206\u5355\u5355\u8bcd\u548c\u6807\u70b9\u7b26\u53f7":93,"\u7528\u4e8e\u521d\u59cb\u5316\u53c2\u6570\u548c\u8bbe\u7f6e":67,"\u7528\u4e8e\u5c06\u4e0b\u4e00\u884c\u7684\u6570\u636e\u8f93\u5165\u51fd\u6570\u6807\u8bb0\u6210\u4e00\u4e2apydataprovider2":2,"\u7528\u4e8e\u5c06\u53c2\u6570\u4f20\u9012\u7ed9\u7f51\u7edc\u914d\u7f6e":77,"\u7528\u4e8e\u6307\u5b9a\u5176\u8981\u5173\u8054\u7684layer":50,"\u7528\u4e8e\u6307\u5b9a\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6":75,"\u7528\u4e8e\u653e\u7f6e":73,"\u7528\u4e8e\u6784\u6210\u65b0\u7684\u8bcd\u8868":85,"\u7528\u4e8e\u6ce8\u518c\u6ca1\u6709\u53cd\u5411\u7684op":68,"\u7528\u4e8e\u7a00\u758f\u8bad\u7ec3\u4e2d":75,"\u7528\u4e8e\u7edf\u8ba1\u8bcd\u9891\u7684bow\u6a21\u578b\u7279\u5f81":93,"\u7528\u4e8e\u81ea\u5b9a\u4e49\u6bcf\u6761\u6570\u636e\u7684batch":2,"\u7528\u4e8e\u83b7\u53d6\u7279\u5b9alayer\u4e0a\u4e00\u65f6\u95f4\u6b65\u7684\u8f93\u51fa":50,"\u7528\u4e8e\u8ba1\u7b97\u7f16\u7801\u5411\u91cf\u7684\u52a0\u6743\u548c":64,"\u7528\u4e8e\u8bbe\u7f6e\u8bad\u7ec3\u7b97\u6cd5":86,"\u7528\u4e8e\u8bfb\u53d6\u8bad\u7ec3":73,"\u7528\u4e8e\u96c6\u7fa4\u901a\u4fe1\u901a\u9053\u7684\u7aef\u53e3\u6570":73,"\u7528\u53cc\u5411\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7f16\u7801":64,"\u7528\u591a\u5bf9\u6548\u679c\u5b8c\u5168\u76f8\u540c\u7684":61,"\u7528\u6237":73,"\u7528\u62371\u7684\u7279\u5f81":91,"\u7528\u6237\u4e0a\u4f20\u6570\u636e\u540e":21,"\u7528\u6237\u4e5f\u53ef\u4ee5\u4e0a\u4f20label":21,"\u7528\u6237\u4e5f\u53ef\u4ee5\u5728\u81ea\u5df1\u719f\u6089\u7684\u5f00\u53d1\u5e73\u53f0\u4e0a":59,"\u7528\u6237\u4e5f\u53ef\u4ee5\u5728c":1,"\u7528\u6237\u53ea\u9700\u5b9a\u4e49rnn\u5728\u4e00\u4e2a\u65f6\u95f4\u6b65\u5185\u5b8c\u6210\u7684\u8ba1\u7b97":63,"\u7528\u6237\u53ea\u9700\u6267\u884c":92,"\u7528\u6237\u53ea\u9700\u6267\u884c\u4ee5\u4e0b\u547d\u4ee4\u5c31\u53ef\u4ee5\u4e0b\u8f7d\u5e76\u5904\u7406\u539f\u59cb\u6570\u636e":92,"\u7528\u6237\u53ef\u4ee5\u4f7f\u7528\u5f00\u53d1\u955c\u50cf\u4ee3\u66ff\u914d\u7f6e\u672c\u5730\u73af\u5883":54,"\u7528\u6237\u53ef\u4ee5\u4f7f\u7528ssh\u767b\u5f55\u5230\u8fd9\u53f0\u670d\u52a1\u5668\u4e0a\u5e76\u6267\u884c":54,"\u7528\u6237\u53ef\u4ee5\u5206\u522b\u67e5\u770b\u6700\u65b0\u7684":70,"\u7528\u6237\u53ef\u4ee5\u53c2\u8003sphinx\u6559\u7a0b\u8fdb\u884c\u4e66\u5199":70,"\u7528\u6237\u53ef\u4ee5\u5728\u8f93\u51fa\u7684\u6587\u672c\u6a21\u578b\u4e2d\u770b\u5230":85,"\u7528\u6237\u53ef\u4ee5\u5b89\u5168\u7684\u91ca\u653e\u67d0\u4e2ac":34,"\u7528\u6237\u53ef\u4ee5\u628a\u81ea\u5df1\u7684\u6570\u636e\u5206\u4eab\u7ed9\u522b\u4eba":21,"\u7528\u6237\u53ef\u4ee5\u6839\u636e\u8bad\u7ec3\u65e5\u5fd7":89,"\u7528\u6237\u53ef\u4ee5\u76f4\u63a5\u4f7f\u7528\u8fd9\u4e2a\u52a8\u6001\u5e93\u6765\u5f15\u5165paddl":34,"\u7528\u6237\u53ef\u4ee5\u81ea\u5b9a\u4e49beam":75,"\u7528\u6237\u53ef\u4ee5\u8bbe\u7f6e":77,"\u7528\u6237\u53ef\u4ee5\u901a\u8fc7\u7b80\u5355\u4f7f\u7528python\u63a5\u53e3":1,"\u7528\u6237\u53ef\u4ee5\u901a\u8fc7\u7f51\u9875\u6d4f\u89c8\u6587\u6863":54,"\u7528\u6237\u53ef\u5728\u8c03\u7528cmake\u7684\u65f6\u5019\u8bbe\u7f6e\u5b83\u4eec":53,"\u7528\u6237\u53ef\u5728cmake\u7684\u547d\u4ee4\u884c\u4e2d":53,"\u7528\u6237\u53ef\u5c06":58,"\u7528\u6237\u53ef\u6839\u636e\u81ea\u5df1\u7684\u7f16\u8bd1\u76ee\u6807\u67b6\u6784":58,"\u7528\u6237\u53ef\u81ea\u884c\u524d\u5f80":59,"\u7528\u6237\u53ef\u81ea\u884c\u524d\u5f80\u4e0b\u8f7d\u9884\u7f16\u8bd1\u597d\u7684\u7248\u672c":58,"\u7528\u6237\u53ef\u901a\u8fc7\u4ea4\u53c9\u7f16\u8bd1\u7684\u65b9\u5f0f":58,"\u7528\u6237\u53ef\u901a\u8fc7ssh\u7b49\u65b9\u5f0f\u767b\u5f55\u5230raspberri":59,"\u7528\u6237\u5728\u4f7f\u7528\u8fd9\u4e00\u7c7brecurr":50,"\u7528\u6237\u5728\u4f7f\u7528paddlepaddl":46,"\u7528\u6237\u5728\u672c\u5730\u8f6c\u6362\u597d\u518d\u4e0a\u4f20":21,"\u7528\u6237\u5b9a\u4e49\u7684\u53c2\u6570":2,"\u7528\u6237\u5c06\u53c2\u6570\u8f7d\u5165":51,"\u7528\u6237\u5c06\u914d\u7f6e\u4e0e\u8bad\u7ec3\u6570\u636e\u5207\u5206\u597d\u653e\u5728\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u9884\u5148\u5206\u914d\u597d\u7684\u76ee\u5f55\u4e2d":80,"\u7528\u6237\u5e94\u8be5\u63d0\u4f9b\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6":92,"\u7528\u6237\u5f3a\u5236\u6307\u5b9a\u7279\u5b9a\u7684python\u7248\u672c":46,"\u7528\u6237\u6307\u5b9a\u65b0\u7684\u5b57\u5178\u7684\u8def\u5f84":85,"\u7528\u6237\u6587\u4ef6\u4e2d\u6709\u56db\u79cd\u7c7b\u578b\u7684\u5b57\u6bb5":91,"\u7528\u6237\u6587\u4ef6\u53ef\u80fd\u662f\u6bd4\u8f83\u5927\u7684":28,"\u7528\u6237\u7279\u5f81":91,"\u7528\u6237\u8fd8\u53ef\u4ee5\u6839\u636e\u6982\u7387\u5206\u5e03\u77e9\u9635\u5b9e\u73b0\u67f1\u641c\u7d22\u6216\u7ef4\u7279\u6bd4\u89e3\u7801":92,"\u7528\u6237\u8fd8\u53ef\u6839\u636e\u81ea\u5df1\u7684\u9700\u6c42\u8bbe\u7f6e\u5176\u4ed6\u7f16\u8bd1\u53c2\u6570":[58,59],"\u7528\u6237\u901a\u8fc7\u53c2\u6570":[50,51],"\u7528\u6237\u901a\u8fc7c":34,"\u7528\u6237\u9700\u8981\u5728\u7f51\u7edc\u914d\u7f6e\u4e2d\u6307\u5b9a":77,"\u7528\u6237\u9700\u8981\u5728cmake\u65f6\u624b\u52a8\u8bbe\u7f6e\u8fd9\u4e9b\u503c":[58,59],"\u7528\u6237\u9700\u8981\u6307\u5b9a\u672c\u673a\u4e0apython\u7684\u8def\u5f84":46,"\u7528\u6237\u9700\u8981\u63d0\u524d\u51c6\u5907\u597d\u4ea4\u53c9\u7f16\u8bd1\u73af\u5883":[58,59],"\u7528\u6237\u9884\u6d4b\u7684\u547d\u4ee4\u884c\u754c\u9762\u5982\u4e0b":91,"\u7528\u6237id":90,"\u7528\u6237id\u8303\u56f4\u4ece1\u52306040":90,"\u7528\u672c\u673a\u7684\u7b2c\u4e00\u4e2a":65,"\u7528\u6765\u4ece\u53c2\u6570\u670d\u52a1\u5668\u9884\u53d6\u53c2\u6570\u77e9\u9635\u76f8\u5e94\u7684\u884c":67,"\u7528\u6765\u4f30\u8ba1\u7ebf\u6027\u51fd\u6570\u7684\u53c2\u6570w":52,"\u7528\u6765\u5177\u4f53\u63cf\u8ff0":91,"\u7528\u6765\u5177\u4f53\u8bf4\u660e\u6570\u636e\u96c6\u7684\u5b57\u6bb5\u548c\u6587\u4ef6\u683c\u5f0f":91,"\u7528\u6765\u8ba1\u7b97\u6a21\u578b\u7684\u8bef\u5dee":52,"\u7528\u8fd9\u4e2a\u955c\u50cf\u521b\u5efa\u7684\u5bb9\u5668\u9700\u8981\u6709\u4ee5\u4e0b\u4e24\u4e2a\u529f\u80fd":80,"\u7531":[50,63],"\u7531\u4e8e\u4e0a\u4e00\u6b65cmake\u914d\u7f6e\u4e2d":59,"\u7531\u4e8e\u4e0d\u540c\u7684paddle\u7684\u7248\u672c\u53ef\u80fd\u9700\u8981\u4e0d\u540c\u7684\u4f9d\u8d56\u548c\u5de5\u5177":54,"\u7531\u4e8e\u5b83\u5185\u90e8\u5305\u542b\u4e86\u6bcf\u7ec4\u6570\u636e\u4e2d\u7684\u6240\u6709\u53e5\u5b50":61,"\u7531\u4e8e\u5bb9\u5668\u4e4b\u95f4\u5171\u4eabnet":78,"\u7531\u4e8e\u5bf9parameters\u7684\u66f4\u65b0\u9700\u8981\u83b7\u53d6parameters\u5185\u5b58\u7684":20,"\u7531\u4e8e\u5df2\u7ecf\u77e5\u9053\u4e86\u771f\u5b9e\u7b54\u6848":52,"\u7531\u4e8e\u610f\u5916\u7684\u526f\u672c\u8bb0\u5f55\u548c\u6d4b\u8bd5\u8bb0\u5f55":90,"\u7531\u4e8e\u6211\u4eec\u60f3\u8981\u7684\u53d8\u6362\u662f\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217":61,"\u7531\u4e8e\u6211\u4eec\u652f\u6301\u8bad\u7ec3\u6570\u636e\u6709\u4e0d\u540c\u7684\u6279\u6b21\u5927\u5c0f":67,"\u7531\u4e8e\u6570\u636e\u8bb8\u53ef\u7684\u539f\u56e0":92,"\u7531\u4e8e\u6807\u51c6\u7684\u7ffb\u8bd1\u7ed3\u679c\u5df2\u7ecf\u4e0b\u8f7d\u5230\u8fd9\u91cc":94,"\u7531\u4e8e\u6bcf\u4e2a\u5377\u79ef\u5c42\u540e\u9762\u8fde\u63a5\u7684\u662fbatch":87,"\u7531\u4e8e\u8fd9\u4e2a\u5730\u5740\u4f1a\u88abdataprovider\u4f7f\u7528":1,"\u7531\u4e8e\u8fd9\u6837\u505a\u53ef\u4ee5\u907f\u514d\u5f88\u591a\u6b7b\u9501\u95ee\u9898":2,"\u7531\u4e8e\u96c6\u7fa4\u4e2d\u540c\u65f6\u5b58\u5728\u4e24\u53f0\u673a\u5668\u6545\u969c\u7684\u6982\u7387\u6781\u4f4e":20,"\u7531\u4e8e\u987a\u5e8f\u8c03\u7528\u8fd9\u4e9bgenerator\u4e0d\u4f1a\u51fa\u73b0\u4e0a\u8ff0\u95ee\u9898":2,"\u7531\u4e8ec":33,"\u7531\u4e8echunk\u6bd4\u8f83\u5c0f":28,"\u7531\u4e8eeigen":69,"\u7531\u4e8epaddlepaddle\u5df2\u7ecf\u5b9e\u73b0\u4e86\u4e30\u5bcc\u7684\u7f51\u7edc\u5c42":52,"\u7531\u4e8epypi":42,"\u7531\u4e8estep":63,"\u7531\u4e8etensor\u7684rank\u662f\u6a21\u677f\u53c2\u6570":69,"\u7531\u4e8etest_data\u5305\u542b\u4e24\u6761\u9884\u6d4b\u6570\u636e":4,"\u7531\u8bcd\u8bed\u6784\u6210\u7684\u53e5\u5b50":60,"\u7531grouplen":90,"\u7533\u8bf7\u7528\u6237\u7a7a\u95f4":28,"\u7535\u5f711\u7684\u7279\u5f81":91,"\u7535\u5f71\u4fe1\u606f\u4ee5\u53ca\u7535\u5f71\u8bc4\u5206":90,"\u7535\u5f71\u540d\u5b57\u6bb5":91,"\u7535\u5f71\u540d\u79f0":90,"\u7535\u5f71\u548c\u7528\u6237":91,"\u7535\u5f71\u548c\u7528\u6237\u6709\u8bb8\u591a\u7684\u7279\u5f81":91,"\u7535\u5f71\u5927\u90e8\u5206\u662f\u624b\u5de5\u8f93\u5165\u6570\u636e":90,"\u7535\u5f71\u7279\u5f81":91,"\u7535\u5f71\u7c7b\u578b":90,"\u7535\u5f71\u7c7b\u578b\u5982\u7b26\u5408\u591a\u79cd\u7528\u7ba1\u9053\u7b26\u53f7":90,"\u7535\u5f71id":90,"\u7535\u5f71id\u8303\u56f4\u4ece1\u52303952":90,"\u7535\u8111":61,"\u767e\u4e07\u6570\u636e\u96c6":90,"\u7684":[61,65,66,69,73,79,80,89,93],"\u768410\u7ef4\u6574\u6570\u6807\u7b7e":2,"\u7684\u4e00\u4e2a\u7b80\u5355\u8c03\u7528\u5982\u4e0b":63,"\u7684\u4e00\u4e2a\u7ebf\u6027\u51fd\u6570":52,"\u7684\u4e00\u79cd":94,"\u7684\u4e3a0":75,"\u7684\u4e3b\u8981\u90e8\u5206":92,"\u7684\u4efb\u4e00\u4e00\u79cd":49,"\u7684\u4efb\u52a1":94,"\u7684\u4f5c\u7528\u662f\u5ef6\u8fdf\u5206\u914d\u5185\u5b58":69,"\u7684\u4f7f\u7528\u793a\u4f8b\u5982\u4e0b":60,"\u7684\u503c":[58,59],"\u7684\u503c\u81ea\u52a8\u63a8\u5bfc\u5f97\u5230":58,"\u7684\u504f\u7f6e\u5411\u91cf":67,"\u7684\u5177\u4f53\u8ba1\u7b97\u903b\u8f91":68,"\u7684\u5185\u5b58":49,"\u7684\u5185\u5bb9\u5982\u4e0b\u6240\u793a":94,"\u7684\u5185\u5bb9\u6765\u5b9a\u5236imag":80,"\u7684\u5185\u6838block\u4f7f\u7528\u60c5\u51b5":72,"\u7684\u5206\u7c7b\u4efb\u52a1\u4e2d\u8d62\u5f97\u4e86\u7b2c\u4e00\u540d":87,"\u7684\u522b\u540d":[5,6,8],"\u7684\u5355\u5143\u6d4b\u8bd5":68,"\u7684\u5355\u8bcd\u7ea7\u522b\u7684\u5377\u79ef\u795e\u7ecf\u7f51\u7edc":91,"\u7684\u53cd\u5411\u4f20\u64ad\u5c06\u4f1a\u6253\u5370\u65e5\u5fd7\u4fe1\u606f":75,"\u7684\u53d8\u6362\u77e9\u9635":67,"\u7684\u53e5\u5b50\u7684\u60c5\u611f":93,"\u7684\u540d\u5b57":2,"\u7684\u540d\u79f0\u76f8\u540c":64,"\u7684\u540e\u7f00":90,"\u7684\u5411\u91cf":67,"\u7684\u542f\u52a8\u53c2\u6570":80,"\u7684\u542f\u52a8\u53c2\u6570\u5e76\u6267\u884c\u8fdb\u7a0b":80,"\u7684\u547d\u540d\u98ce\u683c\u5e76\u4e0d\u80fd\u9002\u5e94\u5176\u4ed6\u7b2c\u4e09\u65b9\u8bed\u8a00":33,"\u7684\u5730\u5740":78,"\u7684\u5730\u65b9":66,"\u7684\u5747\u5300\u5206\u5e03":51,"\u7684\u5934\u6587\u4ef6":33,"\u7684\u5b9e\u73b0":68,"\u7684\u5dee\u8ddd\u4e0d\u65ad\u51cf\u5c0f":52,"\u7684\u5e73\u5747\u503c":60,"\u7684\u5e8f\u5217\u5f62\u72b6\u4e00\u81f4":61,"\u7684\u5f00\u53d1\u5de5\u4f5c\u90fd\u5e94\u8be5\u5728\u4e00\u4e2a\u65b0\u7684\u5206\u652f\u4e0a\u5b8c\u6210":66,"\u7684\u5f00\u53d1\u6d41\u7a0b":65,"\u7684\u5f00\u59cb\u8bf7\u52a0\u4e0a\u5b8f\u5b9a\u4e49":68,"\u7684\u603b":73,"\u7684\u63a5\u53e3\u6837\u5f0f":33,"\u7684\u63cf\u8ff0\u8bf4\u660e\u4e2d":66,"\u7684\u64cd\u4f5c":69,"\u7684\u6570\u636e\u6d41\u56fe":21,"\u7684\u6570\u636e\u8bfb\u53d6\u811a\u672c\u548c\u7c7b\u4f3c\u4e8e":89,"\u7684\u6570\u76ee\u4e00\u81f4":60,"\u7684\u6587\u4ef6\u4e5f\u5e26\u5230\u65b0\u5206\u652f\u4e0a":66,"\u7684\u65b9\u5f0f":54,"\u7684\u65b9\u6cd5\u5df2\u88ab\u8bc1\u660e\u662f\u4e00\u79cd\u5f3a\u5927\u7684\u6a21\u578b":94,"\u7684\u65b9\u7a0b":67,"\u7684\u65f6\u95f4\u6b65\u4fe1\u606f\u6210\u6b63\u6bd4":49,"\u7684\u66f4\u8be6\u7ec6\u51c6\u786e\u7684\u5b9a\u4e49":61,"\u7684\u6700\u5c0f\u503c":75,"\u7684\u6700\u65b0\u4ee3\u7801\u5e76\u66f4\u65b0\u5f53\u524d\u5206\u652f":66,"\u7684\u6784\u9020\u51fd\u6570":68,"\u7684\u67b6\u6784\u7684\u793a\u4f8b":64,"\u7684\u6837\u5f0f":66,"\u7684\u6838\u5fc3\u662f\u8bbe\u8ba1step\u51fd\u6570\u7684\u8ba1\u7b97\u903b\u8f91":63,"\u7684\u6bb5\u843d\u5b9a\u4e49\u4e3a\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":63,"\u7684\u6d4b\u8bd5\u6570\u636e\u96c6":92,"\u7684\u6e90\u7801\u4ee5\u53ca\u751f\u6210\u6587\u6863\u9700\u8981\u591a\u79cd\u5f00\u53d1\u5de5\u5177":66,"\u7684\u6e90\u7801\u91cc\u4f7f\u7528\u4e86":33,"\u7684\u7248\u672c":65,"\u7684\u7248\u672c\u53f7":85,"\u7684\u7279\u5f81":87,"\u7684\u72b6\u6001":63,"\u7684\u7528\u6237\u53c2\u8003":73,"\u7684\u76ee\u5f55\u7ed3\u6784\u4e3a":91,"\u7684\u771f\u5b9e\u5173\u7cfb\u4e3a":52,"\u7684\u77e9\u9635":[49,67],"\u7684\u795e\u7ecf\u7f51\u7edc\u673a\u5668\u7ffb\u8bd1":94,"\u7684\u7a20\u5bc6\u5411\u91cf\u4f5c\u4e3a\u8f93\u5165":67,"\u7684\u7aef\u5230\u7aef\u7cfb\u7edf\u6765\u89e3\u51b3srl\u4efb\u52a1":92,"\u7684\u7b2ci\u4e2a\u503c":67,"\u7684\u7b2cj\u4e2a\u503c":67,"\u7684\u7cfb\u7edf":65,"\u7684\u7d22\u5f15\u6587\u4ef6\u5f15\u7528\u8bad\u7ec3":73,"\u7684\u7ed3\u6784\u5982\u4e0b":91,"\u7684\u7ef4\u5ea6":85,"\u7684\u7f29\u5199":28,"\u7684\u884c\u6570\u5e94\u8be5\u4e00\u81f4":94,"\u7684\u89c4\u8303":33,"\u7684\u89d2\u5ea6":21,"\u7684\u8ba1\u7b97\u4ee3\u7801":69,"\u7684\u8ba1\u7b97\u8fc7\u7a0b\u4e66\u5199\u66f4\u52a0\u7b80\u5355":68,"\u7684\u8bad\u7ec3\u6a21\u578b\u811a\u672c":89,"\u7684\u8bdd":49,"\u7684\u8def\u5f84\u4e2d":93,"\u7684\u8f93\u5165":63,"\u7684\u8f93\u51fa":72,"\u7684\u8f93\u51fa\u4fe1\u606f\u5165\u624b\u662f\u4e2a\u4e0d\u9519\u7684\u9009\u62e9":72,"\u7684\u8f93\u51fa\u51fd\u6570\u8fd4\u56de\u7684\u662f\u4e0b\u4e00\u4e2a\u65f6\u523b\u8f93\u51fa\u8bcd\u7684":64,"\u7684\u8f93\u51fa\u683c\u5f0f":61,"\u7684\u8f93\u51fa\u88ab\u7528\u4f5c":64,"\u7684\u8f93\u51fab\u662f\u4e00\u4e2a":49,"\u7684\u8fd0\u884c\u73af\u5883":65,"\u7684\u8fd4\u56de\u503c\u4e00\u81f4":90,"\u7684\u8fdc\u7a0b\u4ed3\u5e93\u7684\u540d\u5b57":66,"\u7684\u90e8\u5206":73,"\u7684\u914d\u7f6e":85,"\u7684\u914d\u7f6e\u5199\u5230\u914d\u7f6e\u6587\u4ef6\u4e2d":21,"\u7684\u96c6\u88c5\u7bb1\u6280\u672f":65,"\u7684\u9875\u9762\u5220\u9664\u8fdc\u7a0b\u4ed3\u5e93\u7684\u5206\u652f":66,"\u7684cpu":68,"\u7684docker\u955c\u50cf":54,"\u76ee\u524d":[63,92],"\u76ee\u524d\u4f7f\u7528":66,"\u76ee\u524d\u53ea\u652f\u6301":58,"\u76ee\u524d\u53ea\u8003\u8651\u52a8\u6001\u6269\u5bb9trainer\u6570\u91cf":20,"\u76ee\u524d\u5d4c\u5165python\u89e3\u91ca\u5668":33,"\u76ee\u524d\u5fc5\u987b\u8bbe\u7f6e\u6210":[58,59],"\u76ee\u524d\u6211\u4eec\u7528cephfs\u6765\u642d\u5efa":28,"\u76ee\u524d\u652f\u6301\u4e24\u79cd":60,"\u76ee\u524d\u652f\u6301fail":75,"\u76ee\u524d\u8be5\u53c2\u6570\u4ec5\u7528\u4e8eaucvalidationlayer\u548cpnpairvalidationlayer\u5c42":75,"\u76ee\u524d\u8fd8\u672a\u652f\u6301":63,"\u76ee\u524dpaddle\u7684\u8fdb\u7a0b\u6a21\u578b\u662fc":33,"\u76ee\u524dpaddlepaddle\u7684develop\u5206\u652f\u7684\u6587\u6863\u662f\u81ea\u52a8\u89e6\u53d1\u66f4\u65b0\u7684":70,"\u76ee\u5f55":[58,59,65,73,79,80,93],"\u76ee\u5f55\u4e0b":[34,54,67,89,94],"\u76ee\u5f55\u4e0b\u627e\u5230":89,"\u76ee\u5f55\u4e0b\u65b0\u589e\u7684":68,"\u76ee\u5f55\u4e0b\u7684\u751f\u6210\u6587\u4ef6\u7528\u4e8e\u6df1\u5ea6\u5b66\u4e60\u76f8\u5173android":58,"\u76ee\u5f55\u4e0b\u7684demo\u8bad\u7ec3\u51fa\u6765":4,"\u76ee\u5f55\u4e0b\u7684python\u5305":46,"\u76ee\u5f55\u4e2d":[73,91],"\u76ee\u5f55\u4e2d\u4f1a\u5305\u542b":[58,59],"\u76ee\u5f55\u4e2d\u7684":[72,73],"\u76ee\u5f55\u4e2dpaddl":80,"\u76ee\u5f55\u4f1a\u51fa\u73b0\u5982\u4e0b\u51e0\u4e2a\u65b0\u7684\u6587\u4ef6":92,"\u76ee\u5f55\u548c":[58,59],"\u76ee\u5f55\u5c31\u6210\u4e3a\u4e86\u5171\u4eab\u5b58\u50a8":80,"\u76ee\u5f55\u7ed3\u6784\u5982\u4e0b":94,"\u76ee\u5f55\u91cc\u63d0\u4f9b\u4e86\u8be5\u6570\u636e\u7684\u4e0b\u8f7d\u811a\u672c\u548c\u9884\u5904\u7406\u811a\u672c":89,"\u76ee\u6807":94,"\u76ee\u6807\u51fd\u6570\u662f\u6807\u7b7e\u7684\u4ea4\u53c9\u71b5":92,"\u76ee\u6807\u5411\u91cf":64,"\u76ee\u6807\u5b57\u5178":94,"\u76ee\u6807\u673a\u7248protobuf\u5e93":[58,59],"\u76ee\u6807\u67b6\u6784abi":58,"\u76f4\u5230\u8bad\u7ec3\u6536\u655b\u4e3a\u6b62":51,"\u76f4\u5230\u903c\u8fd1\u771f\u5b9e\u89e3":52,"\u76f4\u63a5\u4f7f\u7528c\u8bed\u8a00\u7684":33,"\u76f4\u63a5\u5220\u9664\u8fd9\u4e2a\u53c2\u6570\u5373\u53ef":34,"\u76f4\u63a5\u5bfc\u51fa\u5230c\u7684\u63a5\u53e3\u6bd4\u8f83\u56f0\u96be":33,"\u76f4\u63a5\u7f16\u8bd1raspberri":59,"\u76f4\u63a5\u8fd0\u884c":54,"\u76f4\u63a5\u8fd4\u56de\u8ba1\u7b97\u7ed3\u679c":4,"\u76f4\u63a5\u8fdb\u5165\u8bad\u7ec3\u6a21\u578b\u7ae0\u8282":89,"\u76f8\u5173\u6587\u6863":59,"\u76f8\u5173\u6587\u6863\u6240\u8ff0":59,"\u76f8\u5173\u6982\u5ff5\u662f":2,"\u76f8\u5173\u7684\u9e1f\u7c7b\u6570\u636e\u96c6\u53ef\u4ee5\u4ece\u5982\u4e0b\u5730\u5740\u4e0b\u8f7d":86,"\u76f8\u5173\u8bba\u6587":92,"\u76f8\u53cd":94,"\u76f8\u540c\u540d\u5b57\u7684\u53c2\u6570":51,"\u76f8\u5bf9":61,"\u76f8\u5bf9\u4e8epaddlepaddle\u7a0b\u5e8f\u8fd0\u884c\u65f6\u7684\u8def\u5f84":1,"\u76f8\u5bf9mnist\u800c\u8a00":2,"\u76f8\u5e94\u7684\u6570\u636e\u8bfb\u53d6\u811a\u672c\u548c\u8bad\u7ec3\u6a21\u578b\u811a\u672c":89,"\u76f8\u5e94\u7684\u6570\u636e\u8fed\u4ee3\u5668\u5982\u4e0b":92,"\u76f8\u5e94\u7684\u6807\u8bb0\u53e5\u5b50\u662f":92,"\u76f8\u5f53":61,"\u76f8\u6bd4":68,"\u770b\u5f53\u524dmpi\u96c6\u7fa4\u662f\u5426\u652f\u6301resourc":47,"\u77e9\u9635":74,"\u77e9\u9635\u4e58\u6cd5\u7684\u516c\u5f0f":68,"\u7814\u7a76\u4eba\u5458\u5206\u6790\u4e86\u51e0\u4e2a\u5173\u4e8e\u6d88\u8d39\u8005\u4fe1\u5fc3\u548c\u653f\u6cbb\u89c2\u70b9\u7684\u8c03\u67e5":93,"\u7814\u7a76\u751f":90,"\u786e\u4fdd\u7f16\u8bd1\u5668\u9009\u9879":66,"\u78c1\u76d8\u4e0d\u591f":65,"\u78c1\u76d8\u7a7a\u95f4\u4e0d\u8db3\u7b49":47,"\u793a":89,"\u793a\u4f8b":[49,51,87],"\u793a\u4f8b3\u5bf9\u4e8e\u5355\u5c42rnn\u548c\u53cc\u5c42rnn\u6570\u636e\u5b8c\u5168\u76f8\u540c":61,"\u793a\u4f8b3\u7684\u914d\u7f6e\u4f7f\u7528\u4e86\u5355\u5c42rnn\u548c\u53cc\u5c42rnn":61,"\u793a\u4f8b3\u7684\u914d\u7f6e\u5206\u522b\u4e3a":61,"\u793a\u4f8b\u4ee3\u7801\u5982\u4e0b":49,"\u793a\u4f8b\u5982\u4e0b":51,"\u793e\u533a\u53c2\u4e0e\u56f0\u96be":33,"\u793e\u533a\u8d21\u732e\u4ee3\u7801\u5b66\u4e60\u6210\u672c\u9ad8":33,"\u795e\u7ecf\u7f51\u7edc\u4e2d\u7684\u53c2\u6570":20,"\u795e\u7ecf\u7f51\u7edc\u4e5f\u9700\u8981\u4e00\u4e9b\u7279\u5b9a\u7684layer\u4f5c\u4e3a\u8f93\u5165\u63a5\u53e3":56,"\u795e\u7ecf\u7f51\u7edc\u53c2\u6570\u4ee5\u53ca\u8fed\u4ee3\u65b9\u7a0b":56,"\u795e\u7ecf\u7f51\u7edc\u5728\u8bad\u7ec3\u7684\u65f6\u5019":49,"\u795e\u7ecf\u7f51\u7edc\u672c\u8d28\u4e0a\u662f\u4e00\u4e2a\u8ba1\u7b97\u56fe":69,"\u795e\u7ecf\u7f51\u7edc\u673a\u5668\u7ffb\u8bd1":94,"\u795e\u7ecf\u7f51\u7edc\u7684\u7f51\u7edc\u7ed3\u6784\u4e2d\u5177\u6709\u6709\u5411\u73af\u7ed3\u6784":61,"\u795e\u7ecf\u7f51\u7edc\u7684\u8bad\u7ec3\u672c\u8eab\u662f\u4e00\u4e2a\u975e\u5e38\u6d88\u8017\u5185\u5b58\u548c\u663e\u5b58\u7684\u5de5\u4f5c":49,"\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e":52,"\u79bb":61,"\u79bb\u7ebf\u6279\u5904\u7406":21,"\u79d1\u5b66\u5bb6":90,"\u79d1\u5e7b\u7247":90,"\u79f0\u4e3a":[64,66],"\u79f0\u4e3a\u5f00\u53d1\u955c\u50cf":54,"\u79f0\u4e3a\u751f\u4ea7\u955c\u50cf":54,"\u79f0\u4e4b\u4e3a\u53cc\u5c42\u5e8f\u5217\u7684\u4e00\u4e2a\u5b50\u5e8f\u5217":60,"\u79f0\u4e4b\u4e3a\u96c6\u675f\u5927\u5c0f":75,"\u79f0\u4f5c\u6709kernel":68,"\u79f0\u4f5ckernel":68,"\u7a00\u758f\u6570\u636e\u7684\u683c\u5f0f":67,"\u7a00\u758f\u66f4\u65b0\u7684\u7aef\u53e3\u6570\u91cf":80,"\u7a00\u758f\u768401\u5411\u91cf":[2,56],"\u7a00\u758f\u7684\u5411\u91cf":[2,56],"\u7a00\u758f\u77e9\u9635\u7684\u4e58\u79ef\u5e94\u7528\u4e8e\u524d\u5411\u4f20\u64ad\u8fc7\u7a0b":77,"\u7a0b\u5e8f\u4ece\u6b64\u76ee\u5f55\u62f7\u8d1d\u6587\u4ef6\u5230\u5bb9\u5668\u5185\u8fdb\u884c\u8bad\u7ec3":80,"\u7a0b\u5e8f\u505c\u6b62":75,"\u7a0b\u5e8f\u5458":90,"\u7a0b\u5e8f\u76f4\u63a5\u9000\u51fa":75,"\u7a20\u5bc6\u5411\u91cf":67,"\u7a20\u5bc6\u66f4\u65b0\u7684\u7aef\u53e3\u6570\u91cf":80,"\u7a20\u5bc6\u7684\u6d6e\u70b9\u6570\u5411\u91cf":[2,56],"\u7a97\u6237":61,"\u7acb\u523b\u9000\u51fa":49,"\u7aef\u53e3":[47,73],"\u7aef\u53e3\u6570":73,"\u7aef\u53e3\u9644\u52a0\u5230\u4e3b\u673a\u540d\u4e0a":73,"\u7aef\u81ea\u5b9a\u4e49\u4e00\u4e2a":1,"\u7aef\u8bfb\u53d6\u6570\u636e":49,"\u7b2c":61,"\u7b2c\u4e00\u4e2a":66,"\u7b2c\u4e00\u4e2a\u53c2\u6570":68,"\u7b2c\u4e00\u4e2a\u53c2\u6570\u662fsettings\u5bf9\u8c61":2,"\u7b2c\u4e00\u4e2a\u65b9\u6cd5\u7684\u597d\u5904\u662f\u7b80\u5355\u5feb\u6377":54,"\u7b2c\u4e00\u4e2a\u6837\u672c\u540c\u65f6encode\u4e24\u6761\u6570\u636e\u6210\u4e24\u4e2a\u5411\u91cf":61,"\u7b2c\u4e00\u4e2apass\u4f1a\u4ecepython\u7aef\u8bfb\u53d6\u6570\u636e":2,"\u7b2c\u4e00\u4e2atag\u4e3a":42,"\u7b2c\u4e00\u5929":61,"\u7b2c\u4e00\u6b65\u662f\u8fd0\u884c":54,"\u7b2c\u4e00\u7ae0\u8282":56,"\u7b2c\u4e00\u884c\u4ece":93,"\u7b2c\u4e00\u884c\u5b58\u7684\u662f\u56fe\u50cf":87,"\u7b2c\u4e00\u884c\u662f":85,"\u7b2c\u4e00\u884c\u7684":94,"\u7b2c\u4e00\u90e8\u5206\u5b9a\u4e49\u4e86\u6570\u636e\u8f93\u5165":52,"\u7b2c\u4e00\u90e8\u5206\u662f\u56fe\u7247\u7684\u6807\u7b7e":2,"\u7b2c\u4e09":93,"\u7b2c\u4e09\u4e2a\u53c2\u6570":68,"\u7b2c\u4e09\u5217\u662f\u751f\u6210\u7684\u82f1\u8bed\u5e8f\u5217":94,"\u7b2c\u4e09\u6b65":87,"\u7b2c\u4e09\u6b65\u5b8c\u6210\u540e":42,"\u7b2c\u4e8c":93,"\u7b2c\u4e8c\u4e2a":49,"\u7b2c\u4e8c\u4e2a\u4e3a":42,"\u7b2c\u4e8c\u4e2a\u53c2\u6570":68,"\u7b2c\u4e8c\u4e2a\u65b9\u6cd5\u7684\u597d\u5904\u662fdockerfile\u53ef\u4ee5\u628a\u6574\u4e2a\u751f\u6210\u6d41\u7a0b\u63cf\u8ff0\u5f88\u6e05\u695a":54,"\u7b2c\u4e8c\u5217\u662f\u96c6\u675f\u641c\u7d22\u7684\u5f97\u5206":94,"\u7b2c\u4e8c\u6b65":[85,87],"\u7b2c\u4e8c\u6b65\u662f\u8fd0\u884c":54,"\u7b2c\u4e8c\u7c7b":50,"\u7b2c\u4e8c\u884c\u5b58\u7684\u662f\u56fe\u50cf":87,"\u7b2c\u4e8c\u90e8\u5206\u4e3b\u8981\u662f\u9009\u62e9\u5b66\u4e60\u7b97\u6cd5":52,"\u7b2c\u4e8c\u90e8\u5206\u662f28":2,"\u7b2c\u56db\u4e2a\u53c2\u6570":68,"\u7b2ci\u884c\u7b2cj\u5217\u7684\u6570\u503c":67,"\u7b49":[34,47,68],"\u7b49\u4e8e\u6837\u672c\u6570":49,"\u7b49\u5168\u90e8\u9759\u6001\u5e93\u4e2d\u7684\u76ee\u6807\u6587\u4ef6\u5168\u90e8\u6253\u5305\u540e\u4ea7\u751f\u7684\u6587\u4ef6":34,"\u7b49\u5176\u4ed6":94,"\u7b49\u53c2\u6570":80,"\u7b49\u591a\u79cd\u516c\u6709\u4e91\u73af\u5883":78,"\u7b49\u5f85\u8fd9\u4e2a\u7a0b\u5e8f\u6267\u884c\u6210\u529f\u5e76\u8fd4\u56de0\u5219\u6210\u529f\u9000\u51fa":78,"\u7b49\u6587\u4ef6":34,"\u7b49\u7b49":[89,94],"\u7b49\u90fd\u5c5e\u4e8e\u4e00\u4e2a\u547d\u540d\u7a7a\u95f4":78,"\u7b80\u4ecb":71,"\u7b80\u5199":68,"\u7b80\u5355\u4ecb\u7ecd\u9700\u8981\u7528\u5230\u57fa\u7c7b":68,"\u7b80\u5355\u603b\u7ed3op\u9700\u8981\u5305\u542b\u7684\u5185\u5bb9\u5982\u4e0b":68,"\u7b80\u5355\u6765\u8bf4":72,"\u7b80\u5355\u7684\u5168\u8fde\u63a5\u7f51\u7edc":51,"\u7b80\u5355\u7684\u57fa\u4e8e\u5b57\u6bcd\u7684\u8bcd\u5d4c\u5165":91,"\u7b80\u5355\u7684\u6027\u80fd\u5206\u6790":72,"\u7b80\u5355\u7684\u6574\u4e2a\u8bcd\u5d4c\u5165":91,"\u7b80\u5355\u7684pydataprovider2\u6837\u4f8b\u5c31\u8bf4\u660e\u5b8c\u6bd5\u4e86":2,"\u7b80\u5355\u7684yaml\u6587\u4ef6\u5982\u4e0b":79,"\u7b80\u76f4":61,"\u7b97\u6cd5":[49,52,64,93],"\u7b97\u6cd5\u4e2d\u7684beam\u5927\u5c0f":64,"\u7b97\u6cd5\u914d\u7f6e":93,"\u7ba1\u7406\u4eba\u5458":90,"\u7ba1\u7406\u5458":90,"\u7c7b\u4f3c":[34,60],"\u7c7b\u4f3c\u5730":92,"\u7c7b\u4f5c\u4e3a\u53c2\u6570\u7684\u62bd\u8c61":67,"\u7c7b\u522b\u4e2a\u6570":86,"\u7c7b\u522b\u4e2d\u7684\u53c2\u6570\u53ef\u7528\u4e8e\u6240\u6709\u573a\u5408":74,"\u7c7b\u522bid":89,"\u7c7b\u522bid\u548c\u6587\u672c\u4fe1\u606f\u7528":89,"\u7c7b\u540d\u548cc":33,"\u7c7b\u578b":[33,68,75,91],"\u7c7b\u578b\u4e3a":68,"\u7c7b\u578b\u4ecd\u7136\u4e3aeigenvector":69,"\u7c7b\u578b\u53ef\u4ee5\u662fpaddlepaddle\u652f\u6301\u7684\u4efb\u610f\u8f93\u5165\u6570\u636e\u7c7b\u578b":60,"\u7c7b\u578b\u540d\u4e3a":68,"\u7c7b\u578b\u662fnumpy\u7684ndarrai":49,"\u7c7b\u578b\u662fsparse_binary_vector":[2,56],"\u7c7b\u578b\u662fsparse_float_vector":[2,56],"\u7c7b\u578b\u6765\u8bbe\u7f6e":2,"\u7c7b\u578b\u7684":61,"\u7c7b\u578b\u7b49\u662f\u5426\u5408\u6cd5":68,"\u7c7b\u7684\u5b9a\u4e49\u5199\u5728":68,"\u7c7b\u7684\u6784\u9020\u51fd\u6570\u548c\u6790\u6784\u51fd\u6570":67,"\u7c7b\u91cd\u5199":68,"\u7c7b\u9700\u8981\u5b9e\u73b0\u521d\u59cb\u5316":67,"\u7cfb\u6570":68,"\u7cfb\u7edf\u4f1a\u5bf9\u65b0\u589e\u7684op\u81ea\u52a8\u7ed1\u5b9apython":68,"\u7cfb\u7edf\u4f1a\u6839\u636e\u6587\u4ef6\u540d\u81ea\u52a8\u6784\u5efaop\u548c\u5176\u5bf9\u5e94\u7684python\u6269\u5c55":68,"\u7ea2\u697c\u68a6":85,"\u7eaa\u5f55\u7247":90,"\u7eafcpu\u955c\u50cf\u4ee5\u53cagpu\u955c\u50cf\u90fd\u4f1a\u7528\u5230avx\u6307\u4ee4\u96c6":54,"\u7ebf\u6027\u56de\u5f52\u7684\u8f93\u5165\u662f\u4e00\u6279\u70b9":52,"\u7ebf\u6027\u56de\u5f52\u7684\u8f93\u51fa\u662f\u4ece\u8fd9\u6279\u70b9\u4f30\u8ba1\u51fa\u6765\u7684\u53c2\u6570":52,"\u7ebf\u6027\u8ba1\u7b97\u7f51\u7edc\u5c42":52,"\u7ebf\u7a0bid\u53f7":77,"\u7ec6\u8282\u63cf\u8ff0":76,"\u7ecf\u5e38\u4f1a\u6d88\u8017\u657010gb\u7684\u5185\u5b58\u548c\u6570gb\u7684\u663e\u5b58":49,"\u7ed3\u5408":78,"\u7ed3\u5c3e":68,"\u7ed3\u675f\u6807\u8bb0":64,"\u7ed3\u6784\u5982\u4e0b":93,"\u7ed3\u6784\u5982\u4e0b\u56fe":85,"\u7ed3\u679c\u4fdd\u5b58\u5728":92,"\u7ed3\u679c\u53d1\u73b0\u5b83\u4eec\u4e0e\u540c\u65f6\u671f\u7684twitter\u6d88\u606f\u4e2d\u7684\u60c5\u7eea\u8bcd\u9891\u7387\u76f8\u5173":93,"\u7ed3\u679c\u8f93\u51fa\u5230":65,"\u7ed3\u8bba":33,"\u7ed9":61,"\u7ed9\u4e2a\u7b80\u5355\u7684":66,"\u7ed9\u51fa\u56fe\u7247\u5c3a\u5bf8":86,"\u7ed9\u51fa\u8f93\u5165\u6570\u636e\u6240\u5728\u8def\u5f84":86,"\u7ed9\u5b9a\u52a8\u8bcd":92,"\u7ed9\u5b9a\u7684\u6587\u672c\u53ef\u4ee5\u662f\u4e00\u4e2a\u6587\u6863":93,"\u7ed9\u5b9aencoder\u8f93\u51fa\u548c\u5f53\u524d\u8bcd":63,"\u7edf\u4e00\u7528":21,"\u7ee7\u627f\u81ea":68,"\u7ee7\u627f\u81eaoperatorbas":68,"\u7ee7\u7eed\u8bad\u7ec3\u6216\u9884\u6d4b":2,"\u7ef4\u57fa\u767e\u79d1\u4e2d\u6587\u9875\u9762":61,"\u7ef4\u57fa\u767e\u79d1\u9875\u9762":61,"\u7ef4\u5ea6\u4e3aword_dim":89,"\u7ef4\u5ea6\u662f\u7c7b\u522b\u4e2a\u6570":89,"\u7ef4\u5ea6\u662f\u8bcd\u5178\u5927\u5c0f":89,"\u7ef4\u62a4":78,"\u7ef4\u7a7a\u95f4":64,"\u7ef4\u7a7a\u95f4\u5b8c\u6210":64,"\u7f13\u5b58\u6c60\u7684\u51cf\u5c0f":49,"\u7f13\u5b58\u8bad\u7ec3\u6570\u636e\u5230\u5185\u5b58":2,"\u7f16\u5199\u597d\u6570\u636e\u63d0\u4f9b\u811a\u672c\u540e":91,"\u7f16\u5199\u5b8cyaml\u6587\u4ef6\u540e":80,"\u7f16\u5199\u672c\u6b21\u8bad\u7ec3\u7684yaml\u6587\u4ef6":80,"\u7f16\u53f7":91,"\u7f16\u53f7\u4ece0\u5f00\u59cb":49,"\u7f16\u53f7\u5b57\u6bb5":91,"\u7f16\u7801\u5411\u91cf":64,"\u7f16\u7801\u5668\u8f93\u51fa":64,"\u7f16\u7801\u6e90\u5e8f\u5217":64,"\u7f16\u89e3\u7801\u6a21\u578b\u5c06\u4e00\u4e2a\u6e90\u8bed\u53e5\u7f16\u7801\u4e3a\u4e00\u4e2a\u5b9a\u957f\u7684\u5411\u91cf":94,"\u7f16\u8bd1":[54,65,66],"\u7f16\u8bd1\u548c\u5b89\u88c5paddlepaddl":59,"\u7f16\u8bd1\u548c\u5b89\u88c5paddlepaddle\u9884\u6d4b\u5e93":58,"\u7f16\u8bd1\u548c\u6267\u884c":68,"\u7f16\u8bd1\u5668":[58,59],"\u7f16\u8bd1\u5668\u6ca1\u6709":33,"\u7f16\u8bd1\u578b\u8bed\u8a00":33,"\u7f16\u8bd1\u5b89\u88c5\u4e0e\u5355\u5143\u6d4b\u8bd5":48,"\u7f16\u8bd1\u5b8c\u6210\u4e4b\u540e":70,"\u7f16\u8bd1\u5de5\u5177\u94fe\u6240\u5728\u7684\u7edd\u5bf9\u8def\u5f84":59,"\u7f16\u8bd1\u6210\u529f\u540e":68,"\u7f16\u8bd1\u6210\u52a8\u6001\u5e93":75,"\u7f16\u8bd1\u6d41\u7a0b\u4e3b\u8981\u63a8\u8350\u9ad8\u7ea7\u7528\u6237\u67e5\u770b":55,"\u7f16\u8bd1\u751f\u6210":70,"\u7f16\u8bd1\u8fd9\u4e2a\u7248\u672c\u7684docker\u53d1\u884c\u955c\u50cf":42,"\u7f16\u8bd1\u8fd9\u4e2a\u7248\u672c\u7684python":42,"\u7f16\u8bd1\u8fd9\u4e2a\u7248\u672c\u7684ubuntu":42,"\u7f16\u8bd1\u9009\u9879":53,"\u7f16\u8bd1c":34,"\u7f16\u8bd1master\u5206\u652f\u7684docker\u53d1\u884c\u955c\u50cf":42,"\u7f16\u8bd1paddlepaddle\u548c\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5":71,"\u7f16\u8bd1ubuntu\u7684deb\u5305":42,"\u7f16\u8f91":78,"\u7f29\u653e\u53c2\u6570":87,"\u7f51\u7edc":[92,93],"\u7f51\u7edc\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf":78,"\u7f51\u7edc\u540d\u79f0":89,"\u7f51\u7edc\u5c42\u53ef\u4ee5\u6709\u591a\u4e2a\u8f93\u5165":67,"\u7f51\u7edc\u5c42\u7684\u6807\u8bc6\u7b26\u4e3a":67,"\u7f51\u7edc\u5c42\u7684\u7c7b\u578b":67,"\u7f51\u7edc\u5c42\u7684\u7ec6\u8282\u53ef\u4ee5\u901a\u8fc7\u4e0b\u9762\u8fd9\u4e9b\u4ee3\u7801\u7247\u6bb5\u6765\u6307\u5b9a":67,"\u7f51\u7edc\u5c42\u7684\u8f93\u51fa\u662f\u7ecf\u8fc7\u6fc0\u6d3b\u51fd\u6570\u4e4b\u540e\u7684\u503c":75,"\u7f51\u7edc\u5c42\u914d\u7f6e\u5305\u542b\u4ee5\u4e0b\u51e0\u9879":67,"\u7f51\u7edc\u63a5\u53e3\u5361":73,"\u7f51\u7edc\u6a21\u5757":87,"\u7f51\u7edc\u6a21\u578b\u5c06\u8f93\u51fa\u6807\u7b7e\u7684\u6982\u7387\u5206\u5e03":92,"\u7f51\u7edc\u7684\u8bad\u7ec3\u8fc7\u7a0b":93,"\u7f51\u7edc\u7ed3\u6784\u5982\u4e0b\u56fe\u6240\u793a":91,"\u7f51\u7edc\u901a\u4fe1":67,"\u7f51\u7edc\u914d\u7f6e":[73,89,93],"\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6":[87,92],"\u800c":[50,52,64,68,79,91],"\u800c\u4e0d\u4f1a\u6539\u53d8\u539f\u6709tensor\u7684shape\u4fe1\u606f":69,"\u800c\u4e0d\u4f7f\u7528imdb\u6570\u6910\u96c6\u4e2d\u7684imdb":93,"\u800c\u4e0d\u5fc5\u5728\u610fpaddl":34,"\u800c\u4e0d\u652f\u6301pypy\u89e3\u91ca\u5668":33,"\u800c\u4e0d\u662f\u4f7f\u7528\u540c\u6b65":73,"\u800c\u4e0d\u662f\u5728layer\u91cc\u5b9e\u73b0":50,"\u800c\u4e0d\u662f\u6e90\u7801\u76ee\u5f55\u91cc":46,"\u800c\u4e0d\u662f\u7279\u5f81\u7684\u96c6\u5408":61,"\u800c\u4e0d\u662f\u7ec4\u5408\u4e0a\u4e0b\u6587\u7ea7\u522b\u4fe1\u606f":93,"\u800c\u4e0d\u66b4\u9732\u6982\u5ff5\u7684\u5b9e\u73b0":34,"\u800c\u4e0d\u7528\u5173\u5fc3\u6570\u636e\u5982\u4f55\u4f20\u8f93":2,"\u800c\u4e14":94,"\u800c\u4e14\u5728\u4f20\u8f93\u7684\u8fc7\u7a0b\u4e2d\u4e5f\u53ef\u80fd\u51fa\u73b0\u7f51\u7edc\u4e0d\u7a33\u5b9a\u7684\u60c5\u51b5":28,"\u800c\u4e4b\u524d\u7684\u53c2\u6570\u5c06\u4f1a\u88ab\u5220\u9664":75,"\u800c\u4ece\u5e94\u7528\u7684\u89d2\u5ea6":72,"\u800c\u4f18\u5316\u6027\u80fd\u7684\u9996\u8981\u4efb\u52a1":72,"\u800c\u5176\u4ed6\u5c42\u4f7f\u7528cpu\u8ba1\u7b97":77,"\u800c\u51fa\u73b0\u9636\u6bb5\u6027\u7684\u8fd0\u884c\u505c\u6ede":20,"\u800c\u53cc\u5c42rnn\u662f\u53ef\u4ee5\u5904\u7406\u8fd9\u79cd\u8f93\u5165\u6570\u636e\u7684\u7f51\u7edc\u7ed3\u6784":61,"\u800c\u53ea\u9700\u8981\u83b7\u5f97recurr":50,"\u800c\u53f3\u56fe\u7684\u74f6\u9888\u8fde\u63a5\u6a21\u5757\u7528\u4e8e50\u5c42":87,"\u800c\u5728\u8ba1\u7b97\u7ed3\u675f\u4e4b\u540e":69,"\u800c\u5728cpp\u91cc\u9762\u5b9e\u73b0\u8fd9\u4e2ac\u7684\u63a5\u53e3":33,"\u800c\u591a\u8bed\u8a00\u63a5\u53e3\u9700\u8981\u76f4\u63a5\u8bfb\u53d6\u751f\u6210\u7684\u4e8c\u8fdb\u5236":33,"\u800c\u5927\u591a\u6570\u65b9\u6cd5\u53ea\u662f\u5229\u7528n":93,"\u800c\u5b89\u88c5\u5305":46,"\u800c\u5b89\u88c5\u5305\u662f":46,"\u800c\u5bf9\u4e8e\u53cc\u5c42\u5e8f\u5217":61,"\u800c\u5bf9\u4e8e\u6bcf\u4e00\u4e2a\u5185\u5c42\u7279\u5f81\u6570\u636e\u800c\u8a00":61,"\u800c\u5bf9\u4e8egolang":33,"\u800c\u5bf9\u4e8egolang\u9519\u8bef\u5904\u7406\u5e94\u8be5\u4f7f\u7528\u8fd4\u56de\u503c":33,"\u800c\u5c06\u8fd9\u4e2a\u6bb5\u843d\u7684\u6bcf\u4e00\u53e5\u8bdd\u7528lstm\u7f16\u7801\u6210\u4e00\u4e2a\u5411\u91cf":61,"\u800c\u5f53\u524d\u5df2\u7ecf\u67095":72,"\u800c\u662f\u76f4\u63a5\u4ece\u5185\u5b58\u7684\u7f13\u5b58\u91cc\u8bfb\u53d6\u6570\u636e":49,"\u800c\u662f\u76f4\u63a5\u4fee\u6539paddl":34,"\u800c\u662f\u76f4\u63a5\u7528api\u7684\u63a5\u53e3\u8fdc\u7a0b\u8bbf\u95ee":21,"\u800c\u66f4\u6df1\u5165\u7684\u5206\u6790":72,"\u800c\u6709\u4e9b\u53c2\u6570\u9700\u8981\u5728\u96c6\u7fa4\u591a\u673a\u8bad\u7ec3\u4e2d\u4f7f\u7528\u7b49":74,"\u800c\u6ca1\u6709\u77ed\u65f6\u8bb0\u5fc6\u7684\u635f\u5931":93,"\u800c\u6e90\u5e8f\u5217\u7684\u7f16\u7801\u5411\u91cf\u53ef\u4ee5\u88ab\u65e0\u8fb9\u754c\u7684memory\u8bbf\u95ee":64,"\u800c\u795e\u7ecf\u7f51\u7edc\u662f\u6211\u4eec\u8981\u642d\u5efa\u7684\u5b9d\u5854":56,"\u800c\u7a00\u758f\u66f4\u65b0\u5728\u53cd\u5411\u4f20\u64ad\u4e4b\u540e\u7684\u6743\u91cd\u66f4\u65b0\u65f6\u8fdb\u884c":77,"\u800c\u8ba1\u7b97\u8fc7\u7a0b\u662f\u7531":69,"\u800c\u8fd9\u4e00\u53e5\u8bdd\u5c31\u53ef\u4ee5\u8868\u793a\u6210\u8fd9\u4e9b\u4f4d\u7f6e\u7684\u6570\u7ec4":61,"\u800c\u8fd9\u4e2acontext\u53ef\u80fd\u4f1a\u975e\u5e38\u5927":2,"\u800c\u8fd9\u6bcf\u4e00\u4e2a\u6570\u7ec4\u5143\u7d20":61,"\u800c\u975e\u76f4\u63a5\u56de\u590d\u7684\u65b9\u5f0f":66,"\u800c\u975e\u9759\u6001\u52a0\u8f7dcuda\u52a8\u6001\u5e93":53,"\u800ceigenvector":69,"\u800cpaddlepaddle\u5219\u4f1a\u5e2e\u7528\u6237\u505a\u4ee5\u4e0b\u5de5\u4f5c":2,"\u800crnn\u662f\u6700\u6d41\u884c\u7684\u9009\u62e9":63,"\u800cswig\u53ea\u80fd\u7b80\u5355\u7684\u66b4\u9732c":33,"\u800ctrainer\u9700\u8981\u8bfb\u53d6\u8bad\u7ec3\u6570\u636e\u8fdb\u884c\u8bad\u7ec3":56,"\u800cweight":86,"\u800cy_predict\u662f\u63a5\u6536x\u4f5c\u4e3a\u8f93\u5165":56,"\u804c\u4e1a":90,"\u804c\u4e1a\u4ece\u4e0b\u9762\u6240\u5217\u4e2d\u9009\u62e9":90,"\u80fd\u591f\u5904\u7406\u53cc\u5c42\u5e8f\u5217":63,"\u80fd\u591f\u5bf9\u53cc\u5411\u5e8f\u5217\u8fdb\u884c\u5904\u7406\u7684\u6709":63,"\u80fd\u591f\u627e\u5230\u8fd9\u91cc\u4f7f\u7528\u7684\u6240\u6709\u6570\u636e":89,"\u80fd\u591f\u8bb0\u5f55\u4e0a\u4e00\u4e2asubseq":63,"\u80fd\u591f\u9488\u5bf9cpu\u548cgpu\u7684\u8ba1\u7b97\u505a\u66f4\u591a\u4f18\u5316":50,"\u80fd\u83b7\u53d6":73,"\u811a\u672c":[65,73,86,91],"\u811a\u672c\u4fdd\u5b58\u5728":86,"\u811a\u672c\u5f00\u59cb\u65f6":80,"\u811a\u672c\u63d0\u4f9b\u4e86\u4e00\u4e2a\u9884\u6d4b\u63a5\u53e3":93,"\u811a\u672c\u65f6\u9700\u8981\u52a0\u4e0a":93,"\u811a\u672c\u8fd0\u884c\u5b8c\u6210\u540e":86,"\u81ea\u52a8\u5173\u95ed\u5bf9\u5e94\u7684":66,"\u81ea\u52a8\u5730\u5c06\u8fd9\u4e9b\u9009\u9879\u5e94\u7528\u5230":73,"\u81ea\u52a8\u5b8c\u6210\u8fd9\u4e00\u8fc7\u7a0b":63,"\u81ea\u52a8\u6302\u8f7d\u5206\u5e03\u5f0f\u5b58\u50a8\u76ee\u5f55":20,"\u81ea\u52a8\u751f\u6210":70,"\u81ea\u52a8\u83b7\u53d6\u4e0a\u4e00\u4e2a\u751f\u6210\u7684\u8bcd":64,"\u81ea\u5e95\u5411\u4e0a\u6cd5":89,"\u81ea\u6b64":58,"\u81ea\u7136\u4e5f\u5c31\u6709\u7ba1\u7406\u5458\u6743\u9650":65,"\u81ea\u7136\u8bed\u8a00\u5904\u7406\u7b49":77,"\u81ea\u7531\u804c\u4e1a\u8005":90,"\u81f3\u4e8e\u4e3a\u4ec0\u4e48\u9700\u8981c":34,"\u81f3\u6b64":[2,61,66],"\u8212\u9002":61,"\u826f\u597d\u7684\u6587\u6863":33,"\u827a\u672f\u5bb6":90,"\u8282\u70b9\u4e2d\u7684":73,"\u82e5":67,"\u82e5\u5728paddlepaddle\u7f16\u8bd1\u65f6":51,"\u82e5\u5e0c\u671b\u6700\u5feb\u7684\u6267\u884c\u901f\u5ea6":[58,59],"\u82e5\u5e72\u4e2a\u53e5\u5b50\u6784\u6210\u4e00\u4e2a\u6bb5\u843d":60,"\u82e5\u6709\u4e0d\u4e00\u81f4\u4e4b\u5904":72,"\u82e5\u6709\u5fc5\u8981":67,"\u82e5\u6ca1\u6709\u663e\u5f0f\u8bbe\u7f6e":58,"\u82e5\u73af\u5883\u53d8\u91cf":[58,59],"\u82e5\u8981\u5bf9\u8fd9\u51e0\u4e2alayer\u4f7f\u7528dropout":50,"\u82e5\u8f93\u51fa\u662f\u5355\u5c42\u5e8f\u5217":60,"\u82e5\u8f93\u51fa\u662f\u53cc\u5c42\u5e8f\u5217":60,"\u82f1\u6587\u6587\u6863":70,"\u82f1\u6587\u6587\u6863\u76ee\u5f55":70,"\u82f1\u8bed":94,"\u8303\u56f4":77,"\u83b7\u53d6":66,"\u83b7\u53d6\u5229\u7528":89,"\u83b7\u53d6\u5b57\u5178\u7ef4\u5ea6":93,"\u83b7\u53d6\u5f53\u524d\u7cfb\u7edf\u652f\u6301\u7684python\u5305\u7684\u540e\u7f00":46,"\u83b7\u53d6\u6700\u65b0\u7684\u68c0\u67e5\u70b9\u7684\u6587\u4ef6uuid":20,"\u83b7\u53d6\u6e90\u7801":65,"\u83b7\u53d6\u8be5\u6761\u6837\u672c\u7c7b\u522bid":89,"\u83b7\u53d6\u901a\u8fc7":93,"\u83b7\u53d6trainer":80,"\u83b7\u5f97\u53c2\u6570\u5c3a\u5bf8":67,"\u865a\u62df\u673a\u4e0a":65,"\u867d\u7136":52,"\u867d\u7136\u4e0d\u9f13\u52b1\u8fd9\u6837":34,"\u867d\u7136\u6bcf\u4e2agenerator\u5728\u6ca1\u6709\u8c03\u7528\u7684\u65f6\u5019":2,"\u867d\u7136\u8fd9\u4e9b\u6587\u4ef6\u5e76\u975e\u90fd\u9700\u8981\u96c6\u7fa4\u8bad\u7ec3":73,"\u867d\u7136paddle\u770b\u8d77\u6765\u5305\u542b\u4e86\u4f17\u591a\u53c2\u6570":74,"\u884c":85,"\u884c\u4f18\u5148\u6b21\u5e8f\u5b58\u50a8":87,"\u884c\u5185\u4f7f\u7528":2,"\u884c\u653f\u5de5\u4f5c":90,"\u884c\u955c\u50cf\u5305\u62ec\u7eafcpu\u7248\u672c\u548cgpu\u7248\u672c\u4ee5\u53ca\u5176\u5bf9\u5e94\u7684\u975eavx\u7248\u672c":54,"\u8865\u5145\u4e0a\u6b21\u7684commit":66,"\u8868\u660e\u4e86\u8fd9\u4e9b\u884c\u7684\u6807\u53f7":67,"\u8868\u660e\u8fd9\u4e2a\u5c42\u7684\u4e00\u4e2a\u5b9e\u4f8b\u662f\u5426\u9700\u8981\u504f\u7f6e":67,"\u8868\u793a":68,"\u8868\u793a\u4e00\u4e2akubernetes\u96c6\u7fa4\u4e2d\u7684\u4e00\u4e2a\u5de5\u4f5c\u8282\u70b9":78,"\u8868\u793a\u4e3adeviceid":77,"\u8868\u793a\u5728\u96c6\u7fa4\u4f5c\u4e1a":73,"\u8868\u793a\u5973\u6027":90,"\u8868\u793a\u5c06\u5916\u5c42\u7684outer_mem\u4f5c\u4e3a\u5185\u5c42memory\u7684\u521d\u59cb\u72b6\u6001":61,"\u8868\u793a\u5f53\u524d\u96c6\u7fa4\u4f5c\u4e1a\u7684\u8282\u70b9":73,"\u8868\u793a\u6570\u636e\u7c7b\u578b":68,"\u8868\u793a\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":73,"\u8868\u793a\u751f\u6210\u6570\u636e\u7684\u5e8f\u5217id":94,"\u8868\u793a\u7528\u4e8e\u8bad\u7ec3\u6216\u9884\u6d4b":2,"\u8868\u793a\u7537\u6027":90,"\u8868\u793a\u7684\u6bcf\u4e2a\u5355\u8bcd":89,"\u8868\u793a\u7b2c0\u4e2abatch\u5230\u5f53\u524dbatch\u7684\u5206\u7c7b\u9519\u8bef":93,"\u8868\u793a\u8bad\u7ec3\u4e86xx\u4e2a\u6837\u672c":93,"\u8868\u793a\u8bad\u7ec3\u4e86xx\u4e2abatch":93,"\u8868\u793a\u8bbe\u5907\u7c7b\u578b":68,"\u8868\u793a\u8bfb\u8005\u6240\u4f7f\u7528\u7684docker\u955c\u50cf\u4ed3\u5e93\u5730\u5740":80,"\u8868\u793a\u8fc7\u4e8620\u4e2abatch":89,"\u8868\u793a\u8fc7\u4e862560\u4e2a\u6837\u672c":89,"\u8868\u793a\u8fd9\u4e2ajob\u7684\u540d\u5b57":80,"\u88ab":66,"\u88ab\u6269\u5c55\u4e3a\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":60,"\u88ab\u653e\u5728":67,"\u88ab\u79f0\u4e3a":64,"\u897f\u90e8\u7247":90,"\u8981\u4e0b\u8f7d\u548c\u89e3\u538b\u6570\u636e\u96c6":91,"\u8981\u4e0b\u8f7d\u89e3\u538b\u8fd9\u4e2a\u6a21\u578b":94,"\u8981\u4f7f\u7528\u547d\u4ee4\u884c\u5206\u6790\u5de5\u5177":72,"\u8981\u5728\u5df2\u6709\u7684kubernetes\u96c6\u7fa4\u4e0a\u8fdb\u884cpaddlepaddle\u7684\u5206\u5e03\u5f0f\u8bad\u7ec3":80,"\u8981\u5728\u6240\u6709\u8282\u70b9\u4e0a\u5b58\u5728":73,"\u8981\u5bf9\u4e00\u4e2a\u56fe\u7247\u7684\u8fdb\u884c\u5206\u7c7b\u9884\u6d4b":86,"\u8981\u5c06\u5b57\u6bb5\u914d\u7f6e\u6587\u4ef6\u8f6c\u5316\u4e3ameta\u914d\u7f6e\u6587\u4ef6":91,"\u8981\u6c42\u5355\u5c42\u5e8f\u5217\u542b\u6709\u5143\u7d20\u7684\u6570\u76ee":60,"\u8981\u751f\u6210\u7684\u76ee\u6807\u5e8f\u5217":63,"\u8981\u8c03\u7528":67,"\u89e3\u51b3\u529e\u6cd5\u662f":46,"\u89e3\u51b3\u65b9\u6848\u662f":51,"\u89e3\u538b":94,"\u89e3\u6790\u5668\u80fd\u901a\u8fc7\u6587\u4ef6\u7684\u6269\u5c55\u540d\u81ea\u52a8\u8bc6\u522b\u6587\u4ef6\u7684\u683c\u5f0f":91,"\u89e3\u6790\u6570\u636e\u96c6\u4e2d\u7684\u6bcf\u4e00\u4e2a\u5b57\u6bb5":91,"\u89e3\u6790\u6a21\u578b\u914d\u7f6e\u6587\u4ef6":4,"\u89e3\u6790\u73af\u5883\u53d8\u91cf\u5f97\u5230":80,"\u89e3\u6790\u8bad\u7ec3\u6a21\u578b\u65f6\u7528\u7684\u914d\u7f6e\u6587\u4ef6":4,"\u89e3\u7801\u5668\u4f7f\u7528":64,"\u89e3\u7801\u5668\u57fa\u4e8e\u7f16\u7801\u6e90\u5e8f\u5217\u548c\u6700\u540e\u751f\u6210\u7684\u76ee\u6807\u8bcd\u9884\u6d4b\u4e0b\u4e00\u76ee\u6807\u8bcd":64,"\u89e3\u7801\u5668\u662f\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":64,"\u89e3\u7801\u5668\u6839\u636e\u4e0a\u4e0b\u6587\u5411\u91cf\u9884\u6d4b\u51fa\u4e00\u4e2a\u76ee\u6807\u5355\u8bcd":94,"\u89e3\u91ca":89,"\u89e3\u91ca\u578b\u8bed\u8a00\u53ea\u80fd\u8c03\u7528\u52a8\u6001\u5e93":33,"\u89e3\u91ca\u6027\u8bed\u8a00\u5b9e\u9645\u8fd0\u884c\u7684\u4e8c\u8fdb\u5236\u662f\u89e3\u91ca\u5668\u672c\u8eab":33,"\u8ba1\u7b97":64,"\u8ba1\u7b97\u504f\u7f6e\u7684\u68af\u5ea6":67,"\u8ba1\u7b97\u5355\u5143\u662f\u795e\u7ecf\u7f51\u7edc\u7684\u6838\u5fc3":52,"\u8ba1\u7b97\u53cd\u5411rnn\u7684\u7b2c\u4e00\u4e2a\u5b9e\u4f8b":64,"\u8ba1\u7b97\u53d8\u6362\u77e9\u9635\u7684\u5927\u5c0f\u548c\u683c\u5f0f":67,"\u8ba1\u7b97\u5f53\u524d\u5c42\u6743\u91cd\u7684\u68af\u5ea6":67,"\u8ba1\u7b97\u6548\u7387\u66f4\u9ad8":50,"\u8ba1\u7b97\u6bcf\u4e2a\u8bcd\u7684\u8bcd\u5411\u91cf":64,"\u8ba1\u7b97\u6fc0\u6d3b\u51fd\u6570\u7684\u68af\u5ea6":67,"\u8ba1\u7b97\u7684\u7ec6\u8282\u5c06\u5728\u4e0b\u9762\u7684\u5c0f\u8282\u7ed9\u51fa":67,"\u8ba1\u7b97\u8bef\u5dee\u51fd\u6570":52,"\u8ba1\u7b97\u8f6c\u6362\u77e9\u9635\u548c\u8f93\u5165\u7684\u68af\u5ea6":67,"\u8ba1\u7b97\u8f93\u5165\u548c\u53c2\u6570\u7684\u68af\u5ea6":67,"\u8ba1\u7b97\u8f93\u5165\u5c42\u7684\u504f\u5dee":67,"\u8ba1\u7b97\u8f93\u51fa":67,"\u8ba1\u7b97\u8fd9\u4e2a\u6587\u4ef6\u7684md5":20,"\u8ba1\u7b97\u9700\u8981\u7684\u6570\u636e\u5b58\u653e\u5728":69,"\u8ba9\u6a21\u578b\u80fd\u591f\u5f97\u5230\u8bad\u7ec3\u66f4\u65b0":89,"\u8ba9\u795e\u7ecf\u7f51\u7edc\u53ef\u4ee5\u8fdb\u884c\u8bad\u7ec3\u6216\u9884\u6d4b":1,"\u8ba9\u8fd9\u4e2a\u793a\u4f8b\u53d8\u5f97\u66f4\u597d":91,"\u8ba9paddle\u6838\u5fc3\u4e2d":34,"\u8bad\u7ec3":[74,93],"\u8bad\u7ec3\u4e0e\u5e94\u7528":0,"\u8bad\u7ec3\u4efb\u52a1\u7684\u8fd0\u884c\u53ef\u80fd\u4f1a\u5360\u6ee1trainer\u548cparamet":20,"\u8bad\u7ec3\u4f5c\u4e1a":73,"\u8bad\u7ec3\u53ca\u6d4b\u8bd5\u8bef\u5dee\u66f2\u7ebf\u56fe\u4f1a\u88ab":86,"\u8bad\u7ec3\u53ef\u4ee5\u8bbe\u7f6e\u4e3atrue":92,"\u8bad\u7ec3\u540e":92,"\u8bad\u7ec3\u548c\u7eaf\u4f7f\u7528":42,"\u8bad\u7ec3\u5931\u8d25\u65f6\u53ef\u4ee5\u68c0\u67e5\u9519\u8bef\u65e5\u5fd7":73,"\u8bad\u7ec3\u597d\u4e00\u4e2a\u6df1\u5c42\u795e\u7ecf\u7f51\u7edc\u901a\u5e38\u8981\u8017\u8d39\u975e\u5e38\u957f\u7684\u65f6\u95f4":72,"\u8bad\u7ec3\u5b8c\u6210\u540e":52,"\u8bad\u7ec3\u6570\u636e\u548c\u6d4b\u8bd5\u6570\u636e\u7684\u76ee\u5f55":94,"\u8bad\u7ec3\u6570\u636e\u662f":2,"\u8bad\u7ec3\u6570\u636e\u6709\u95ee\u9898":49,"\u8bad\u7ec3\u6570\u6910\u96c6":93,"\u8bad\u7ec3\u65f6":80,"\u8bad\u7ec3\u65f6\u6240\u9700\u8bbe\u7f6e\u7684\u4e3b\u8981\u53c2\u6570\u5982\u4e0b":89,"\u8bad\u7ec3\u65f6\u9ed8\u8ba4shuffl":2,"\u8bad\u7ec3\u6a21\u578b\u4e4b\u524d":93,"\u8bad\u7ec3\u6a21\u578b\u540e":64,"\u8bad\u7ec3\u6a21\u578b\u6b63\u786e\u6027":42,"\u8bad\u7ec3\u7684\u635f\u5931\u51fd\u6570\u9ed8\u8ba4\u6bcf\u969410\u4e2abatch\u6253\u5370\u4e00\u6b21":94,"\u8bad\u7ec3\u7684\u811a\u672c\u662f":92,"\u8bad\u7ec3\u7ed3\u675f\u540e\u67e5\u770b\u8f93\u51fa\u7ed3\u679c":80,"\u8bad\u7ec3\u811a\u672c":89,"\u8bad\u7ec3\u811a\u672c\u7b49\u7b49":89,"\u8bad\u7ec3\u81f3\u591a":91,"\u8bad\u7ec3\u8282\u70b9\u6570\u91cf":80,"\u8bad\u7ec3\u8bed\u8a00\u6a21\u578b\u8ddd\u79bb":49,"\u8bad\u7ec3\u8f6e\u6b21":89,"\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u53c2\u6570\u6216\u8005\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u7684\u68af\u5ea6\u5c3a\u5ea6\u8fc7\u5927":49,"\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u6d4b\u8bd5test_period":74,"\u8bad\u7ec3\u8fc7\u7a0b\u662f\u5426\u4e3a\u672c\u5730\u6a21\u5f0f":75,"\u8bad\u7ec3\u8fc7\u7a0b\u662f\u5426\u4f7f\u7528gpu":75,"\u8bad\u7ec3\u914d\u7f6e\u4e2d\u7684\u8bbe\u5907\u5c5e\u6027\u5c06\u4f1a\u65e0\u6548":75,"\u8bad\u7ec3\u96c6\u548c\u6d4b\u8bd5\u96c6\u6587\u4ef6\u5217\u8868":93,"\u8bad\u7ec3\u96c6\u5df2\u7ecf\u505a\u4e86\u968f\u673a\u6253\u4e71\u6392\u5e8f\u800c\u6d4b\u8bd5\u96c6\u6ca1\u6709":93,"\u8bad\u7ec3\u96c6\u5df2\u7ecf\u968f\u673a\u6253\u4e71":93,"\u8bad\u7ec3\u96c6\u5e73\u5747\u503c":86,"\u8bad\u7ec3dot_period":74,"\u8bb0\u5f55\u4e0b\u6240\u6709\u5931\u8d25\u7684\u4f8b\u5b50":42,"\u8bb0\u5fc6\u6a21\u5757":64,"\u8bba\u6587":87,"\u8bbe\u4e3a\u5df2\u90e8\u7f72\u7684\u5de5\u4f5c\u7a7a\u95f4\u76ee\u5f55":73,"\u8bbe\u4e3a\u672c\u5730":73,"\u8bbe\u5b9a":50,"\u8bbe\u7f6e":[34,49,50],"\u8bbe\u7f6e\u4e3a":[59,67],"\u8bbe\u7f6e\u4e3a\u4e0d\u540c\u7684\u503c":50,"\u8bbe\u7f6e\u4e3atrue\u4f7f\u7528\u672c\u5730\u8bad\u7ec3\u6216\u8005\u4f7f\u7528\u96c6\u7fa4\u4e0a\u7684\u4e00\u4e2a\u8282\u70b9":75,"\u8bbe\u7f6e\u4e3atrue\u4f7f\u7528gpu\u6a21\u5f0f":75,"\u8bbe\u7f6e\u4e86\u76f8\u540c\u7684\u53d6\u503c":50,"\u8bbe\u7f6e\u4efb\u52a1\u7684\u6a21\u5f0f\u4e3a\u6d4b\u8bd5":94,"\u8bbe\u7f6e\u4fdd\u5b58\u6a21\u578b\u7684\u8f93\u51fa\u8def\u5f84":94,"\u8bbe\u7f6e\u5168\u5c40\u5b66\u4e60\u7387":93,"\u8bbe\u7f6e\u5176\u53c2\u6570\u5c5e\u6027":51,"\u8bbe\u7f6e\u5185\u5b58\u4e2d\u6682\u5b58\u7684\u6570\u636e\u6761\u6570":2,"\u8bbe\u7f6e\u5185\u5b58\u4e2d\u6700\u5c0f\u6682\u5b58\u7684\u6570\u636e\u6761\u6570":2,"\u8bbe\u7f6e\u53c2\u6570\u7684\u540d\u5b57":51,"\u8bbe\u7f6e\u547d\u4ee4\u884c\u53c2\u6570":[49,71,91],"\u8bbe\u7f6e\u5b57\u5178\u6587\u4ef6":93,"\u8bbe\u7f6e\u5b66\u4e60\u7387\u8870\u51cf\u56e0\u5b50\u5206\u6bb5\u51fd\u6570":51,"\u8bbe\u7f6e\u5de5\u4f5c\u6a21\u5f0f\u4e3a\u8bad\u7ec3":93,"\u8bbe\u7f6e\u5e73\u5747sgd\u7a97\u53e3":93,"\u8bbe\u7f6e\u6210":51,"\u8bbe\u7f6e\u6210\u4e00\u4e2a\u5c0f\u4e00\u4e9b\u7684\u503c":49,"\u8bbe\u7f6e\u6570\u636e\u904d\u5386\u6b21\u6570":92,"\u8bbe\u7f6e\u6807\u7b7e\u7c7b\u522b\u5b57\u5178":93,"\u8bbe\u7f6e\u6a21\u578b\u8def\u5f84":93,"\u8bbe\u7f6e\u7684\u547d\u4ee4\u884c\u53c2\u6570":93,"\u8bbe\u7f6e\u795e\u7ecf\u7f51\u7edc\u7684\u914d\u7f6e\u6587\u4ef6":94,"\u8bbe\u7f6e\u7c7b\u522b\u6570":93,"\u8bbe\u7f6e\u7ebf\u7a0b\u6570":[92,93],"\u8bbe\u7f6e\u7f51\u7edc\u914d\u7f6e":93,"\u8bbe\u7f6e\u8f93\u51fa\u7684\u5c3a\u5bf8":67,"\u8bbe\u7f6e\u8f93\u51fa\u8def\u5f84\u4ee5\u4fdd\u5b58\u8bad\u7ec3\u5b8c\u6210\u7684\u6a21\u578b":93,"\u8bbe\u7f6e\u8f93\u51fatensor\u7684\u5f62\u72b6":68,"\u8bbe\u7f6e\u8fd9\u4e2apydataprovider2\u8fd4\u56de\u4ec0\u4e48\u6837\u7684\u6570\u636e":2,"\u8bbe\u7f6e\u9ed8\u8ba4\u8bbe\u5907\u53f7\u4e3a0":77,"\u8bbe\u7f6ebatch":93,"\u8bbe\u7f6ecpu\u7ebf\u7a0b\u6570\u6216\u8005gpu\u8bbe\u5907\u6570":94,"\u8bbe\u7f6egpu":75,"\u8bbe\u7f6epass":93,"\u8bbe\u7f6epasses\u7684\u6570\u91cf":94,"\u8bbf\u95ee\u5230\u4e3b\u673a\u4e0a\u7684\u6587\u4ef6":54,"\u8bbf\u95eekubernetes\u7684\u63a5\u53e3\u6765\u67e5\u8be2\u6b64job\u5bf9\u5e94\u7684\u6240\u6709pod\u4fe1\u606f":80,"\u8bc4\u4ef7\u9884\u6d4b\u7684\u6548\u679c":52,"\u8bc4\u4f30\u8be5\u4ea7\u54c1\u7684\u8d28\u91cf":89,"\u8bc4\u5206":[90,91],"\u8bc4\u5206\u6587\u4ef6\u7684\u6bcf\u4e00\u884c\u4ec5\u4ec5\u63d0\u4f9b\u7535\u5f71\u6216\u7528\u6237\u7684\u7f16\u53f7\u6765\u4ee3\u8868\u76f8\u5e94\u7684\u7535\u5f71\u6216\u7528\u6237":91,"\u8bc4\u5206\u88ab\u8c03\u6574\u4e3a5\u661f\u7684\u89c4\u6a21":90,"\u8bc4\u5ba1\u4eba\u4e00\u822c\u4e0d\u505a\u8bc4\u5ba1":66,"\u8bc4\u5ba1\u4eba\u7684\u6bcf\u4e2a\u610f\u89c1\u90fd\u5fc5\u987b\u56de\u590d":66,"\u8bc4\u5ba1\u4eba\u9700\u8981\u9010\u4e00\u67e5\u770b\u6bcf\u4e2acommit\u624d\u80fd\u77e5\u9053\u505a\u4e86\u54ea\u4e9b\u4fee\u6539":66,"\u8bc4\u8bba\u6846\u4e2d\u52a0\u4e0a":66,"\u8bc6\u522b\u6570\u5b57":42,"\u8bcd\u5178\u7684key\u548c":68,"\u8bcd\u5411\u91cf":[42,85],"\u8bcd\u5411\u91cf\u6a21\u578b":88,"\u8bcd\u5411\u91cf\u6a21\u578b\u540d\u79f0":85,"\u8bcd\u672c\u8eab\u548c\u8bcd\u9891":85,"\u8bcd\u9891\u6700\u9ad8\u7684":94,"\u8bd5\u7740\u8ba9\u8f93\u51fa\u7684\u5206\u6790\u6570\u636e\u548c\u7406\u8bba\u503c\u5bf9\u5e94":72,"\u8be5":[73,92],"\u8be5\u51fd\u6570\u5177\u6709\u4e24\u4e2a\u53c2\u6570":2,"\u8be5\u51fd\u6570\u5728\u521d\u59cb\u5316\u7684\u65f6\u5019\u4f1a\u88ab\u8c03\u7528":2,"\u8be5\u51fd\u6570\u7684\u529f\u80fd\u662f":2,"\u8be5\u53c2\u6570\u5728\u7f51\u7edc\u914d\u7f6e\u7684output":75,"\u8be5\u53c2\u6570\u5728\u96c6\u7fa4\u63d0\u4ea4\u73af\u5883\u4e2d\u81ea\u52a8\u8bbe\u7f6e":75,"\u8be5\u53c2\u6570\u5df2\u7ecf\u5728\u96c6\u7fa4\u63d0\u4ea4\u73af\u5883\u4e2d\u5b8c\u6210\u8bbe\u7f6e":75,"\u8be5\u53c2\u6570\u5fc5\u987b\u80fd\u88abflag":75,"\u8be5\u53c2\u6570\u6307\u793a\u662f\u5426\u6253\u5370\u65e5\u5fd7\u622a\u65ad\u4fe1\u606f":75,"\u8be5\u53c2\u6570\u6307\u793a\u662f\u5426\u6253\u5370\u9519\u8bef\u622a\u65ad\u65e5\u5fd7":75,"\u8be5\u53c2\u6570\u7528\u4e8e\u6307\u5b9a\u52a8\u6001\u5e93\u8def\u5f84":75,"\u8be5\u53c2\u6570\u7684\u610f\u601d\u662f\u8bad\u7ec3num":75,"\u8be5\u53c2\u6570\u9ed8\u8ba4\u4e3anull":75,"\u8be5\u5bf9\u8c61\u5177\u6709\u4ee5\u4e0b\u4e24\u4e2a\u5c5e\u6027":2,"\u8be5\u5c42\u4ec5\u9700\u8981\u8fd9\u4e9b\u975e\u96f6\u6837\u672c\u4f4d\u7f6e\u6240\u5bf9\u5e94\u7684\u53d8\u6362\u77e9\u9635\u7684\u90a3\u4e9b\u884c":67,"\u8be5\u5c42\u795e\u7ecf\u5143\u4e2a\u6570":89,"\u8be5\u622a\u65ad\u4f1a\u5f71\u54cd":75,"\u8be5\u6279\u6b21\u7684\u8f93\u5165\u4e2d\u4ec5\u6709\u4e00\u4e2a\u5b50\u96c6\u662f\u975e\u96f6\u7684":67,"\u8be5\u63a5\u53e3\u4f7f\u7528\u591a\u7ebf\u7a0b\u8bfb\u53d6\u6570\u636e":2,"\u8be5\u63a5\u53e3\u53ef\u7528\u4e8e\u9884\u6d4b\u548c\u5b9a\u5236\u5316\u8bad\u7ec3":53,"\u8be5\u6570\u636e\u53ca\u6709\u5f88\u591a\u4e0d\u540c\u7684\u7248\u672c":90,"\u8be5\u6570\u636e\u96c6":85,"\u8be5\u6570\u636e\u96c6\u4e8e2003\u5e742\u6708\u53d1\u5e03":90,"\u8be5\u6570\u636e\u96c6\u5305\u542b\u4e00\u4e9b\u7528\u6237\u4fe1\u606f":90,"\u8be5\u6570\u76ee\u662f\u63d0\u524d\u5b9a\u4e49\u597d\u7684":75,"\u8be5\u6587\u4ef6\u53ef\u4ee5\u4ece\u5b57\u6bb5\u914d\u7f6e\u6587\u4ef6\u751f\u6210":91,"\u8be5\u6587\u4ef6\u662f\u7531cpickle\u4ea7\u751f\u7684":87,"\u8be5\u6587\u4ef6\u662fpython\u7684pickle\u5bf9\u8c61":91,"\u8be5\u6587\u4ef6\u8d1f\u8d23\u4ea7\u751f\u56fe\u7247\u6570\u636e\u5e76\u4f20\u9012\u7ed9paddle\u7cfb\u7edf":86,"\u8be5\u6a21\u578b\u4f9d\u7136\u4f7f\u7528\u903b\u8f91\u56de\u5f52\u5206\u7c7b\u7f51\u7edc\u7684\u6846\u67b6":89,"\u8be5\u6a21\u578b\u5728\u957f\u8bed\u53e5\u7ffb\u8bd1\u7684\u573a\u666f\u4e0b\u6548\u679c\u63d0\u5347\u66f4\u52a0\u660e\u663e":94,"\u8be5\u6a21\u578b\u7684\u7f51\u7edc\u914d\u7f6e\u5982\u4e0b":52,"\u8be5\u6a21\u578b\u7684\u8bf4\u660e\u5982\u4e0b\u56fe\u6240\u793a":64,"\u8be5\u6a21\u578b\u7f51\u7edc\u53ea\u662f\u7528\u4e8e\u8fdb\u884cdemo\u5c55\u793apaddle\u5982\u4f55\u5de5\u4f5c":91,"\u8be5\u793a\u4f8b\u5c06\u5c55\u793apaddle\u5982\u4f55\u8fdb\u884c\u8bcd\u5411\u91cf\u5d4c\u5165":91,"\u8be5\u793a\u4f8b\u7684\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e\u6587\u4ef6":91,"\u8be5\u7b97\u6cd5\u6bcf\u6279\u91cf":52,"\u8be5\u7c7b\u7684":68,"\u8be5\u7c7b\u7684\u5b9e\u73b0\u7ec6\u8282\u5728":67,"\u8be5\u7f16\u8bd1\u5de5\u5177\u94fe\u9700\u8981\u7cfb\u7edfglibc\u652f\u63012":59,"\u8be5\u811a\u672c\u4ec5\u4ec5\u662f\u5f00\u59cb\u4e00\u4e2apaddle\u8bad\u7ec3\u8fc7\u7a0b":91,"\u8be5\u811a\u672c\u4f1a\u751f\u6210\u4e00\u4e2adot\u6587\u4ef6":87,"\u8be5\u811a\u672c\u5c06\u8f93\u51fa\u9884\u6d4b\u5206\u7c7b\u7684\u6807\u7b7e":86,"\u8be5\u8bed\u53e5\u4f1a\u4e3a\u6bcf\u4e2a\u5c42\u521d\u59cb\u5316\u5176\u6240\u9700\u8981\u7684\u53d8\u91cf\u548c\u8fde\u63a5":67,"\u8be5github\u4ed3\u5e93\u4e2d\u5305\u542b\u82e5\u5e72\u4e2a\u9884\u7f16\u8bd1\u597d\u7684":59,"\u8be5layer\u662f\u901a\u8fc7\u53c2\u6570":50,"\u8be6\u7ec6\u4ecb\u7ecd\u53ef\u4ee5\u53c2\u8003":61,"\u8be6\u7ec6\u4ecb\u7ecd\u8bf7\u53c2\u8003\u8bbe\u8ba1\u6587\u6863":68,"\u8be6\u7ec6\u4fe1\u606f\u8bf7\u68c0\u67e5":73,"\u8be6\u7ec6\u5185\u5bb9\u8bf7\u53c2\u89c1":89,"\u8be6\u7ec6\u53ef\u53c2\u8003":66,"\u8be6\u7ec6\u5730\u5c55\u793a\u4e86\u6574\u4e2a\u7279\u5f81\u63d0\u53d6\u7684\u8fc7\u7a0b":87,"\u8be6\u7ec6\u6587\u6863\u53c2\u8003":49,"\u8be6\u7ec6\u7684\u53c2\u6570\u89e3\u91ca":89,"\u8be6\u7ec6\u7684cmake\u4f7f\u7528\u65b9\u6cd5\u53ef\u4ee5\u53c2\u8003":53,"\u8be6\u7ec6\u89c1":60,"\u8be6\u7ec6\u8bbe\u8ba1":28,"\u8bed\u4e49\u89d2\u8272\u6807\u6ce8":[88,92],"\u8bed\u610f\u89d2\u8272\u6807\u6ce8":42,"\u8bed\u8a00\u6a21\u578b":85,"\u8bf4\u660e":[20,53,80],"\u8bf4\u660e\u63d0\u4ea4\u7684\u4ee3\u7801\u5b58\u5728\u95ee\u9898":66,"\u8bf4\u660e\u6bcf\u4e2a\u7279\u5f81\u6587\u4ef6\u5177\u4f53\u5b57\u6bb5\u662f":91,"\u8bf4\u660e\u8fd9\u4e2a\u5c42\u7684\u8f93\u5165":67,"\u8bf7\u4e0d\u8981\u521b\u5efa\u7a7a\u7684":68,"\u8bf7\u4fdd\u8bc1travi":66,"\u8bf7\u5148\u4f7f\u7528":[58,59],"\u8bf7\u53c2\u7167\u7f51\u7edc\u914d\u7f6e\u7684\u6587\u6863\u4e86\u89e3\u66f4\u8be6\u7ec6\u7684\u4fe1\u606f":77,"\u8bf7\u53c2\u8003":[2,34,46,49,56,61,67,68,89],"\u8bf7\u53c2\u8003\u5982\u4e0b\u8868\u683c":89,"\u8bf7\u53c2\u8003\u9875\u9762":91,"\u8bf7\u53c2\u8003layer\u6587\u6863":86,"\u8bf7\u53c2\u89c1":66,"\u8bf7\u53c2\u9605":64,"\u8bf7\u53c2\u9605\u60c5\u611f\u5206\u6790\u7684\u6f14\u793a\u4ee5\u4e86\u89e3\u6709\u5173\u957f\u671f\u77ed\u671f\u8bb0\u5fc6\u5355\u5143\u7684\u66f4\u591a\u4fe1\u606f":92,"\u8bf7\u5728\u8be5pull":66,"\u8bf7\u60a8\u6bcf\u6b21\u63d0\u4ea4\u4ee3\u7801\u65f6":66,"\u8bf7\u60a8\u9075\u5b88\u4ee5\u4e0b\u7ea6\u5b9a":66,"\u8bf7\u628a\u6587\u6863\u91cc\u547d\u4ee4\u4e2d\u7684paddlepaddl":54,"\u8bf7\u6307\u5b9a\u8be5\u76ee\u5f55":75,"\u8bf7\u663e\u793a\u5730\u8c03\u7528":68,"\u8bf7\u67e5\u770b":85,"\u8bf7\u6ce8\u610f":[54,64,68,79,85],"\u8bf7\u6ce8\u610f\u6bcf\u4e2acommit\u7684\u540d\u79f0":66,"\u8bf7\u6ce8\u610f\u8fd9\u4e2a\u547d\u4ee4\u7ed3\u5c3e\u5904\u7684":65,"\u8bf7\u6ce8\u610fcommit\u7684\u6570\u91cf":66,"\u8bf7\u76f4\u63a5\u586b\u51450":51,"\u8bf7\u770b\u4e0b\u9762\u7684\u4f8b\u5b50":77,"\u8bf7\u786e\u4fdd":66,"\u8bf7\u7ed9\u51fa\u603b\u4f53\u7684\u4fee\u6539\u60c5\u51b5":66,"\u8bf7\u7ed9\u51fa\u60a8\u81ea\u5df1\u7684\u53cd\u9a73\u7406\u7531":66,"\u8bf7\u8bb0\u4f4f":73,"\u8bf7\u8fd0\u884c":54,"\u8bf7\u9009\u62e9\u5408\u9002\u7684\u8bcd\u6c47":66,"\u8bf7\u9009\u62e9\u6b63\u786e\u7684\u7248\u672c":46,"\u8bf7\u9075\u5b88":66,"\u8bf7\u91c7\u7528":66,"\u8bf8\u5982\u56fe\u50cf\u5206\u7c7b":77,"\u8bfb\u53d612\u4e2a\u91c7\u6837\u6570\u636e\u8fdb\u884c\u968f\u673a\u68af\u5ea6\u8ba1\u7b97\u6765\u66f4\u65b0\u66f4\u65b0":52,"\u8bfb\u53d6\u6570\u636e":2,"\u8bfb\u53d6\u6bcf\u4e00\u884c":2,"\u8bfb\u53d6volume\u4e2d\u7684\u6570\u636e\u8fdb\u884c\u8fd9\u6b21\u5206\u5e03\u5f0f\u8bad\u7ec3":80,"\u8bfb\u8005\u53ef\u4ee5\u67e5\u770b":80,"\u8bfb\u8005\u9700\u8981\u66ff\u6362\u6210\u81ea\u5df1\u4f7f\u7528\u7684\u4ed3\u5e93\u5730\u5740":80,"\u8c03\u7528":[67,68,86],"\u8c03\u7528\u4e00\u6b21":2,"\u8c03\u7528\u4e0a\u9762\u7684process\u51fd\u6570\u83b7\u5f97\u89c2\u6d4b\u6570\u636e":52,"\u8c03\u7528\u5bf9\u5e94":69,"\u8c03\u7528\u65b9\u6cd5\u89c1c":58,"\u8c03\u7528\u7684\u51fd\u6570\u662f\u5426\u652f\u6301\u4e0d\u540c\u8bbe\u5907":68,"\u8c03\u7528\u7684pydataprovider2\u662f":2,"\u8c03\u7528\u7b2c\u4e8c\u6b21\u7684\u65f6\u5019":2,"\u8c03\u7528\u8be5\u51fd\u6570\u540e":67,"\u8c03\u7528\u8fd9\u4e2apydataprovider2\u7684\u65b9\u6cd5":2,"\u8c13\u8bcd\u4e0a\u4e0b\u6587":92,"\u8d1f\u6837\u672c":89,"\u8d1f\u9762\u7684\u8bc4\u8bba\u7684\u5f97\u5206\u5c0f\u4e8e\u7b49\u4e8e4":93,"\u8d1f\u9762\u8bc4\u4ef7\u6837\u672c":93,"\u8d21\u732e\u6587\u6863":70,"\u8d44\u6e90\u5bf9\u8c61\u7684\u540d\u5b57\u662f\u552f\u4e00\u7684":78,"\u8d77":61,"\u8def\u5f84\u4e0b":[52,87],"\u8df3\u8f6c\u5230":66,"\u8df3\u8fc7":49,"\u8f6c\u4e3ajpeg\u6587\u4ef6\u5e76\u5b58\u5165\u7279\u5b9a\u7684\u76ee\u5f55":86,"\u8f6c\u6362\u751f\u6210\u7684\u6587\u4ef6\u540d\u4f1a\u662f\u4ee5\u4e0b\u683c\u5f0f":21,"\u8f6c\u6362\u8fc7\u6765\u7684":87,"\u8f6e":91,"\u8f83":61,"\u8f93\u5165":[60,64],"\u8f93\u5165\u548c\u8f93\u51fa\u90fd\u662f\u5355\u5c42\u5e8f\u5217":63,"\u8f93\u5165\u548c\u8f93\u51fa\u90fd\u662f\u53cc\u5c42\u5e8f\u5217":63,"\u8f93\u5165\u56fe\u7247\u7684\u9ad8\u5ea6\u53ca\u5bbd\u5ea6":86,"\u8f93\u5165\u5c42\u5c3a\u5bf8":87,"\u8f93\u5165\u5e8f\u5217\u4e2d\u5143\u7d20\u7684\u603b\u6570":49,"\u8f93\u5165\u6570\u636e\u4e3a\u4e00\u4e2a\u5b8c\u6574\u7684\u65f6\u95f4\u5e8f\u5217":61,"\u8f93\u5165\u6570\u636e\u4e3a\u5728\u5355\u5c42rnn\u6570\u636e\u91cc\u9762":61,"\u8f93\u5165\u6570\u636e\u6574\u4f53\u4e0a\u662f\u4e00\u4e2a\u65f6\u95f4\u5e8f\u5217":61,"\u8f93\u5165\u6570\u636e\u7684\u5b57\u5178\u7ef4\u6570\u662f1\u767e\u4e07":77,"\u8f93\u5165\u6570\u6910\u6240\u5728\u76ee\u5f55":93,"\u8f93\u5165\u6587\u672c":85,"\u8f93\u5165\u6587\u672c\u4e2d\u6ca1\u6709\u5934\u90e8":85,"\u8f93\u5165\u662f\u5426\u662f\u8f6c\u7f6e\u7684":67,"\u8f93\u5165\u662f\u7531\u4e00\u4e2alist\u4e2d\u7684\u7f51\u7edc\u5c42\u5b9e\u4f8b\u7684\u540d\u5b57\u7ec4\u6210\u7684":67,"\u8f93\u5165\u7279\u5f81\u56fe\u7684\u901a\u9053\u6570\u76ee":87,"\u8f93\u5165\u7684":85,"\u8f93\u5165\u7684\u539f\u59cb\u6570\u636e\u96c6\u8def\u5f84":94,"\u8f93\u5165\u7684\u540d\u5b57":67,"\u8f93\u5165\u7684\u5927\u5c0f":67,"\u8f93\u5165\u7684\u6587\u672c\u683c\u5f0f\u5982\u4e0b":85,"\u8f93\u5165\u7684\u6587\u672c\u8bcd\u5411\u91cf\u6a21\u578b\u540d\u79f0":85,"\u8f93\u5165\u7684\u7c7b\u578b":67,"\u8f93\u5165\u8bcd\u5178":68,"\u8f93\u5165\u95e8":93,"\u8f93\u5165\u9884\u6d4b\u6837\u672c":93,"\u8f93\u5165n\u4e2a\u5355\u8bcd":89,"\u8f93\u51fa":[60,64,68],"\u8f93\u51fa\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":63,"\u8f93\u51fa\u4e00\u4e2a\u53cc\u5c42\u5e8f\u5217":63,"\u8f93\u51fa\u4e3an\u4e2aword_dim\u7ef4\u5ea6\u5411\u91cf":89,"\u8f93\u51fa\u51fd\u6570":64,"\u8f93\u51fa\u5e8f\u5217\u7684\u7c7b\u578b":60,"\u8f93\u51fa\u5e8f\u5217\u7684\u8bcd\u8bed\u6570\u548c\u8f93\u5165\u5e8f\u5217\u4e00\u81f4":63,"\u8f93\u51fa\u5e94\u8be5\u7c7b\u4f3c\u5982\u4e0b":91,"\u8f93\u51fa\u6587\u4ef6\u7684\u683c\u5f0f\u8bf4\u660e":85,"\u8f93\u51fa\u65e5\u5fd7\u4fdd\u5b58\u5728\u8def\u5f84":93,"\u8f93\u51fa\u65e5\u5fd7\u8bf4\u660e\u5982\u4e0b":93,"\u8f93\u51fa\u67092\u5217":85,"\u8f93\u51fa\u7279\u5f81\u56fe\u7684\u901a\u9053\u6570\u76ee":87,"\u8f93\u51fa\u7684\u4e8c\u8fdb\u5236\u8bcd\u5411\u91cf\u6a21\u578b\u540d\u79f0":85,"\u8f93\u51fa\u7684\u6587\u672c\u6a21\u578b\u540d\u79f0":85,"\u8f93\u51fa\u7684\u68af\u5ea6":75,"\u8f93\u51fa\u76ee\u5f55":87,"\u8f93\u51fa\u7ed3\u679c\u53ef\u80fd\u4f1a\u968f\u7740\u5bb9\u5668\u7684\u6d88\u8017\u800c\u88ab\u5220\u9664":79,"\u8fbe\u5230\u5bb9\u707e\u7684\u76ee\u7684":20,"\u8fc7\u4e86\u4e00\u4e2a\u5f88\u7b80\u5355\u7684recurrent_group":61,"\u8fc7\u5b8c\u6240\u6709\u8bad\u7ec3\u6570\u636e\u5373\u4e3a\u4e00\u4e2apass":49,"\u8fd0":54,"\u8fd0\u884c\u4e00\u4e2a":65,"\u8fd0\u884c\u4e0b\u9762\u547d\u4ee4\u5373\u53ef":91,"\u8fd0\u884c\u4e0b\u9762\u547d\u4ee4\u53ef\u4ee5\u8fdb\u884c\u7f16\u8bd1":68,"\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4\u6765\u4e0b\u8f7d\u8fd9\u4e2a\u811a\u672c":94,"\u8fd0\u884c\u4ee5\u4e0b\u6307\u4ee4":54,"\u8fd0\u884c\u4ee5\u4e0b\u7684\u547d\u4ee4\u4e0b\u8f7d\u548c\u83b7\u53d6\u6211\u4eec\u7684\u5b57\u5178\u548c\u9884\u8bad\u7ec3\u6a21\u578b":85,"\u8fd0\u884c\u4ee5\u4e0b\u7684\u547d\u4ee4\u4e0b\u8f7d\u6570\u636e\u96c6":85,"\u8fd0\u884c\u4ee5\u4e0b\u8bad\u7ec3\u547d\u4ee4":52,"\u8fd0\u884c\u4ee5\u53ca\u53d1\u5e03\u60a8\u7684ai\u7a0b\u5e8f":54,"\u8fd0\u884c\u5206\u5e03\u5f0f\u4f5c\u4e1a":73,"\u8fd0\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3":[49,71,89],"\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5":[54,65],"\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5\u6d4b\u65f6\u9700\u8981\u7f16\u8bd1\u6574\u4e2a\u5de5\u7a0b":68,"\u8fd0\u884c\u5931\u8d25":77,"\u8fd0\u884c\u5b8c\u4ee5\u4e0a\u547d\u4ee4":85,"\u8fd0\u884c\u5b8c\u6210\u540e":73,"\u8fd0\u884c\u5b8c\u811a\u672c":93,"\u8fd0\u884c\u6210\u529f\u4ee5\u540e":85,"\u8fd0\u884c\u6210\u529f\u540e\u76ee\u5f55":93,"\u8fd0\u884c\u65e5\u5fd7":73,"\u8fd0\u884c\u7684\u4e00\u4e9b\u53c2\u6570\u901a\u8fc7\u8fd9\u79cd\u65b9\u5f0f\u4f20\u9012\u5230\u5bb9\u5668\u5185":80,"\u8fd1":61,"\u8fd1\u671f\u63d0\u51fa\u7684nmt\u6a21\u578b\u901a\u5e38\u90fd\u5c5e\u4e8e\u7f16\u89e3\u7801\u6a21\u578b":94,"\u8fd4\u56de":[7,8,9,12,14],"\u8fd4\u56de0":2,"\u8fd4\u56de8\u4e2a\u7279\u5f81list\u548c1\u4e2a\u6807\u7b7elist":92,"\u8fd4\u56de\u4e00\u6761\u5b8c\u6574\u7684\u6837\u672c":2,"\u8fd4\u56de\u6570\u636e\u7684\u6bcf\u4e00\u6761\u6837\u672c\u7ed9":91,"\u8fd4\u56de\u65f6":2,"\u8fd4\u56de\u7684\u662f":[2,56],"\u8fd4\u56de\u7684\u987a\u5e8f\u9700\u8981\u548cinput_types\u4e2d\u5b9a\u4e49\u7684\u987a\u5e8f\u4e00\u81f4":2,"\u8fd4\u56de\u7b2c\u4e8c\u6b65":42,"\u8fd4\u56de\u7b2ci\u4e2a\u8f93\u5165\u77e9\u9635":67,"\u8fd4\u56de\u7c7b\u578b":[7,8,9,12,14],"\u8fd4\u56depython\u7aef\u7684\u8ba1\u7b97\u7ed3\u679c":68,"\u8fd8\u4f1a":61,"\u8fd8\u4f1a\u5f3a\u5236\u8bbe\u7f6e\u4e00\u4e9bpaddlepaddle\u53c2\u6570\u7684\u503c":58,"\u8fd8\u4f1a\u8f93\u51fa\u4e00\u4e2a":66,"\u8fd8\u53ef\u4ee5\u901a\u8fc7\u51cf\u5c0f\u5b66\u4e60\u5f8b\u6216\u8005\u5bf9\u6570\u636e\u8fdb\u884c\u5f52\u4e00\u5316\u5904\u7406\u6765\u89e3\u51b3\u8fd9\u7c7b\u95ee\u9898":49,"\u8fd8\u662f":61,"\u8fd8\u662f\u4ece":21,"\u8fd8\u662f\u865a\u62df\u673a":65,"\u8fd8\u6709":61,"\u8fd8\u80fd\u5904\u7406\u5176\u4ed6\u7528\u6237\u81ea\u5b9a\u4e49\u7684\u6570\u636e":93,"\u8fd8\u91c7\u7528\u4e86\u4e24\u4e2a\u5176\u4ed6\u7279\u5f81":92,"\u8fd8\u9700\u8981\u8fdb\u884c\u9884\u5904\u7406":86,"\u8fd8\u9700\u8981\u91cd\u5199":68,"\u8fd9":[49,61,89],"\u8fd98\u79cdlearning_rate_schedule\u53ca\u5176\u5bf9\u5e94\u5b66\u4e60\u7387\u8ba1\u7b97\u65b9\u5f0f\u5982\u4e0b":51,"\u8fd9\u4e00\u5757\u7684\u8017\u65f6\u6bd4\u4f8b\u771f\u7684\u592a\u9ad8":72,"\u8fd9\u4e00\u5c42\u8fdb\u884c\u5c01\u88c5":34,"\u8fd9\u4e00\u6982\u5ff5\u4e0d\u518d\u7410\u788e":34,"\u8fd9\u4e00\u8ba1\u7b97\u5355\u5143":50,"\u8fd9\u4e00\u8fc7\u7a0b\u5bf9\u7528\u6237\u662f\u5b8c\u5168\u900f\u660e\u7684":63,"\u8fd9\u4e09\u4e2a\u5206\u652f":42,"\u8fd9\u4e09\u4e2a\u6b65\u9aa4\u53ef\u914d\u7f6e\u4e3a":89,"\u8fd9\u4e0e\u672c\u5730\u8bad\u7ec3\u76f8\u540c":73,"\u8fd9\u4e24\u4e2a\u6587\u4ef6\u5939\u4e0b\u5404\u81ea\u670910\u4e2a\u5b50\u6587\u4ef6\u5939":86,"\u8fd9\u4e24\u4e2a\u6807\u51c6":92,"\u8fd9\u4e2a":[61,65,78],"\u8fd9\u4e2a\u4efb\u52a1\u7684\u914d\u7f6e\u4e3a":49,"\u8fd9\u4e2a\u4efb\u52a1\u7684dataprovider\u4e3a":49,"\u8fd9\u4e2a\u4f8b\u5b50\u6709\u4e24\u5904\u4e0d\u540c":68,"\u8fd9\u4e2a\u51fd\u6570\u7684":64,"\u8fd9\u4e2a\u51fd\u6570\u8fdb\u884c\u53d8\u6362":61,"\u8fd9\u4e2a\u51fd\u6570\u9700\u8981\u8bbe\u7f6e":64,"\u8fd9\u4e2a\u52a8\u6001\u5e93\u7684\u8fde\u63a5\u53c2\u6570\u4e0epaddle\u7684\u5176\u4ed6\u4e8c\u8fdb\u5236":34,"\u8fd9\u4e2a\u53c2\u6570\u4e5f\u4e0d\u4f1a\u4e00\u5e76\u5220\u9664":34,"\u8fd9\u4e2a\u5411\u91cf\u4e0e\u6e90\u4e2d\u641c\u7d22\u51fa\u7684\u4f4d\u7f6e\u548c\u6240\u6709\u4e4b\u524d\u751f\u6210\u7684\u76ee\u6807\u5355\u8bcd\u6709\u5173":94,"\u8fd9\u4e2a\u5730\u5740\u5219\u4e3a\u5b83\u7684\u7edd\u5bf9\u8def\u5f84\u6216\u76f8\u5bf9\u8def\u5f84":1,"\u8fd9\u4e2a\u5730\u5740\u6765\u8868\u793a\u6b64\u6b65\u9aa4\u6240\u6784\u5efa\u51fa\u7684\u955c\u50cf":80,"\u8fd9\u4e2a\u57fa\u7c7b":67,"\u8fd9\u4e2a\u5934\u6587\u4ef6\u4e0d\u5047\u8bbe\u5176\u4ed6\u6587\u4ef6\u7684\u5f15\u7528\u987a\u5e8f":34,"\u8fd9\u4e2a\u5b57\u5178\u662f\u6574\u6570\u6807\u7b7e\u548c\u5b57\u7b26\u4e32\u6807\u7b7e\u7684\u4e00\u4e2a\u5bf9\u5e94":93,"\u8fd9\u4e2a\u5e8f\u5217\u7684\u6bcf\u4e2a\u5143\u7d20\u53c8\u662f\u4e00\u4e2a\u5e8f\u5217":63,"\u8fd9\u4e2a\u63a5\u53e3\u9700\u8981\u505a\u5230":33,"\u8fd9\u4e2a\u6570\u636e\u4e5f\u88ab\u5355\u5c42rnn\u7f51\u7edc\u76f4\u63a5\u4f7f\u7528":61,"\u8fd9\u4e2a\u6570\u91cf\u79f0\u4e3abeam":94,"\u8fd9\u4e2a\u6587\u4ef6\u5177\u6709\u72ec\u7279\u7684\u8bed\u6cd5":33,"\u8fd9\u4e2a\u663e\u793a\u5668\u5f88\u68d2":89,"\u8fd9\u4e2a\u6a21\u578b\u5bf9\u4e8e\u7f16\u89e3\u7801\u6a21\u578b\u6765\u8bf4":94,"\u8fd9\u4e2a\u76ee\u5f55\u4e2d\u9664\u4e86":34,"\u8fd9\u4e2a\u795e\u7ecf\u7f51\u7edc\u5355\u5143\u5c31\u53ebmemori":61,"\u8fd9\u4e2a\u7a0b\u5e8f\u662f\u60a8\u5728\u5f00\u53d1\u673a\u4e0a\u4f7f\u7528\u5f00\u53d1\u955c\u50cf\u5b8c\u6210\u5f00\u53d1":54,"\u8fd9\u4e2a\u7c7b\u7684\u53c2\u6570\u5305\u62ec":67,"\u8fd9\u4e2a\u7c7b\u9700\u8981\u7ee7\u627f":67,"\u8fd9\u4e2a\u7cfb\u7edf\u5c06srl\u4efb\u52a1\u89c6\u4e3a\u5e8f\u5217\u6807\u6ce8\u95ee\u9898":92,"\u8fd9\u4e2a\u7ed3\u6784\u4f53\u4e2d\u7684\u53e6\u4e00\u4e2a\u9879\u76ee\u662f":34,"\u8fd9\u4e2a\u7ed3\u6784\u4f53\u5305\u542b\u4e24\u4e2a\u9879\u76ee":34,"\u8fd9\u4e2a\u811a\u672c\u8c03\u7528":65,"\u8fd9\u4e2a\u8282\u70b9\u53ef\u4ee5\u662f\u7269\u7406\u673a\u6216\u8005\u865a\u62df\u673a":78,"\u8fd9\u4e2a\u8868\u683c":78,"\u8fd9\u4e2a\u8f93\u5165\u4e0d\u53c2\u4e0e":68,"\u8fd9\u4e2a\u8fc7\u7a0b\u5bf9\u7528\u6237\u4e5f\u662f\u900f\u660e\u7684":63,"\u8fd9\u4e2a\u8fc7\u7a0b\u5c31\u662f\u6a21\u578b\u7684\u8bad\u7ec3\u8fc7\u7a0b":52,"\u8fd9\u4e2a\u8fc7\u7a0b\u9664\u4e86\u7f16\u8bd1paddlepaddle\u4e3a":66,"\u8fd9\u4e2a\u914d\u7f6e\u4e0e":85,"\u8fd9\u4e2a\u914d\u7f6e\u6587\u4ef6":78,"\u8fd9\u4e2a\u914d\u7f6e\u662f\u5426\u7528\u6765\u751f\u6210":94,"\u8fd9\u4e2a\u955c\u50cf\u5305\u542b\u4e86paddle\u76f8\u5173\u7684\u5f00\u53d1\u5de5\u5177\u4ee5\u53ca\u7f16\u8bd1\u548c\u8fd0\u884c\u73af\u5883":54,"\u8fd9\u4e2a\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u751f\u6210\u4e00\u7cfb\u5217\u6743\u91cd":64,"\u8fd9\u4e2a\u95ee\u9898\u662fpydataprovider\u8bfb\u6570\u636e\u65f6\u5019\u7684\u903b\u8f91\u95ee\u9898":2,"\u8fd9\u4e2a\u9759\u6001\u5e93\u5305\u542b\u4e86paddle\u7684\u5168\u90e8\u7b26\u53f7":34,"\u8fd9\u4e2adataprovider\u8f83\u590d\u6742":2,"\u8fd9\u4e2ainstance\u53ef\u4ee5\u662f\u5355\u4e2a\u503c":21,"\u8fd9\u4e2aissu":65,"\u8fd9\u4e2ajob\u624d\u7b97\u6210\u529f\u7ed3\u675f":80,"\u8fd9\u4e2alayer\u7684\u8f93\u51fa\u4f1a\u4f5c\u4e3a\u6574\u4e2a":63,"\u8fd9\u4e5f\u4f1a\u6781\u5927\u51cf\u5c11\u6570\u636e\u8bfb\u5165\u7684\u8017\u65f6":49,"\u8fd9\u4e9b":73,"\u8fd9\u4e9b\u51fd\u6570\u4f1a\u5c06\u5bf9\u5e94\u5185\u5bb9\u6dfb\u52a0\u5230":68,"\u8fd9\u4e9b\u5206\u5e03\u5f0f\u5b58\u50a8\u670d\u52a1\u901a\u5e38\u4f1a\u628a\u6570\u636e\u5207\u5272\u6210\u591a\u4e2a\u5206\u7247\u5206\u5e03\u5f0f\u7684\u5b58\u50a8\u5728\u591a\u4e2a\u8282\u70b9\u4e4b\u4e0a":21,"\u8fd9\u4e9b\u53c2\u6570\u7684\u5177\u4f53\u63cf\u8ff0":80,"\u8fd9\u4e9b\u53c2\u6570\u7684\u7b80\u77ed\u4ecb\u7ecd\u5982\u4e0b":91,"\u8fd9\u4e9b\u540d\u5b57\u5fc5\u987b\u8981\u5199\u5bf9":67,"\u8fd9\u4e9b\u6570\u636e\u4f1a\u88ab\u7528\u6765\u66f4\u65b0\u53c2\u6570":49,"\u8fd9\u4e9b\u6570\u636e\u4f7f\u7528\u7684\u5185\u5b58\u4e3b\u8981\u548c\u4e24\u4e2a\u53c2\u6570\u6709\u5173\u7cfb":49,"\u8fd9\u4e9b\u6587\u4ef6\u5c06\u4f1a\u88ab\u4fdd\u5b58\u5728":87,"\u8fd9\u4e9b\u6a21\u578b\u90fd\u662f\u7531\u539f\u4f5c\u8005\u63d0\u4f9b\u7684\u6a21\u578b":87,"\u8fd9\u4e9b\u7279\u5f81\u503c\u4e0e\u4e0a\u8ff0\u4f7f\u7528c":87,"\u8fd9\u4e9b\u7279\u5f81\u548c\u6807\u7b7e\u5b58\u50a8\u5728":92,"\u8fd9\u4e9b\u7279\u5f81\u6570\u636e\u4e4b\u95f4\u7684\u987a\u5e8f\u662f\u6709\u610f\u4e49\u7684":61,"\u8fd9\u4efd\u6559\u7a0b\u5c55\u793a\u4e86\u5982\u4f55\u5728paddlepaddle\u4e2d\u5b9e\u73b0\u4e00\u4e2a\u81ea\u5b9a\u4e49\u7684\u7f51\u7edc\u5c42":67,"\u8fd9\u4efd\u7b80\u77ed\u7684\u4ecb\u7ecd\u5c06\u5411\u4f60\u5c55\u793a\u5982\u4f55\u5229\u7528paddlepaddle\u6765\u89e3\u51b3\u4e00\u4e2a\u7ecf\u5178\u7684\u7ebf\u6027\u56de\u5f52\u95ee\u9898":52,"\u8fd9\u4f1a\u63d0\u793a\u5f53\u524d\u76ee\u5f55\u7684\u4e00\u4e9b\u53d8\u5316":66,"\u8fd9\u4f1a\u7ed9\u8bc4\u5ba1\u4eba\u5e26\u6765\u5f88\u5927\u56f0\u6270":66,"\u8fd9\u4f1a\u81ea\u52a8\u8fdb\u884c\u7f51\u7edc\u914d\u7f6e\u4e2d\u58f0\u660e\u7684\u6fc0\u6d3b\u64cd\u4f5c":67,"\u8fd9\u4f7f\u5f97nmt\u6a21\u578b\u5f97\u4ee5\u89e3\u653e\u51fa\u6765":94,"\u8fd9\u4fbf\u662f\u4e00\u79cd\u53cc\u5c42rnn\u7684\u8f93\u5165\u6570\u636e":61,"\u8fd9\u51e0\u4e2a\u7f16\u8bd1\u9009\u9879\u7684\u8bbe\u7f6e":53,"\u8fd9\u53e5\u8868\u793a\u4f7f\u7528\u57fa\u7c7b":68,"\u8fd9\u53ef\u4ee5\u5e2e\u60a8\u7701\u6389\u82b1\u4e00\u5c0f\u65f6\u5b89\u88c5\u548c\u914d\u7f6e\u5404\u79cd\u5f00\u53d1\u5de5\u5177":65,"\u8fd9\u53ef\u4ee5\u8ba9\u5176\u4ed6\u4eba\u77e5\u9053\u8fd9\u6b21\u63d0\u4ea4\u505a\u4e86\u54ea\u4e9b\u6539\u53d8":66,"\u8fd9\u53ef\u4ee5\u901a\u8fc7":66,"\u8fd9\u548c\u5355\u5c42rnn\u7684\u914d\u7f6e\u662f\u7b49\u4ef7\u7684":61,"\u8fd9\u56db\u4e2a\u7b80\u5355\u7684\u7279\u5f81\u662f\u6211\u4eec\u7684srl\u7cfb\u7edf\u6240\u9700\u8981\u7684":92,"\u8fd9\u56db\u6761\u6570\u636e\u540c\u65f6\u5904\u7406\u7684\u53e5\u5b50\u6570\u91cf\u4e3a":61,"\u8fd9\u5728\u5f88\u5927\u7a0b\u5ea6\u4e0a\u4f18\u4e8e\u5148\u524d\u7684\u6700\u5148\u8fdb\u7684\u7cfb\u7edf":92,"\u8fd9\u5728\u6784\u9020\u975e\u5e38\u590d\u6742\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u65f6\u662f\u6709\u7528\u7684":64,"\u8fd9\u5bf9\u4e8e\u901a\u5e38\u7684java\u7684\u5f00\u53d1\u8005\u6765\u8bf4":33,"\u8fd9\u5c06\u4f1a\u5bfc\u81f4\u5355\u5143\u6d4b\u8bd5\u51fa\u9519":68,"\u8fd9\u5c06\u4f1a\u5bfc\u81f4\u7f16\u8bd1\u51fa\u9519":68,"\u8fd9\u5c06\u82b1\u8d39\u6570\u5206\u949f\u7684\u65f6\u95f4":94,"\u8fd9\u5df2\u7ecf\u5728":93,"\u8fd9\u610f\u5473\u7740":64,"\u8fd9\u610f\u5473\u7740\u6a21\u578b\u5728\u8bad\u7ec3\u6570\u636e\u4e0a\u4e0d\u65ad\u7684\u6539\u8fdb":52,"\u8fd9\u610f\u5473\u7740\u9664\u4e86\u6307\u5b9adevic":77,"\u8fd9\u65f6":49,"\u8fd9\u65f6\u5728\u4f7f\u7528":51,"\u8fd9\u65f6\u8fdb\u884c\u77e9\u9635\u4e58\u6cd5\u8fd0\u7b97\u5c31\u53ef\u80fd\u5bfc\u81f4\u6d6e\u70b9\u6570\u6ea2\u51fa":49,"\u8fd9\u662f\u4e00\u4e2a\u57fa\u4e8e\u7edf\u8ba1\u7684\u673a\u5668\u7ffb\u8bd1\u7cfb\u7edf":93,"\u8fd9\u662f\u4e00\u79cd\u6309\u5df2\u8bad\u7ec3\u6837\u672c\u6570\u5206\u6bb5\u53d6\u503c\u7684\u5b66\u4e60\u7387\u9000\u706b\u65b9\u6cd5":51,"\u8fd9\u662f\u4e00\u79cd\u6309\u5df2\u8bad\u7ec3pass\u6570\u5206\u6bb5\u53d6\u503c\u7684\u5b66\u4e60\u7387\u9000\u706b\u65b9\u6cd5":51,"\u8fd9\u662f\u4e00\u79cd\u975e\u5e38\u7075\u6d3b\u7684\u6570\u636e\u7ec4\u7ec7\u65b9\u5f0f":60,"\u8fd9\u662f\u56e0\u4e3a":33,"\u8fd9\u662f\u56e0\u4e3a\u5b83\u53d1\u6398\u51fa\u4e86\u56fe\u7247\u7684\u4e24\u7c7b\u91cd\u8981\u4fe1\u606f":86,"\u8fd9\u662f\u5f00\u6e90\u793e\u533a\u7684\u57fa\u672c\u793c\u8c8c":66,"\u8fd9\u662f\u666e\u901a\u7684\u5355\u5c42\u65f6\u95f4\u5e8f\u5217\u7684dataprovider\u4ee3\u7801":61,"\u8fd9\u662f\u76ee\u524dcmake\u5bfb\u627epython\u7684\u903b\u8f91\u5b58\u5728\u7f3a\u9677":46,"\u8fd9\u662f\u96c6\u675f\u641c\u7d22\u7684\u7ed3\u679c":94,"\u8fd9\u6765\u81ea\u4e8epaddlepaddle\u7684\u5185\u5b58\u4e2d":94,"\u8fd9\u6837":[34,52,73],"\u8fd9\u6837\u4fdd\u8bc1":42,"\u8fd9\u6837\u4fdd\u8bc1\u8fd0\u884c\u7ed3\u675f\u4e4b\u540e\u7684":65,"\u8fd9\u6837\u505a\u53ef\u4ee5\u6781\u5927\u7684\u51cf\u5c11\u5185\u5b58\u5360\u7528":49,"\u8fd9\u6837\u53ef\u4ee5\u51cf\u5c0fgpu\u5185\u5b58":77,"\u8fd9\u6837\u5982\u679c\u9047\u5230\u95ee\u9898":65,"\u8fd9\u6837\u5bb9\u5668\u7684":80,"\u8fd9\u6837\u5c31\u4f1a\u751f\u6210\u4e24\u4e2a\u6587\u4ef6":91,"\u8fd9\u6837\u5c31\u53ef\u4ee5\u5728\u4e91\u7aef\u6267\u884c\u591a\u79cd\u6570\u636e\u7c7b\u8ba1\u7b97\u4efb\u52a1":21,"\u8fd9\u6837\u5df2\u7ecf\u4f20\u8f93\u6210\u529f\u7684\u90e8\u5206\u5c31\u4e0d\u7528\u91cd\u65b0\u4f20\u8f93\u4e86":28,"\u8fd9\u6837\u5f00\u53d1\u4eba\u5458\u53ef\u4ee5\u4ee5\u4e00\u81f4\u7684\u65b9\u5f0f\u5728\u4e0d\u540c\u7684\u5e73\u53f0\u4e0a\u5de5\u4f5c":54,"\u8fd9\u6837\u5f53\u8be5pull":66,"\u8fd9\u6837\u6211\u4eec\u53ef\u4ee5\u76f4\u63a5\u5728docker\u4e2d\u8fd0\u884c\u9700\u8981\u7684\u7a0b\u5e8f\u800c\u4e0d\u9700\u8981\u5b89\u88c5\u540e\u5728\u6267\u884c":54,"\u8fd9\u6837\u7684\u88c5\u9970\u5668":67,"\u8fd9\u6837\u7684\u8bdd":79,"\u8fd9\u6837\u7684\u8bdd\u6bcf\u4f4d\u7528\u6237\u5728\u6d4b\u8bd5\u6587\u4ef6\u4e2d\u5c06\u4e0e\u8bad\u7ec3\u6587\u4ef6\u542b\u6709\u540c\u6837\u7684\u4fe1\u606f":91,"\u8fd9\u6b63\u662f\u5b83\u4eec\u901f\u5ea6\u5feb\u7684\u539f\u56e0":72,"\u8fd9\u6bb5\u7b80\u77ed\u7684\u914d\u7f6e\u5c55\u793a\u4e86paddlepaddle\u7684\u57fa\u672c\u7528\u6cd5":52,"\u8fd9\u7528\u4e8e\u5728\u591a\u7ebf\u7a0b\u548c\u591a\u673a\u4e0a\u66f4\u65b0\u53c2\u6570":67,"\u8fd9\u79cd\u521d\u59cb\u5316\u65b9\u5f0f\u5728\u4e00\u822c\u60c5\u51b5\u4e0b\u4e0d\u4f1a\u4ea7\u751f\u5f88\u5dee\u7684\u7ed3\u679c":51,"\u8fd9\u79cd\u60c5\u51b5\u4e0b\u4e0d\u9700\u8981\u91cd\u5199\u8be5\u51fd\u6570":67,"\u8fd9\u79cd\u60c5\u51b5\u5e38\u5e38\u53d1\u751f\u5728":49,"\u8fd9\u79cd\u65b9\u5f0f\u5bf9\u5185\u5b58\u6d88\u8017\u8f83\u5927":50,"\u8fd9\u79cd\u65b9\u5f0f\u5fc5\u987b\u4f7f\u7528paddle\u5b58\u50a8\u7684\u6a21\u578b\u8def\u5f84\u683c\u5f0f":77,"\u8fd9\u79cd\u751f\u6210\u6280\u672f\u53ea\u7528\u4e8e\u7c7b\u4f3c\u89e3\u7801\u5668\u7684\u751f\u6210\u8fc7\u7a0b":64,"\u8fd9\u79cd\u7c7b\u578b\u7684\u8f93\u5165\u5fc5\u987b\u901a\u8fc7":63,"\u8fd9\u79cd\u96c6\u7fa4\u8282\u70b9\u7ba1\u7406\u65b9\u5f0f\u4f1a\u5728\u5c06\u6765\u4f7f\u7528":80,"\u8fd9\u7bc7\u6587\u6863":66,"\u8fd9\u7bc7\u6587\u6863\u4ecb\u7ecd\u57fa\u4e8e":65,"\u8fd9\u7bc7\u6587\u6863\u5c06\u4ee5linux":59,"\u8fd9\u7bc7\u6587\u7ae0":[65,94],"\u8fd9\u7ec4\u8bed\u4e49\u76f8\u540c\u7684\u793a\u4f8b\u914d\u7f6e\u5982\u4e0b":61,"\u8fd9\u901a\u8fc7\u83b7\u5f97\u53cd\u5411\u5faa\u73af\u7f51\u7edc\u7684\u7b2c\u4e00\u4e2a\u5b9e\u4f8b":64,"\u8fd9\u90fd\u9700\u8981\u8fd9\u4e2a\u63a5\u53e3\u6309\u7167\u7ea6\u5b9a\u4fd7\u6210\u7684\u89c4\u5219\u6765\u6ce8\u91ca\u5b8c\u5907":33,"\u8fd9\u91cc":[51,54,64,66,78,80,87,92],"\u8fd9\u91cc\u4e0d\u518d\u8d58\u8ff0":68,"\u8fd9\u91cc\u4ecb\u7ecd\u5982\u751f\u6210\u9020\u8fd9\u4e2a\u5f00\u53d1\u955c\u50cf":54,"\u8fd9\u91cc\u4ee5":89,"\u8fd9\u91cc\u4f7f\u7528\u4e00\u4e2a\u57fa\u4e8emomentum\u7684\u968f\u673a\u68af\u5ea6\u4e0b\u964d":52,"\u8fd9\u91cc\u4f7f\u7528\u4e86\u4e09\u79cd\u7f51\u7edc\u5355\u5143":52,"\u8fd9\u91cc\u4f7f\u7528\u4e86paddlepaddle\u7684python\u63a5\u53e3\u6765\u52a0\u8f7d\u6570\u6910":93,"\u8fd9\u91cc\u4f7f\u7528\u4e86paddlepaddle\u9884\u5b9a\u4e49\u597d\u7684rnn\u5904\u7406\u51fd\u6570":61,"\u8fd9\u91cc\u4f7f\u7528\u7b80\u5355\u7684":49,"\u8fd9\u91cc\u5229\u7528\u5b83\u5efa\u6a21\u53d8\u91cf\u4e4b\u95f4\u7684\u7ebf\u6027\u5173\u7cfb":52,"\u8fd9\u91cc\u53ea\u52a0\u8f7d":94,"\u8fd9\u91cc\u53ea\u7b80\u5355\u4ecb\u7ecd\u4e86\u5355\u673a\u8bad\u7ec3":89,"\u8fd9\u91cc\u5c06\u4ecb\u7ecdpaddlepaddle\u7684\u57fa\u672c\u4f7f\u7528\u6982\u5ff5":56,"\u8fd9\u91cc\u5c55\u793a\u5982\u4f55\u4f7f\u7528\u89c2\u6d4b\u6570\u636e\u6765\u62df\u5408\u8fd9\u4e00\u7ebf\u6027\u5173\u7cfb":52,"\u8fd9\u91cc\u6211\u4eec\u4f7f\u7528":91,"\u8fd9\u91cc\u6211\u4eec\u4f7f\u7528\u7684\u662f\u4e00\u4e2a\u5c0f\u7684vgg\u7f51\u7edc":86,"\u8fd9\u91cc\u6211\u4eec\u4f7f\u7528\u7684\u662fgpu\u6a21\u5f0f\u8fdb\u884c\u8bad\u7ec3":86,"\u8fd9\u91cc\u6211\u4eec\u5728movielens\u6570\u636e\u96c6\u63cf\u8ff0\u4e00\u79cd":91,"\u8fd9\u91cc\u6211\u4eec\u5c55\u793a\u4e00\u4efd\u7b80\u5316\u8fc7\u7684\u4ee3\u7801":67,"\u8fd9\u91cc\u6211\u4eec\u7528paddl":54,"\u8fd9\u91cc\u6211\u4eec\u901a\u8fc7\u5728kubernetes\u96c6\u7fa4\u4e0a\u542f\u52a8\u4e00\u4e2ajob\u6765\u4e0b\u8f7d\u5e76\u5207\u5272\u6570\u636e":80,"\u8fd9\u91cc\u6307\u5b9a\u8bcd\u5178":89,"\u8fd9\u91cc\u6570\u636e\u5c42\u6709\u4e24\u4e2a":52,"\u8fd9\u91cc\u662f\u4e00\u4e2a\u4f8b\u5b50":94,"\u8fd9\u91cc\u6709\u4e00\u4e9b\u4e0d\u540c\u7684\u53c2\u6570\u9700\u8981\u6307\u5b9a":94,"\u8fd9\u91cc\u6709\u4e24\u79cd\u6709\u6548\u7684\u89e3\u51b3\u65b9\u6cd5":49,"\u8fd9\u91cc\u68c0\u9a8c\u8fd0\u884c\u65f6\u95f4\u6a21\u578b\u7684\u6536\u655b":73,"\u8fd9\u91cc\u6bcf\u4e2a5\u4e2abatch\u6253\u5370\u4e00\u4e2a\u70b9":94,"\u8fd9\u91cc\u6bcf\u9694100\u4e2abatch\u663e\u793a\u4e00\u6b21\u53c2\u6570\u7edf\u8ba1\u4fe1\u606f":94,"\u8fd9\u91cc\u6bcf\u969410\u4e2abatch\u6253\u5370\u4e00\u6b21\u65e5\u5fd7":94,"\u8fd9\u91cc\u7684\u5217\u51fa\u7684\u548c":86,"\u8fd9\u91cc\u7684eigentensor\u4e4b\u95f4\u7684\u8fd0\u7b97\u53ea\u662f\u6539\u53d8\u4e86\u539f\u6709tensor\u4e2d\u7684\u6570\u636e":69,"\u8fd9\u91cc\u76f4\u63a5\u901a\u8fc7\u9884\u6d4b\u811a\u672c":89,"\u8fd9\u91cc\u7ed9\u51fa\u96c6\u4e2d\u5e38\u89c1\u7684\u90e8\u7f72\u65b9\u6cd5":78,"\u8fd9\u91cc\u8bbe\u7f6e\u4e3a\u4f7f\u7528cpu":94,"\u8fd9\u91cc\u8bbe\u7f6e\u4e3afals":94,"\u8fd9\u91cc\u8bbe\u7f6e\u4e3atrue":94,"\u8fd9\u91cc\u91c7\u7528adam\u4f18\u5316\u65b9\u6cd5":89,"\u8fd9\u91cc\u9700\u8981\u7528\u6237\u989d\u5916\u6ce8\u610f":20,"\u8fdb\u4e3b\u4ed3\u5e93\u540e":66,"\u8fdb\u5165":93,"\u8fdb\u5165\u5bb9\u5668":79,"\u8fdb\u5165\u5f00\u53d1\u955c\u50cf\u5e76\u5f00\u59cb\u5de5\u4f5c":54,"\u8fdb\u7a0b":54,"\u8fdb\u7a0b\u542f\u52a8\u7684\u5fc5\u8981\u53c2\u6570":80,"\u8fdb\u7a0b\u7684":73,"\u8fdb\u7a0b\u7684\u542f\u52a8\u53c2\u6570":80,"\u8fdb\u7a0b\u7684\u8fd0\u884c\u73af\u5883":80,"\u8fdb\u7a0b\u9700\u8981\u7684":80,"\u8fdb\u800c\u8fdb\u884c\u4ee3\u7801\u8bc4\u5ba1":42,"\u8fdb\u884c\u4e86":61,"\u8fdb\u884c\u4f7f\u7528":86,"\u8fdb\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3\u7684\u65b9\u6848":80,"\u8fdb\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3\u7684\u65b9\u6cd5":80,"\u8fdb\u884c\u56de\u590d":66,"\u8fdb\u884c\u5f00\u53d1":66,"\u8fdb\u884c\u62c6\u89e3":61,"\u8fdb\u884c\u6fc0\u6d3b\u64cd\u4f5c":67,"\u8fdb\u884c\u8bbe\u7f6e":68,"\u8fdb\u884c\u8bfb\u5165\u548c\u9884\u5904\u7406\u4ece\u800c\u5f97\u5230\u771f\u5b9e\u8f93\u5165":52,"\u8fdb\u884c\u9884\u6d4b":89,"\u8fdb\u9636\u6307\u5357":[56,83],"\u8fde\u63a5":63,"\u8fde\u63a5\u4e09\u4e2alstm\u9690\u85cf\u5c42":93,"\u9000\u4f11\u4eba\u5458":90,"\u9000\u51fa\u5bb9\u5668":79,"\u9002\u4e2d":61,"\u9002\u5408\u81ea\u5df1\u5b9e\u9a8c":54,"\u9009":61,"\u9009\u62e9":61,"\u9009\u62e9\u6d4b\u8bd5\u7ed3\u679c\u6700\u597d\u7684\u6a21\u578b\u6765\u9884\u6d4b":89,"\u9009\u62e9\u76ee\u6807\u5206\u652f":66,"\u9009\u62e9\u8def\u5f84\u6765\u52a8\u6001\u52a0\u8f7dnvidia":75,"\u9009\u62e9\u8fc7\u540e\u7684":94,"\u9009\u62e9\u9002\u5408\u60a8\u7684\u573a\u666f\u7684\u5408\u9002\u65b9\u6848":78,"\u9009\u81ea\u4e0b\u5217\u7c7b\u578b":90,"\u9009\u9879":[53,65,85],"\u900f\u4f20\u7528\u6237\u8eab\u4efd\u7684\u529e\u6cd5":28,"\u9012\u5f52\u795e\u7ecf\u7f51\u7edc":74,"\u901a\u5e38":[34,68,73,93],"\u901a\u5e38\u4f1a\u4f7f\u7528\u73af\u5883\u53d8\u91cf\u914d\u7f6ejob\u7684\u914d\u7f6e\u4fe1\u606f":80,"\u901a\u5e38\u4f7f\u7528\u7a00\u758f\u8bad\u7ec3\u6765\u52a0\u901f\u8ba1\u7b97\u8fc7\u7a0b":77,"\u901a\u5e38\u505a\u6cd5\u662f\u4ece\u4e00\u4e2a\u6bd4\u8f83\u5927\u7684learning_rate\u5f00\u59cb\u8bd5":51,"\u901a\u5e38\u540d\u5b57\u662f":66,"\u901a\u5e38\u60c5\u51b5\u4e0b":72,"\u901a\u5e38\u6211\u4eec\u4f1a\u5b89\u88c5ceph\u7b49\u5206\u5e03\u5f0f\u6587\u4ef6\u7cfb\u7edf\u6765\u5b58\u50a8\u8bad\u7ec3\u6570\u636e":79,"\u901a\u5e38\u6307\u5c06\u4e00\u4e2a\u6574\u4f53\u62c6\u5206\u6210\u591a\u4efd\u7684\u5176\u4e2d\u7684\u4e00\u4efd":20,"\u901a\u5e38\u6bcf\u4e2ajob\u5305\u62ec\u4e00\u4e2a\u6216\u8005\u591a\u4e2apod":78,"\u901a\u5e38\u7684\u505a\u6cd5\u662f\u4f7f\u7528":64,"\u901a\u5e38\u7684\u505a\u6cd5\u662f\u5c06\u914d\u7f6e\u5b58\u4e8e":67,"\u901a\u5e38\u8981\u6c42\u65f6\u95f4\u6b65\u4e4b\u95f4\u5177\u6709\u4e00\u4e9b\u4f9d\u8d56\u6027":61,"\u901a\u5e38\u90fd\u4f1a\u4f7f\u7528\u4e0b\u9762\u8fd9\u4e9b\u547d\u4ee4\u884c\u53c2\u6570":77,"\u901a\u7528":74,"\u901a\u77e5":61,"\u901a\u8fc7":[49,61,66,67,68,73,89],"\u901a\u8fc7\u4e24\u4e2a\u5d4c\u5957\u7684":63,"\u901a\u8fc7\u4ea4\u53c9\u7f16\u8bd1\u7684\u65b9\u5f0f\u6765\u7f16\u8bd1":59,"\u901a\u8fc7\u4ea4\u66ff\u4f7f\u7528\u5377\u79ef\u548c\u6c60\u5316\u5904\u7406":86,"\u901a\u8fc7\u4f7f\u7528":53,"\u901a\u8fc7\u51fd\u6570":80,"\u901a\u8fc7\u5377\u79ef\u64cd\u4f5c\u4ece\u56fe\u7247\u6216\u7279\u5f81\u56fe\u4e2d\u63d0\u53d6\u7279\u5f81":86,"\u901a\u8fc7\u547d\u4ee4\u884c\u53c2\u6570":49,"\u901a\u8fc7\u5f15\u7528memory\u5f97\u5230\u8fd9\u4e2alayer\u4e0a\u4e00\u4e2a\u65f6\u523b\u7684\u8f93\u51fa":63,"\u901a\u8fc7\u5f15\u7528memory\u5f97\u5230\u8fd9\u4e2alayer\u4e0a\u4e00\u4e2a\u65f6\u523b\u8f93\u51fa":63,"\u901a\u8fc7\u6240\u6709\u5355\u5143\u6d4b\u8bd5":66,"\u901a\u8fc7\u6240\u6709\u8bad\u7ec3\u96c6\u4e00\u6b21\u79f0\u4e3a\u4e00\u904d":93,"\u901a\u8fc7\u67e5\u770b\u4e70\u5bb6\u5bf9\u67d0\u4e2a\u4ea7\u54c1\u7684\u8bc4\u4ef7\u53cd\u9988":89,"\u901a\u8fc7\u6a21\u578b\u63a8\u65adapi\u7684\u5b9e\u73b0\u4f5c\u4e3a\u4e00\u4e2a\u6837\u4f8b":34,"\u901a\u8fc7\u7ec4\u5408\u4e0d\u540c\u7684layer":56,"\u901a\u8fc7\u7f16\u8bd1\u4f1a\u751f\u6210py_paddle\u8f6f\u4ef6\u5305":4,"\u901a\u8fc7\u7f51\u7edc\u5c42\u7684\u6807\u8bc6\u7b26\u6765\u6307\u5b9a":67,"\u901a\u8fc7\u8be5\u53c2\u6570\u53ef\u83b7\u53d6\u5230\u8f93\u5165\u8f93\u51fa\u4ee5\u53ca\u5c5e\u6027":68,"\u901a\u8fc7\u8c03\u7528":4,"\u901a\u8fc7\u914d\u7f6e\u7c7b\u4f3c\u4e8e":89,"\u901a\u8fc7data":63,"\u901a\u8fc7volum":78,"\u903b\u8f91\u5212\u4e0a\u6587\u4ef6\u5206\u5757\u7684\u5355\u4f4d":28,"\u903b\u8f91\u56de\u5f52":89,"\u9047\u5230\u8be5\u9519\u8bef\u65f6":50,"\u9053\u6b49":61,"\u9069":61,"\u9075\u5b88\u4ee5\u4e0b\u7ea6\u5b9a":66,"\u9075\u5faa\u4ee5\u4e0b\u6d41\u7a0b":42,"\u9075\u5faa\u5982\u4e0b\u7684\u683c\u5f0f":90,"\u9075\u5faa\u6587\u7ae0":85,"\u90a3\u4e48":[34,63,67],"\u90a3\u4e480\u5c42\u5e8f\u5217\u5373\u4e3a\u4e00\u4e2a\u8bcd\u8bed":63,"\u90a3\u4e48\u53ef\u4ee5\u8ba4\u4e3a\u8bad\u7ec3\u4e0d\u6536\u655b":51,"\u90a3\u4e48\u5728":68,"\u90a3\u4e48\u5728\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u4e0d\u4f1a\u6267\u884c\u6d4b\u8bd5\u64cd\u4f5c":1,"\u90a3\u4e48\u5982\u4f55\u5224\u65ad\u8bad\u7ec3\u4e0d\u6536\u655b\u5462":51,"\u90a3\u4e48\u5e38\u6570\u8f93\u51fa\u6240\u80fd\u8fbe\u5230\u7684\u6700\u5c0fcost\u662f":51,"\u90a3\u4e48\u5f53check\u51fa\u6570\u636e\u4e0d\u5408\u6cd5\u65f6":2,"\u90a3\u4e48\u6211\u4eec\u4e5f\u5c31\u4e0d\u9700\u8981":65,"\u90a3\u4e48\u6211\u4eec\u53ef\u4ee5\u5224\u65ad\u4e3a\u8bad\u7ec3\u4e0d\u6536\u655b":51,"\u90a3\u4e48\u6211\u4eec\u53ef\u4ee5\u901a\u8fc7\u6536\u96c6\u5e02\u573a\u4e0a\u623f\u5b50\u7684\u5927\u5c0f\u548c\u4ef7\u683c":52,"\u90a3\u4e48\u63a8\u8350\u4f7f\u7528":64,"\u90a3\u4e48\u63a8\u8350\u4f7f\u7528\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u65b9\u6cd5":64,"\u90a3\u4e48\u6536\u655b\u53ef\u80fd\u5f88\u6162":51,"\u90a3\u4e48\u6700\u597d\u5c06\u6570\u636e\u6587\u4ef6\u5728\u6bcf\u6b21\u8bfb\u53d6\u4e4b\u524d\u505a\u4e00\u6b21shuffl":49,"\u90a3\u4e48\u7528\u6237\u9700\u8981\u62c9\u53d6\u6240\u6709\u7684\u8fdc\u7a0b\u5206\u652f\u5230\u672c\u673a":46,"\u90a3\u4e48\u7f16\u8bd1\u8fc7\u7a0b\u53ea\u4f1a\u4ea7\u751f":65,"\u90a3\u4e48\u8bad\u7ec3\u6709\u53ef\u80fd\u4e0d\u6536\u655b":51,"\u90a3\u4e48\u8be5\u4f18\u5316\u7b97\u6cd5\u81f3\u5c11\u9700\u8981":49,"\u90a3\u4e48fc1\u548cfc2\u5c42\u5c06\u4f1a\u4f7f\u7528\u7b2c1\u4e2agpu\u6765\u8ba1\u7b97":77,"\u90a3\u4e48paddlepaddle\u4f1a\u6839\u636elayer\u7684\u58f0\u660e\u987a\u5e8f":2,"\u90a3\u4e5f\u5c31\u4e0d\u9700\u8981\u6025\u7740\u4f18\u5316\u6027\u80fd\u5566":72,"\u90a3\u4f30\u8ba1\u8fd9\u91cc\u7684\u6f5c\u529b\u5c31\u6ca1\u5565\u597d\u6316\u7684\u4e86":72,"\u90a3\u51cf\u5c11\u5b66\u4e60\u738710\u500d\u7ee7\u7eed\u8bd5\u9a8c":51,"\u90a3\u5bb9\u5668\u5c31\u662f\u8fd9\u4e2a\u7a0b\u5e8f\u8fd0\u884c\u65f6\u4ea7\u751f\u7684":54,"\u90a3\u6211\u4f1a\u671f\u671b\u5206\u6790\u5de5\u5177\u7edf\u8ba1\u5230\u901f\u5ea6\u662f100gb":72,"\u90a3\u7a0b\u5e8f\u5206\u6790\u5de5\u5177\u662f\u5fc5\u4e0d\u53ef\u5c11\u7684\u5229\u5668":72,"\u90ae\u7f16":90,"\u90e8\u7f72\u548c\u914d\u7f6e\u6bd4\u8f83\u7b80\u5355":78,"\u90fd":61,"\u90fd\u4e0d\u9700\u8981":65,"\u90fd\u4f1a\u4ea7\u751f\u5f53\u524d\u5c42\u72b6\u6001\u7684\u6240\u6709\u7ee7\u627f\u7ed3\u679c":75,"\u90fd\u4f7f\u7528\u968f\u673a\u503c\u521d\u59cb\u5316":52,"\u90fd\u53ea\u662f\u4ecb\u7ecd\u53cc\u5c42rnn\u7684api\u63a5\u53e3":61,"\u90fd\u53ef\u4ee5\u8fd0\u884c":65,"\u90fd\u662f\u4e94\u4f4d\u7684\u6570\u5b57":21,"\u90fd\u662f\u5bf9layer1\u5143\u7d20\u7684\u62f7\u8d1d":60,"\u90fd\u662f\u5c06\u6bcf\u4e00\u53e5\u5206\u597d\u8bcd\u540e\u7684\u53e5\u5b50":61,"\u90fd\u662fabi\u8c03\u7528\u6807\u51c6\u7684":33,"\u90fd\u7528":66,"\u90fd\u9700\u8981\u5199\u63d0\u4ea4\u8bf4\u660e":66,"\u90fd\u9700\u8981\u8c03\u7528\u4e00\u6b21":67,"\u914d\u7f6e":93,"\u914d\u7f6e\u4e86\u7f51\u7edc":91,"\u914d\u7f6e\u521b\u5efa\u5b8c\u6bd5\u540e":86,"\u914d\u7f6e\u5982\u4e0b":85,"\u914d\u7f6e\u6253\u5f00":72,"\u914d\u7f6e\u6587\u4ef6":89,"\u914d\u7f6e\u6587\u4ef6\u63a5\u53e3\u662ffc_layer":67,"\u914d\u7f6e\u6587\u4ef6\u91cc\u52a0\u4e24\u884c":65,"\u914d\u7f6e\u6a21\u578b\u6587\u4ef6":85,"\u914d\u7f6e\u7684\u65b9\u6cd5\u53c2\u8003":28,"\u914d\u7f6e\u7b49\u6587\u4ef6\u7684\u76ee\u5f55\u89c6\u4e3a":73,"\u914d\u7f6e\u7b80\u5355\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u7684\u4f8b\u5b50":64,"\u914d\u7f6e\u7f51\u7edc\u5c42\u7684\u8f93\u5165":67,"\u914d\u7f6eapi":60,"\u9152\u5e97":61,"\u91c7\u6837\u5c42":91,"\u91c7\u7528":92,"\u91c7\u7528\u53e6\u4e00\u79cd\u65b9\u6cd5\u6765\u5806\u53e0lstm\u5c42":92,"\u91c7\u7528\u5747\u5300\u5206\u5e03\u6216\u8005\u9ad8\u65af\u5206\u5e03\u521d\u59cb\u5316":75,"\u91c7\u7528multi":51,"\u91ca\u653e\u5bf9paramters\u5185\u5b58\u7684\u9501\u5b9a":20,"\u91cc":65,"\u91cc\u4ecb\u7ecd\u4e86\u7528paddle\u6e90\u7801\u4e2d\u7684\u811a\u672c\u4e0b\u8f7d\u8bad\u7ec3\u6570\u636e\u7684\u8fc7\u7a0b":79,"\u91cc\u53ef\u4ee5\u6807\u51c6\u5316\u7f16\u8bd1\u73af\u5883":65,"\u91cc\u5b8c\u6210":68,"\u91cc\u6240\u6709\u7684\u7b26\u53f7\u90fd\u5199\u5165\u81ea\u5df1\u7684\u7a0b\u5e8f\u7684\u4e8c\u8fdb\u5236\u6587\u4ef6\u91cc":33,"\u91cc\u6307\u5b9a\u56fe\u50cf\u6570\u636e\u5217\u8868":87,"\u91cc\u7684":65,"\u91cc\u7684\u65e5\u5fd7":73,"\u91cc\u7684\u6e90\u7801":65,"\u91cc\u8fd0\u884c\u7684\u7f16\u8bd1\u5de5\u5177\u5b9e\u9645\u4e0a\u90fd\u662f\u5728\u672c\u673a\u7684":65,"\u91cc\u9762":68,"\u91cc\u9762\u6db5\u76d6\u4e86":54,"\u91cc\u9762\u6db5\u76d6\u4e86paddlepaddle\u8fd0\u884c\u6240\u9700\u7684\u6240\u6709\u73af\u5883":54,"\u91cd\u547d\u540d\u6210":33,"\u91cd\u65b0\u7f16\u8bd1paddlepaddl":72,"\u9488\u5bf9\u4e0d\u540c\u5e73\u53f0\u7684\u7f16\u8bd1\u5de5\u5177":59,"\u9488\u5bf9\u4efb\u52a1\u8fd0\u884c\u5b8c\u6210\u540e\u5bb9\u5668\u81ea\u52a8\u9000\u51fa\u7684\u573a\u666f":79,"\u9488\u5bf9\u5185\u5b58\u548c\u663e\u5b58":49,"\u9488\u5bf9\u6587\u672c":91,"\u94fe\u63a5\u4f55\u79cdblas\u5e93\u7b49":53,"\u94fe\u63a5\u5230\u81ea\u5df1\u7684\u7a0b\u5e8f\u91cc":33,"\u94fe\u63a5\u5f85\u8865\u5145":89,"\u9500\u552e":90,"\u9519\u8bef\u5904\u7406":33,"\u9519\u8bef\u5904\u7406\u65b9\u5f0f\u662f\u8fd4\u56de\u503c":33,"\u9519\u8bef\u5904\u7406\u7684\u65b9\u5f0f\u4e5f\u4e0d\u5c3d\u76f8\u540c":33,"\u9519\u8bef\u7387":89,"\u9519\u8bef\u7684define_py_data_sources2\u7c7b\u4f3c":51,"\u952e\u6765\u542f\u52a8\u7f16\u8bd1\u4e86":65,"\u955c\u50cf":54,"\u955c\u50cf\u91cc\u6709":79,"\u957f\u5ea6":49,"\u95e8\u63a7\u5faa\u73af\u5355\u5143\u5355\u6b65\u51fd\u6570\u548c\u8f93\u51fa\u51fd\u6570":64,"\u95e8\u63a7\u5faa\u73af\u5355\u5143\u7684\u8f93\u51fa\u88ab\u7528\u4f5c\u8f93\u51famemori":64,"\u95ee\u9898":52,"\u95f4\u9694":89,"\u9650\u5236\u5957\u63a5\u5b57\u53d1\u9001\u7f13\u51b2\u533a\u7684\u5927\u5c0f":75,"\u9650\u5236\u5957\u63a5\u5b57\u63a5\u6536\u7f13\u51b2\u533a\u7684\u5927\u5c0f":75,"\u9664\u4e86":2,"\u9664\u4e86boot_lay":61,"\u9664\u53bbdata\u5c42":89,"\u9664\u6784\u9020\u67d0\u79cd\u7c7b\u578b\u7684\u51fd\u6570":34,"\u9664\u6b64\u4e4b\u5916":49,"\u9664\u8bcd\u5411\u91cf\u6a21\u578b\u5916\u7684\u53c2\u6570\u5c06\u4f7f\u7528\u6b63\u6001\u5206\u5e03\u968f\u673a\u521d\u59cb\u5316":85,"\u9664\u96f6\u7b49\u95ee\u9898":49,"\u968f\u540e\u53ef\u4ee5\u7528\u8fd9\u4e2a\u5f00\u53d1\u955c\u50cf\u5f00\u59cbbuild":66,"\u968f\u673a\u521d\u59cb\u4e0d\u5b58\u5728\u7684\u53c2\u6570":92,"\u968f\u673a\u6570\u7684\u79cd\u5b50":75,"\u968f\u673a\u6570seed":74,"\u968f\u7740\u8f6e\u6570\u589e\u52a0\u8bef\u5dee\u4ee3\u4ef7\u51fd\u6570\u7684\u8f93\u51fa\u5728\u4e0d\u65ad\u7684\u51cf\u5c0f":52,"\u9694\u5f00":87,"\u96c6":90,"\u96c6\u675f\u641c\u7d22\u4e2d\u7684\u6269\u5c55\u5e7f\u5ea6":94,"\u96c6\u675f\u641c\u7d22\u4f7f\u7528\u5e7f\u5ea6\u4f18\u5148\u641c\u7d22\u6765\u6784\u5efa\u641c\u7d22\u6811":94,"\u96c6\u675f\u641c\u7d22\u4f7f\u7528\u5e7f\u5ea6\u4f18\u5148\u641c\u7d22\u7684\u65b9\u5f0f\u6784\u5efa\u67e5\u627e\u6811":75,"\u96c6\u7fa4\u4e0a\u542f\u52a8\u4e00\u4e2a\u5355\u673a\u4f7f\u7528cpu\u7684paddle\u8bad\u7ec3\u4f5c\u4e1a":79,"\u96c6\u7fa4\u4f5c\u4e1a\u4e2d\u6240\u6709\u8fdb\u7a0b\u7684\u73af\u5883\u8bbe\u7f6e":73,"\u96c6\u7fa4\u4f5c\u4e1a\u5c06\u4f1a\u5728\u51e0\u79d2\u540e\u542f\u52a8":73,"\u96c6\u7fa4\u5de5\u4f5c":73,"\u96c6\u7fa4\u6d4b\u8bd5":74,"\u96c6\u7fa4\u8bad\u7ec3":74,"\u96c6\u7fa4\u8bad\u7ec3\u4e0e\u9884\u6d4b":48,"\u96c6\u7fa4\u8fdb\u7a0b":73,"\u96c6\u7fa4\u901a\u4fe1\u4fe1\u9053\u7684\u7aef\u53e3\u6570":73,"\u96c6\u7fa4\u901a\u4fe1\u901a\u9053\u7684":73,"\u96c6\u7fa4\u901a\u4fe1\u901a\u9053\u7684\u7aef\u53e3\u53f7":73,"\u9700\u52a0\u8be5\u6a21\u677f\u53c2\u6570":68,"\u9700\u5728nvvp\u754c\u9762\u4e2d\u9009\u4e0a\u624d\u80fd\u5f00\u542f":72,"\u9700\u8981":[21,65,68],"\u9700\u8981\u4e3a":68,"\u9700\u8981\u4f7f\u7528\u5176\u5236\u5b9a\u7684\u65b9\u5f0f\u6302\u8f7d\u540e\u5e76\u5bfc\u5165\u6570\u636e":80,"\u9700\u8981\u4fdd\u6301\u5f53\u524d\u5206\u652f\u76ee\u5f55":66,"\u9700\u8981\u4fee\u6539build":42,"\u9700\u8981\u5148\u6302\u8f7d\u5230\u670d\u52a1\u5668node\u4e0a\u518d\u901a\u8fc7kubernet":78,"\u9700\u8981\u5347\u7ea7pip\u7248\u672c\u5230\u6700\u65b0":46,"\u9700\u8981\u53ef\u4ee5\u8de8\u5e73\u53f0\u6267\u884c":28,"\u9700\u8981\u540c\u6b65\u539f\u4ed3\u5e93":66,"\u9700\u8981\u542f\u52a8\u7684\u8282\u70b9\u4e2a\u6570\u4ee5\u53ca":80,"\u9700\u8981\u548c\u8be5op\u7684\u540d\u5b57\u4e00\u6837":68,"\u9700\u8981\u5728":[68,73],"\u9700\u8981\u5728\u521b\u5efa\u5bb9\u5668\u524d\u6302\u8f7d\u5377\u4ee5\u4fbf\u6211\u4eec\u4fdd\u5b58\u8bad\u7ec3\u7ed3\u679c":79,"\u9700\u8981\u5728\u7cfb\u7edf\u91cc\u5148\u5b89\u88c5\u597ddocker\u5de5\u5177\u5305":70,"\u9700\u8981\u5728cmake\u7684\u65f6\u5019":34,"\u9700\u8981\u5b89\u88c5graphviz\u6765\u8f6c\u6362dot\u6587\u4ef6\u4e3a\u56fe\u7247":87,"\u9700\u8981\u5bf9":78,"\u9700\u8981\u5c06\u5176parameter\u8bbe\u7f6e\u6210":49,"\u9700\u8981\u5c06\u6807\u8bb0\u7684\u8bad\u7ec3\u548c\u6d4b\u8bd5\u6837\u672c\u79fb\u52a8\u5230\u53e6\u4e00\u4e2a\u8def\u5f84":93,"\u9700\u8981\u5c06bugfix\u7684\u5206\u652f\u540c\u65f6merge\u5230":42,"\u9700\u8981\u5f15\u7528":34,"\u9700\u8981\u5f3a\u8c03\u7684\u662f":65,"\u9700\u8981\u6307\u5b9a\u4e0e\u67d0\u4e00\u4e2a\u8f93\u5165\u7684\u5e8f\u5217\u4fe1\u606f\u662f\u4e00\u81f4\u7684":61,"\u9700\u8981\u6307\u5b9alayer\u7684\u8f93\u5165\u6765\u6e90":56,"\u9700\u8981\u660e\u786e\u6307\u5b9a":75,"\u9700\u8981\u6709\u4e00\u4e2a\u5916\u90e8\u7684\u5b58\u50a8\u670d\u52a1\u6765\u4fdd\u5b58\u8bad\u7ec3\u6240\u9700\u6570\u636e\u548c\u8bad\u7ec3\u8f93\u51fa":78,"\u9700\u8981\u6709\u7a33\u5b9a\u7684\u5bfc\u51fa\u7b26\u53f7":33,"\u9700\u8981\u6839\u636e\u4e0d\u540c\u7684\u5206\u5e03\u5f0f\u5b58\u50a8\u6765\u7ed1\u5b9a\u4e00\u4e2a":80,"\u9700\u8981\u6ce8\u610f":68,"\u9700\u8981\u6ce8\u610f\u7684\u662f":[42,49,75,91],"\u9700\u8981\u6ce8\u610f\u7684\u662f\u68af\u5ea6\u68c0\u67e5\u4ec5\u4ec5\u9a8c\u8bc1\u4e86\u68af\u5ea6\u7684\u8ba1\u7b97":67,"\u9700\u8981\u6ce8\u610f\u7684\u662fpaddlepaddle\u76ee\u524d\u53ea\u652f\u6301\u5b50\u5e8f\u5217\u6570\u76ee\u4e00\u6837\u7684\u591a\u8f93\u5165\u53cc\u5c42rnn":61,"\u9700\u8981\u7528\u5230\u7684\u7f16\u8bd1\u5de5\u5177\u548c\u7cfb\u7edf\u5e93":58,"\u9700\u8981\u7528\u6237\u663e\u5f0f\u8bbe\u5b9a":50,"\u9700\u8981\u88ab\u66b4\u9732\u5230\u5176\u4ed6\u8bed\u8a00":34,"\u9700\u8981\u8bf7\u7ba1\u7406\u5458\u5b89\u88c5\u548c\u914d\u7f6e\u597d":65,"\u9700\u8981\u9075\u5faa\u4ee5\u4e0b\u7ea6\u5b9a":63,"\u9700\u8981\u91cd\u547d\u540dwheel\u5305\u4e2dplatform\u76f8\u5173\u7684\u540e\u7f00":42,"\u9700\u8981\u989d\u5916\u6ce8\u610f\u7684\u662f":69,"\u975e\u5e38\u6570":67,"\u975e\u96f6\u6570\u5b57\u7684\u4e2a\u6570":67,"\u9762\u5411\u67b6\u6784\u4e3a32\u4f4darm\u67b6\u6784":58,"\u97f3\u4e50\u5267":90,"\u9879\u76ee\u5728\u52aa\u529b\u5f00\u59cb\u652f\u6301\u5176\u4ed6\u4e0d\u9700\u8981":65,"\u987a\u5e8f":61,"\u9884\u5904\u7406\u6570\u636e\u4e00\u822c\u7684\u547d\u4ee4\u4e3a":91,"\u9884\u5904\u7406\u811a\u672c":93,"\u9884\u5b9a\u4e49\u7f51\u7edc":93,"\u9884\u5b9a\u4e49\u7f51\u7edc\u5982\u56fe3\u6240\u793a":93,"\u9884\u63d0\u4ea4\u94a9\u5b50":66,"\u9884\u6d4b\u540e":92,"\u9884\u6d4b\u63a5\u53e3\u811a\u672c":93,"\u9884\u6d4b\u6982\u7387\u53d6\u5e73\u5747":87,"\u9884\u6d4b\u7a0b\u5e8f\u5c06\u8bfb\u53d6\u7528\u6237\u7684\u8f93\u5165":91,"\u9884\u6d4b\u7ed3\u679c\u4ee5\u6587\u672c\u7684\u5f62\u5f0f\u4fdd\u5b58\u5728":89,"\u9884\u6d4b\u811a\u672c\u662f":92,"\u9884\u6d4bid":89,"\u9884\u6d4bimdb\u7684\u672a\u6807\u8bb0\u8bc4\u8bba\u7684\u4e00\u4e2a\u5b9e\u4f8b\u5982\u4e0b":93,"\u9884\u8bad\u7ec3\u6a21\u578b\u4f7f\u7528\u7684\u5b57\u5178\u7684\u8def\u5f84":85,"\u9884\u8bad\u7ec3\u8bcd\u5411\u91cf\u5b57\u5178\u6a21\u578b\u7684\u8def\u5f84":85,"\u989c\u8272\u901a\u9053\u987a\u5e8f\u4e3a":87,"\u989d\u5916\u7684\u53c2\u6570":89,"\u9996\u5148":[2,52,61,64,67,85,87,89,92,93],"\u9996\u5148\u4e0b\u8f7dcifar":86,"\u9996\u5148\u5728\u7cfb\u7edf\u8def\u5f84":53,"\u9996\u5148\u5b89\u88c5\u5e76\u5728\u5f53\u524d\u76ee\u5f55\u8fd0\u884c\u5b83":66,"\u9996\u5148\u5b89\u88c5paddlepaddl":93,"\u9996\u5148\u5b9a\u4e49":68,"\u9996\u5148\u5bf9\u8f93\u5165\u505a\u4e00\u4e2a\u5c0f\u7684\u6270\u52a8":67,"\u9996\u5148\u6211\u4eec\u9700\u8981\u63a8\u5bfc\u8be5\u7f51\u7edc\u5c42\u7684":67,"\u9996\u5148\u662f\u6cd5\u8bed\u5e8f\u5217":94,"\u9996\u5148\u6784\u9020\u5934\u4fe1\u606f":51,"\u9996\u5148\u901a\u8fc7":66,"\u9996\u5148\u9700\u8981\u52a0\u8f7d\u76f8\u5e94\u7684python\u5e93":56,"\u9a71\u52a8":70,"\u9ad8\u4e2d\u6bd5\u4e1a\u751f":90,"\u9ad8\u4eae\u90e8\u5206":61,"\u9ad8\u53ef\u7528":78,"\u9ad8\u5ea6\u652f\u6301\u7075\u6d3b\u548c\u9ad8\u6548\u7684\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e":64,"\u9ad8\u65af\u5206\u5e03":51,"\u9ed1\u8272\u7535\u5f71":90,"\u9ed8\u8ba4":[2,75,94],"\u9ed8\u8ba4256k":28,"\u9ed8\u8ba4\u4e00\u4e2apass\u4fdd\u5b58\u4e00\u6b21\u6a21\u578b":89,"\u9ed8\u8ba4\u4e0d\u663e\u793a":75,"\u9ed8\u8ba4\u4e0d\u8bbe\u7f6e":63,"\u9ed8\u8ba4\u4e3a0":[75,77],"\u9ed8\u8ba4\u4e3a1":[2,77],"\u9ed8\u8ba4\u4e3a100":77,"\u9ed8\u8ba4\u4e3a4096mb":75,"\u9ed8\u8ba4\u4e3a\u4e0d\u4f7f\u7528":91,"\u9ed8\u8ba4\u4e3a\u7b2c\u4e00\u4e2a\u8f93\u5165":63,"\u9ed8\u8ba4\u4e3anull":75,"\u9ed8\u8ba4\u4f1a\u5c06a\u548cb":49,"\u9ed8\u8ba4\u4f7f\u7528concurrentremoteparameterupdat":75,"\u9ed8\u8ba4\u503c":[53,60,77],"\u9ed8\u8ba4\u503c\u4e3a":[58,59],"\u9ed8\u8ba4\u521d\u59cb\u72b6\u4e3a0":63,"\u9ed8\u8ba4\u60c5\u51b5\u4e0b":[51,54,73,93],"\u9ed8\u8ba4\u60c5\u51b5\u4e0b\u4f7f\u7528\u6b64\u7f51\u7edc":93,"\u9ed8\u8ba4\u60c5\u51b5\u4e0b\u6309\u7167float\u7cbe\u5ea6\u8ba1\u7b97":51,"\u9ed8\u8ba4\u6307\u5b9a\u7b2c\u4e00\u4e2a\u8f93\u5165":61,"\u9ed8\u8ba4\u7528\u6765\u5207\u5206\u5355\u8bb0\u548c\u6807\u70b9\u7b26\u53f7":93,"\u9ed8\u8ba4\u7684":79,"\u9ed8\u8ba4\u8bbe\u7f6e\u4e3a\u771f":77,"\u9ed8\u8ba4\u8bbe\u7f6e\u6210\u73af\u5883\u53d8\u91cf":[58,59],"\u9ed8\u8ba4\u8c03\u7528":65,"\u9ed8\u8ba4\u914d\u7f6e\u5982\u4e0b":73,"abstract":43,"adamax\u7b49":89,"amazon\u7535\u5b50\u4ea7\u54c1\u8bc4\u8bba\u6570\u636e":89,"android\u5e73\u53f0\u4e0a\u4f7f\u7528\u7684c":58,"android\u5e73\u53f0\u53ef\u9009\u914d\u7f6e\u53c2\u6570":58,"api\u4e2d\u4f7f\u7528":33,"api\u5bf9\u6bd4\u4ecb\u7ecd":62,"api\u5bfc\u51fa\u7684\u52a8\u6001\u5e93":34,"api\u5bfc\u51fa\u7684\u9759\u6001\u5e93":34,"api\u63a5\u53d7\u7684\u7c7b\u578b\u5168\u662f":34,"api\u63a5\u53e3":[28,78],"api\u63a5\u53e3\u751f\u6210":68,"api\u63a5\u53e3\u7684\u53c2\u6570\u8f6c\u53d1\u7ed9":34,"api\u63a5\u53e3\u7684\u751f\u6210":68,"api\u6587\u6863":58,"api\u65f6":34,"api\u65f6\u6240\u552f\u4e00\u9700\u8981\u5f15\u5165\u7684\u5934\u6587\u4ef6":34,"api\u662f\u591a\u8bed\u8a00api\u7684\u57fa\u7840\u90e8\u5206":34,"api\u66b4\u9732\u7684\u7c7b\u578b":34,"api\u6765\u9884\u6d4b":58,"api\u751f\u6210\u7684\u4e8c\u8fdb\u5236\u6587\u4ef6\u4f1a\u88ab\u5b89\u88c5\u5230":34,"api\u7684\u5934\u6587\u4ef6":[58,59],"api\u7684\u5b9e\u4f8b":34,"api\u7684\u5b9e\u73b0\u7ec6\u8282":34,"api\u7684\u63a5\u53e3":34,"api\u7684\u65f6\u5019\u63a8\u8350paddle\u4e0d\u5d4c\u5165python\u89e3\u91ca\u5668":34,"api\u7684\u7f16\u8bd1\u9009\u9879\u9ed8\u8ba4\u5173\u95ed":34,"api\u76ee\u5f55\u7ed3\u6784\u5982\u4e0a\u56fe\u8868\u6240\u793a":34,"api\u7ea7\u522b":58,"api\u7ea7\u522b\u4e0d\u5c0f\u4e8e21":58,"api\u7ea7\u522b\u4e3a21":58,"api\u83b7\u5f97\u4e86\u795e\u7ecf\u7f51\u7edc\u7684\u53c2\u6570\u5b9e\u4f8b":34,"app\u4e2d":58,"async_sgd\u8fdb\u884c\u8bad\u7ec3\u65f6":51,"atlas\u7684\u8def\u5f84":53,"avx\u7684\u955c\u50cf":54,"awselasticblockstore\u7b49":78,"batch\u4e2d\u5305\u542b":49,"batches\u4e2a\u6279\u6b21\u4fdd\u5b58\u4e00\u6b21\u53c2\u6570":75,"batches\u6b21":75,"bin\u548c\u8bc4\u5206\u6587\u4ef6":91,"blas\u7684\u8def\u5f84":53,"block\u6784\u6210\u4e00\u4e2amodel":20,"book\u4e00\u5b9a\u662f\u60a8\u6700\u597d\u7684\u9009\u62e9":54,"book\u4e2d\u6240\u6709\u7ae0\u8282\u529f\u80fd\u7684\u6b63\u786e\u6027":42,"book\u662f\u4e3a\u7528\u6237\u548c\u5f00\u53d1\u8005\u5236\u4f5c\u7684\u4e00\u4e2a\u4ea4\u4e92\u5f0f\u7684jupyt":54,"book\u7684":56,"book\u7684docker\u955c\u50cf":54,"bool\u578b\u53c2\u6570":2,"boolean":[8,29,32,33],"break":18,"bugfix\u5206\u652f\u4e5f\u662f\u5728\u5f00\u53d1\u8005\u81ea\u5df1\u7684fork\u7248\u672c\u5e93\u7ef4\u62a4":42,"bugfix\u5206\u652f\u9700\u8981\u5206\u522b\u7ed9\u4e3b\u7248\u672c\u5e93\u7684":42,"build\u6307\u4ee4\u6309\u7167dockerfile\u751f\u6210\u955c\u50cf":54,"build\u8fd9\u4e2a\u547d\u4ee4\u7684":54,"byte":[28,51],"c99\u662f\u76ee\u524dc\u6700\u5e7f\u6cdb\u7684\u4f7f\u7528\u6807\u51c6":33,"c\u6709\u6807\u51c6\u7684abi":33,"c\u8bed\u8a00\u662f\u6709\u5bfc\u51fa\u7b26\u53f7\u7684\u6807\u51c6\u7684":33,"caoying\u7684pul":94,"case":[8,22,32,34,38,39,41,72,84],"cc\u4e2d\u7684":69,"cells\u7b49":50,"char":24,"class":[5,6,7,8,9,10,11,12,14,15,17,30,31,33,39,41,43,44,51,67,68,69,93],"cmake\u4e2d\u5c06":72,"cmake\u53c2\u6570\u5982\u4e0b":59,"cmake\u5b98\u65b9\u5bf9android\u5e73\u53f0\u7684\u4ea4\u53c9\u7f16\u8bd1\u63d0\u4f9b\u4e86\u901a\u7528\u7684\u652f\u6301":58,"cmake\u627e\u5230\u7684python\u5e93\u548cpython\u89e3\u91ca\u5668\u7248\u672c\u53ef\u80fd\u6709\u4e0d\u4e00\u81f4\u73b0\u8c61":46,"cmake\u7cfb\u7edf\u5bf9\u4ea4\u53c9\u7f16\u8bd1\u63d0\u4f9b\u4e86\u652f\u6301":[58,59],"cmake\u7f16\u8bd1\u65f6":53,"cmake\u7f16\u8bd1\u7684\u76ee\u6807\u5e73\u53f0":[58,59],"cmake\u914d\u7f6e\u4e2d\u5c06":72,"cmake\u914d\u7f6e\u5b8c\u6210\u540e":[58,59],"com\u4e0b\u8f7d\u7684":54,"conf\u4f5c\u4e3a\u914d\u7f6e":94,"const":[17,22,24,36,43,44,67,68,69],"container\u4e2d":79,"core\u4e2d\u7684\u6a21\u578b\u8fd8\u5728\u4f7f\u7528\u8fd9\u4e2a\u53c2\u6570":34,"core\u4e2d\u8fd9\u4e00\u7c7b\u578b\u63a5\u53e3\u7684\u667a\u80fd\u6307\u9488":34,"core\u662f\u5426\u8fd8\u5728\u4f7f\u7528\u8fd9\u4e2a\u5b9e\u4f8b":34,"core\u6982\u5ff5":34,"cost\u63a5\u6536y_predict\u4e0ey\u4f5c\u4e3a\u8f93\u5165":56,"cost\u8fd8\u5927\u4e8e\u8fd9\u4e2a\u6570":51,"count\u4e2agpu\u4e0a\u4f7f\u7528\u6570\u636e\u5e76\u884c\u6765\u8ba1\u7b97\u67d0\u4e00\u5c42":77,"count\u548cgpu":77,"cuda\u5e93":75,"cudnn\u5e93":[53,75],"dat\u4e2d":91,"data\u5230\u5206\u5e03\u5f0f\u5b58\u50a8\u8865\u5145\u8bad\u7ec3\u6570\u636e":21,"data\u63a5\u53e3\u5206\u914d\u5b9e\u9645\u7684\u5185\u5b58":69,"data\u76ee\u5f55\u4e2d\u5b58\u653e\u5207\u5206\u597d\u7684\u6570\u636e":80,"dataprovider\u5171\u8fd4\u56de\u4e24\u4e2a\u6570\u636e":61,"dataprovider\u5171\u8fd4\u56de\u4e24\u7ec4\u6570\u636e":61,"dataprovider\u662fpaddlepaddle\u8d1f\u8d23\u63d0\u4f9b\u6570\u636e\u7684\u6a21\u5757":1,"dataprovider\u7684\u4ecb\u7ecd":[3,89],"dataprovider\u7f13\u51b2\u6c60\u5185\u5b58":49,"deb\u5305":42,"deb\u5305\u7f16\u8bd1\u95ee\u9898":42,"debian\u955c\u50cf":54,"decay\u5219\u4e3a0":86,"decoder\u5faa\u73af\u5c55\u5f00\u7684\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u4f1a\u5f15\u7528\u5168\u90e8\u7ed3\u679c":63,"decoder\u63a5\u53d7\u4e24\u4e2a\u8f93\u5165":63,"decoder\u6bcf\u6b21\u9884\u6d4b\u4ea7\u751f\u4e0b\u4e00\u4e2a\u6700\u53ef\u80fd\u7684\u8bcd\u8bed":63,"decoer\u67b6\u6784":63,"default":[6,7,8,9,11,12,14,15,17,18,32,35,43,44,45,77,79,80,84,93],"demo\u9884\u6d4b\u8f93\u51fa\u5982\u4e0b":4,"dictionary\u662f\u4ece\u7f51\u7edc\u914d\u7f6e\u4e2d\u4f20\u5165\u7684dict\u5bf9\u8c61":2,"dictionary\u7531\u89e3\u6790\u81ea\u52a8\u751f\u6210":91,"dir\u4e2d\u670916\u4e2a\u5b50\u76ee\u5f55":94,"docker\u5b58\u5728\u95ee\u9898":54,"docker\u5b89\u88c5\u8bf7\u53c2\u8003":70,"docker\u5bb9\u5668\u4e2d":54,"docker\u5bb9\u5668\u4f1a\u8fd0\u884c\u5728\u72ec\u7acb\u7684\u6587\u4ef6\u7cfb\u7edf\u7a7a\u95f4\u4e4b\u4e0a":54,"docker\u7684\u5b98\u7f51":70,"docker\u955c\u50cf\u6765\u670d\u52a1html\u4ee3\u7801":54,"dockerfile\u4f4d\u4e8epaddlepaddl":54,"dockers\u8bbe\u7f6e":54,"double\u7c7b\u578b\u65f6\u4e3a8":51,"dropout\u7684\u6bd4\u4f8b":67,"eigenscalar\u7684\u8f6c\u6362":69,"elec\u6d4b\u8bd5\u96c6":89,"embedding\u6a21\u578b\u9700\u8981\u7a0d\u5fae\u6539\u53d8\u63d0\u4f9b\u6570\u636e\u7684python\u811a\u672c":89,"encode\u6210\u7684\u6700\u540e\u4e00\u4e2a\u5411\u91cf":61,"encoder\u548cdecoder\u53ef\u4ee5\u662f\u80fd\u591f\u5904\u7406\u5e8f\u5217\u7684\u4efb\u610f\u795e\u7ecf\u7f51\u7edc\u5355\u5143":63,"encoder\u8f93\u51fa":63,"entropy\u4f5c\u4e3acost":51,"enum":[22,24,44,45],"evaluator\u7684\u503c\u4f4e\u4e8e0":94,"export":[46,54,86],"f\u4ee3\u8868\u4e00\u4e2a\u6d6e\u70b9\u6570":[2,56],"false\u7684\u60c5\u51b5":2,"fc1\u548cfc2\u5c42\u5728gpu\u4e0a\u8ba1\u7b97":77,"fc3\u5c42\u4f7f\u7528cpu\u8ba1\u7b97":77,"final":[8,9,16,35],"flatten\u65b9\u6cd5\u662f\u628apaddle\u4e2d\u7684\u4e00\u4e2atensor\u8fdb\u884creshape\u64cd\u4f5c":69,"float":[2,6,7,8,10,12,44,52,68,69,72,87],"float\u7b49":77,"from\u65b9\u6cd5\u662f\u628apaddle\u4e2d\u7684\u4e00\u7ef4tensor\u8f6c\u4e3aeigen\u7684\u4e00\u7ef4tensor":69,"from\u662feigentensor\u6a21\u677f\u63d0\u4f9b\u7684\u4e00\u4e2a\u63a5\u53e3":69,"full\u53c2\u6570\u63d0\u4ea4":47,"function":[8,9,10,12,15,17,19,23,24,25,27,31,35,36,37,38,39,40,41,43,44,64,84,93],"function\u4f7f\u7528":50,"gen\u6587\u4ef6\u5939\u4e2d\u7684\u6587\u4ef6\u5217\u8868":94,"generator\u4fbf\u4f1a\u5b58\u4e0b\u5f53\u524d\u7684\u4e0a\u4e0b\u6587":2,"generator\u81f3\u5c11\u9700\u8981\u8c03\u7528\u4e24\u6b21\u624d\u4f1a\u77e5\u9053\u662f\u5426\u505c\u6b62":2,"git\u6d41\u5206\u652f\u6a21\u578b":66,"github\u9996\u9875":66,"golang\u53ef\u4ee5\u4f7f\u7528":33,"golang\u7684":33,"gpu\u4e8c\u8fdb\u5236\u6587\u4ef6":53,"gpu\u5171\u4eabkernel\u5b9e\u73b0\u5728":68,"gpu\u5219\u8fd8\u9700\u8981\u9ad8\u5e76\u884c\u6027":72,"gpu\u53cc\u7f13\u5b58":2,"gpu\u548c\u975eavx\u533a\u5206\u4e86\u5982\u4e0b4\u4e2a\u955c\u50cf":54,"gpu\u5b9e\u73b0\u5171\u4eab\u540c\u4e00\u4e2a":68,"gpu\u5b9e\u73b0\u5728":68,"gpu\u6027\u80fd\u5206\u6790\u4e0e\u8c03\u4f18":71,"gpu\u6267\u884c":69,"gpu\u6838\u5728\u8bad\u7ec3\u914d\u7f6e\u4e2d\u6307\u5b9a":75,"gpu\u7248\u672c\u5e76\u60f3\u4f7f\u7528":92,"gpu\u7684\u4ee3\u7801\u53ef\u4ee5\u590d\u7528":68,"gpu\u7684docker\u955c\u50cf\u7684\u65f6\u5019":46,"gpu\u8ba1\u7b97\u7ed3\u679c":68,"gram\u7ea7\u522b\u7684\u77e5\u8bc6":93,"group\u6559\u7a0b":62,"group\u7684\u5b9e\u73b0\u65b9\u5f0f":50,"gru\u6216lstm":64,"gru\u6a21\u578b":89,"gru\u6a21\u578b\u914d\u7f6e":89,"h\u5e76\u4e0d\u56f0\u96be":33,"hot\u7a20\u5bc6\u5411\u91cf":91,"html\u5373\u53ef\u8bbf\u95ee\u672c\u5730\u6587\u6863":70,"i\u4ee3\u8868\u4e00\u4e2a\u6574\u6570":[2,56],"id\u4e3a0\u7684\u6982\u7387":89,"id\u4e3a1\u7684\u6982\u7387":89,"id\u6307\u5b9a\u4f7f\u7528\u54ea\u4e2agpu\u6838":75,"id\u6307\u5b9a\u7684gpu":77,"id\u65e0\u6548":75,"image\u91cc":79,"images\u6570\u636e\u96c6\u4e0a\u4f20\u5230\u4e91\u7aef\u7684":21,"imdb\u6570\u636e\u96c6\u5305\u542b25":93,"imdb\u6709\u4e24\u4e2a\u6807\u7b7e":93,"imdb\u7684\u6570\u6910\u96c6":93,"import":[2,4,7,8,15,17,18,32,35,40,43,45,52,56,68,85,86,87,93,94],"include\u4e0b\u9700\u8981\u5305\u542bcbla":53,"include\u4e0b\u9700\u8981\u5305\u542bmkl":53,"infer\u63a5\u53e3\u7684\u8fd4\u56de\u503c\u662f\u4e00\u4e2apython":49,"ingress\u9700\u8981\u628apfsclient\u7684\u8eab\u4efd\u4fe1\u606f\u4f20\u7ed9pfsserv":28,"init_hook\u53ef\u4ee5\u4f20\u5165\u4e00\u4e2a\u51fd\u6570":2,"instance\u4e0e\u751f\u6210\u6570\u636e\u96c6\u65f6":21,"instance\u5305\u6db5\u4e24\u4e2a\u503c":21,"instance\u662f\u4e00\u6a21\u4e00\u6837\u7684":21,"int":[2,6,7,8,9,12,17,22,23,24,27,33,34,37,38,44,61,67,69,77,92],"interface\u6587\u4ef6\u7684\u5199\u6cd5\u975e\u5e38":33,"issue\u7f16\u53f7":66,"job\u542f\u52a8\u540e\u4f1a\u521b\u5efa\u8fd9\u4e9bpod\u5e76\u5f00\u59cb\u6267\u884c\u4e00\u4e2a\u7a0b\u5e8f":78,"job\u6216\u8005\u5e94\u7528\u7a0b\u5e8f\u5728\u5bb9\u5668\u4e2d\u8fd0\u884c\u65f6\u751f\u6210\u7684\u6570\u636e\u4f1a\u5728\u5bb9\u5668\u9500\u6bc1\u65f6\u6d88\u5931":78,"job\u662f\u672c\u6b21\u8bad\u7ec3\u5bf9\u5e94\u7684job":80,"job\u7684\u540d\u5b57":80,"kernel\u5b9e\u73b0":68,"kernel\u6ce8\u518ccpu\u5b9e\u73b0\u5728":68,"kernel\u7684\u5b9e\u73b0\u57fa\u4e8eeigen":68,"kubernetes\u4e3a\u8fd9\u6b21\u8bad\u7ec3\u521b\u5efa\u4e863\u4e2apod\u5e76\u4e14\u8c03\u5ea6\u5230\u4e863\u4e2anode\u4e0a\u8fd0\u884c":80,"kubernetes\u5206\u5e03\u5f0f\u8bad\u7ec3":71,"kubernetes\u5355\u673a\u8bad\u7ec3":71,"kubernetes\u53ef\u4ee5\u5728\u7269\u7406\u673a\u6216\u865a\u62df\u673a\u4e0a\u8fd0\u884c":78,"kubernetes\u53ef\u4ee5\u901a\u8fc7yaml\u6587\u4ef6\u6765\u521b\u5efa\u76f8\u5173\u5bf9\u8c61":80,"kubernetes\u5c31\u4f1a\u521b\u5efa3\u4e2apod\u4f5c\u4e3apaddlepaddle\u8282\u70b9\u7136\u540e\u62c9\u53d6\u955c\u50cf":80,"kubernetes\u63d0\u4f9b\u4e86\u591a\u79cd\u96c6\u7fa4\u90e8\u7f72\u7684\u65b9\u6848":78,"kubernetes\u652f\u6301\u591a\u79cdvolum":78,"kubernetes\u6709job\u7c7b\u578b\u7684\u8d44\u6e90\u6765\u652f\u6301":79,"kubernetes\u96c6\u7fa4\u5c31\u662f\u7531node\u8282\u70b9\u4e0emaster\u8282\u70b9\u7ec4\u6210\u7684":78,"label\u662f\u539f\u59cb\u6570\u636e\u4e2d\u5bf9\u4e8e\u6bcf\u4e00\u53e5\u8bdd\u7684\u5206\u7c7b\u6807\u7b7e":61,"labels\u662f\u6bcf\u7ec4\u5185\u6bcf\u4e2a\u53e5\u5b50\u7684\u6807\u7b7e":61,"layer1\u5fc5\u987b\u662f\u4e00\u4e2a0\u5c42\u5e8f\u5217":60,"layer1\u5fc5\u987b\u662f\u4e00\u4e2a\u5355\u5c42\u5e8f\u5217":60,"layer\u4f5c\u4e3a\u4e00\u4e2a\u6574\u4f53\u6765\u5b9e\u73b0":50,"layer\u62ff\u5230\u7684\u7528\u6237\u8f93\u5165":63,"layer\u65f6":50,"layer\u662f\u6211\u4eec\u7684\u79ef\u6728":56,"layer\u7c7b\u53ef\u4ee5\u81ea\u52a8\u8ba1\u7b97\u4e0a\u9762\u7684\u5bfc\u6570":67,"layer\u8ba1\u7b97\u7684\u8f93\u51fa":50,"lib\u4e0b\u9700\u8981\u5305\u542bcblas\u548catlas\u4e24\u4e2a\u5e93":53,"lib\u4e0b\u9700\u8981\u5305\u542bcblas\u5e93":53,"lib\u4e0b\u9700\u8981\u5305\u542bopenblas\u5e93":53,"lib\u76ee\u5f55\u4e0b\u9700\u8981\u5305\u542bmkl_cor":53,"list\u4e2d\u7684\u6bcf\u4e00\u884c\u90fd\u4f20\u9012\u7ed9process\u51fd\u6570":2,"list\u4f5c\u4e3a\u68c0\u67e5\u5217\u8868":42,"list\u5199\u5165\u90a3\u4e2a\u6587\u672c\u6587\u4ef6\u7684\u5730\u5740":2,"list\u548ctest":1,"list\u5982\u4e0b\u6240\u793a":77,"list\u5b58\u653e\u5728\u672c\u5730":1,"list\u6307\u5b9a\u6d4b\u8bd5\u7684\u6a21\u578b\u5217\u8868":77,"long":[8,9,12],"lstm\u67b6\u6784\u7684\u6700\u5927\u4f18\u70b9\u662f\u5b83\u53ef\u4ee5\u5728\u957f\u65f6\u95f4\u95f4\u9694\u5185\u8bb0\u5fc6\u4fe1\u606f":93,"lstm\u6a21\u578b":89,"lstm\u6a21\u578b\u914d\u7f6e":89,"lstm\u7f51\u7edc\u7c7b\u4f3c\u4e8e\u5177\u6709\u9690\u85cf\u5c42\u7684\u6807\u51c6\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":93,"memory\u4e0d\u80fd\u72ec\u7acb\u5b58\u5728":63,"memory\u4e5f\u53ef\u4ee5\u5177\u6709":64,"memory\u4e5f\u53ef\u4ee5\u662f\u5e8f\u5217":64,"memory\u53ea\u80fd\u5728":63,"memory\u53ef\u4ee5\u7f13\u5b58\u4e0a\u4e00\u4e2a\u65f6\u523b\u67d0\u4e00\u4e2a\u795e\u7ecf\u5143\u7684\u8f93\u51fa":61,"memory\u6307\u5411\u4e00\u4e2alay":63,"memory\u662f\u5728\u5355\u6b65\u51fd\u6570\u4e2d\u5faa\u73af\u4f7f\u7528\u7684\u72b6\u6001":64,"memory\u662fpaddlepaddle\u5b9e\u73b0rnn\u65f6\u5019\u4f7f\u7528\u7684\u4e00\u4e2a\u6982\u5ff5":61,"memory\u7684":64,"memory\u7684\u521d\u59cb\u72b6\u6001":63,"memory\u7684\u65f6\u95f4\u5e8f\u5217\u957f\u5ea6\u4e00\u81f4\u7684\u60c5\u51b5":61,"memory\u7684\u66f4\u591a\u8ba8\u8bba\u8bf7\u53c2\u8003\u8bba\u6587":63,"memory\u7684\u8f93\u51fa\u5b9a\u4e49\u5728":64,"memory\u7684i":63,"memory\u9ed8\u8ba4\u521d\u59cb\u5316\u4e3a0":63,"mkl\u7684\u8def\u5f84":53,"mkl_sequential\u548cmkl_intel_lp64\u4e09\u4e2a\u5e93":53,"mnist\u662f\u4e00\u4e2a\u5305\u542b\u670970":2,"mode\u548cattent":94,"mode\u7684python\u51fd\u6570":94,"model\u505a\u5206\u652f\u7ba1\u7406":42,"model\u53ef\u4ee5\u901a\u8fc7":4,"model\u6765\u5b9e\u73b0\u624b\u5199\u8bc6\u522b\u7684\u9884\u6d4b\u4ee3\u7801":4,"movielens\u6570\u636e\u96c6":91,"name\u662f\u4f53\u88c1":91,"name\u662f\u5e74\u9f84":91,"name\u662f\u6027\u522b":91,"name\u662f\u7535\u5f71\u540d":91,"name\u662f\u804c\u4e1a":91,"name\u7ec4\u5408\u53ef\u4ee5\u627e\u5230\u672c\u6b21\u8bad\u7ec3\u9700\u8981\u7684\u6587\u4ef6\u8def\u5f84":80,"ndarray\u7c7b\u578b\u7684\u503c\u548c\u6574\u578b\u7684\u503c":21,"ndk\u4e2d\u5305\u542b\u4e86\u6240\u6709android":58,"new":[8,12,16,17,18,19,22,23,24,25,26,38,39,41,44,66,67,84],"nfs\u7684\u90e8\u7f72\u65b9\u6cd5\u53ef\u4ee5\u53c2\u8003":78,"nmt\u6a21\u578b\u53d7\u5236\u4e8e\u6e90\u8bed\u53e5\u7684\u7f16\u7801":94,"normalization\u5c42":87,"normalization\u5c42\u7684\u53c2\u6570":87,"note\u7684\u4e66\u5199":42,"notebook\u662f\u4e00\u4e2a\u5f00\u6e90\u7684web\u7a0b\u5e8f":54,"null":[8,67,75],"num_gradient_servers\u53c2\u6570":80,"num_samples_processed\u4e3a\u5df2\u8bad\u7ec3\u6837\u672c\u6570":51,"op\u4e0d\u9700\u8981\u5b9a\u4e49opprotomak":68,"op\u5b9a\u4e49":68,"op\u6709\u8ba1\u7b97\u51fd\u6570":68,"op\u6ce8\u518c\u5b9e\u73b0\u5728":68,"op\u8ba1\u7b97\u51fd\u6570\u7684\u57fa\u7c7b":68,"openblas\u7684\u8def\u5f84":53,"opprotomake\u5b9a\u4e49":68,"org\u76ee\u524d\u9075\u5faa":42,"out\u4e0b\u5305\u542b":86,"out\u7684\u6587\u4ef6\u5939":86,"outer_mem\u662f\u4e00\u4e2a\u5b50\u53e5\u7684\u6700\u540e\u4e00\u4e2a\u5411\u91cf":61,"output\u53ef\u4ee5\u662f\u4efb\u610f\u7ef4\u5ea6\u7684tensor":69,"output\u6587\u4ef6\u5939\u5b58\u653e\u8bad\u7ec3\u7ed3\u679c\u4e0e\u65e5\u5fd7":80,"output\u7684\u539f\u6709shape\u4fe1\u606f\u4e0d\u53d8":69,"packages\u91cc\u9762":46,"packages\u91cc\u9762\u7684python\u5305":46,"paddepaddle\u901a\u8fc7\u7f16\u8bd1\u65f6\u6307\u5b9a\u8def\u5f84\u6765\u5b9e\u73b0\u5f15\u7528\u5404\u79cdbla":53,"paddle\u4e00\u4e2a\u52a8\u6001\u5e93\u53ef\u4ee5\u5728\u4efb\u4f55linux\u7cfb\u7edf\u4e0a\u8fd0\u884c":33,"paddle\u4e0b\u8f7d":54,"paddle\u4e2d\u7684\u4e00\u6761pass\u8868\u793a\u8bad\u7ec3\u6570\u636e\u96c6\u4e2d\u6240\u6709\u7684\u6837\u672c\u4e00\u6b21":94,"paddle\u4e2d\u7ecf\u5e38\u4f1a\u5c06\u65f6\u95f4\u5e8f\u5217\u6210\u4e3a":61,"paddle\u4e8c\u8fdb\u5236\u5728\u8fd0\u884c\u65f6\u6355\u83b7\u4e86\u6d6e\u70b9\u6570\u5f02\u5e38":49,"paddle\u5185\u5d4c\u7684python\u89e3\u91ca\u5668\u548c\u5916\u90e8\u4f7f\u7528\u7684python\u5982\u679c\u7248\u672c\u4e0d\u540c":33,"paddle\u5185\u90e8\u7684\u7c7b\u4e3ac":33,"paddle\u662f\u4ece\u5b98\u65b9\u955c\u50cf\u6e90dockerhub":54,"paddle\u66ff\u6362\u6210dock":54,"paddle\u7684\u5404\u7248\u672c\u955c\u50cf\u53ef\u4ee5\u53c2\u8003":79,"paddle\u7684\u591a\u8bed\u8a00\u63a5\u53e3\u5b9e\u73b0\u5305\u62ec\u4e00\u4e0b\u51e0\u4e2a\u65b9\u9762":33,"paddle\u7684\u7c7b\u578b\u5168\u90e8\u9000\u5316\u6210":34,"paddle\u7684\u94fe\u63a5\u65b9\u5f0f\u6bd4\u8f83\u590d\u6742":33,"paddle\u7684c":34,"paddle\u7684dock":79,"paddle\u7684docker\u5f00\u53d1\u955c\u50cf\u5e26\u6709\u4e00\u4e2a\u901a\u8fc7":54,"paddle\u8bad\u7ec3\u4efb\u52a1":21,"paddle\u8def\u5f84\u4e0b":34,"paddle\u955c\u50cf":79,"paddle\u9700\u8981\u4e00\u4e2a\u591a\u8bed\u8a00\u63a5\u53e3":33,"paddle\u9700\u8981\u66b4\u9732\u7684api\u5f88\u591a":34,"paddle\u9759\u6001\u5e93\u94fe\u63a5\u590d\u6742":33,"paddle_\u7c7b\u578b\u540d":34,"paddle_\u7c7b\u578b\u540d_\u51fd\u6570\u540d":34,"paddlepaddle\u4e2d":[60,63],"paddlepaddle\u4e2d\u7684\u4e00\u4e2apass\u610f\u5473\u7740\u5bf9\u6570\u636e\u96c6\u4e2d\u7684\u6240\u6709\u6837\u672c\u8fdb\u884c\u4e00\u6b21\u8bad\u7ec3":93,"paddlepaddle\u4e2d\u7684\u8bb8\u591alayer\u5e76\u4e0d\u5728\u610f\u8f93\u5165\u662f\u5426\u662f\u65f6\u95f4\u5e8f\u5217":61,"paddlepaddle\u4e2d\u8fd8\u5305\u542b":50,"paddlepaddle\u4e2d\u901a\u8fc7reader\u6765\u52a0\u8f7d\u6570\u636e":56,"paddlepaddle\u4e3a\u4ea4\u53c9\u7f16\u8bd1\u63d0\u4f9b\u4e86\u5de5\u5177\u94fe\u914d\u7f6e\u6587\u6863":[58,59],"paddlepaddle\u4e3a\u6df1\u5ea6\u5b66\u4e60\u7814\u7a76\u4eba\u5458\u63d0\u4f9b\u4e86\u4e30\u5bcc\u7684api":56,"paddlepaddle\u4f1a\u5728\u8c03\u7528\u8bfb\u53d6\u6570\u636e\u7684python\u811a\u672c\u4e4b\u524d":89,"paddlepaddle\u4f1a\u81ea\u52a8\u8bbe\u5b9a":50,"paddlepaddle\u4f7f\u7528\u5747\u503c0":51,"paddlepaddle\u4f7f\u7528avx":46,"paddlepaddle\u4f7f\u7528git":42,"paddlepaddle\u4f7f\u7528swig\u5bf9\u5e38\u7528\u7684\u9884\u6d4b\u63a5\u53e3\u8fdb\u884c\u4e86\u5c01\u88c5":4,"paddlepaddle\u4fdd\u5b58\u7684\u6a21\u578b\u53c2\u6570\u6587\u4ef6\u5185\u5bb9\u753116\u5b57\u8282\u5934\u4fe1\u606f\u548c\u7f51\u7edc\u53c2\u6570\u4e24\u90e8\u5206\u7ec4\u6210":51,"paddlepaddle\u4fdd\u5b58\u7684\u6a21\u578b\u53c2\u6570\u6587\u4ef6\u524d16\u5b57\u8282\u4e3a\u5934\u4fe1\u606f":51,"paddlepaddle\u4fdd\u7559\u6dfb\u52a0\u53c2\u6570\u7684\u6743\u529b":2,"paddlepaddle\u5148\u4ece\u4e00\u4e2a\u6587\u4ef6\u5217\u8868\u91cc\u83b7\u5f97\u6570\u636e\u6587\u4ef6\u5730\u5740":52,"paddlepaddle\u53d1\u5e03\u65b0\u7248\u672c\u7684\u65f6\u5019\u90fd\u4f1a\u53d1\u5e03\u5bf9\u5e94\u7248\u672c\u7684\u751f\u4ea7\u955c\u50cf\u4ee5\u53ca\u5f00\u53d1\u955c\u50cf":54,"paddlepaddle\u53ef\u4ee5\u4f7f\u7528cudnn":53,"paddlepaddle\u53ef\u4ee5\u6267\u884c\u7528\u6237\u7684python\u811a\u672c\u7a0b\u5e8f\u6765\u8bfb\u53d6\u5404\u79cd\u683c\u5f0f\u7684\u6570\u636e\u6587\u4ef6":89,"paddlepaddle\u53ef\u4ee5\u6bd4\u8f83\u7b80\u5355\u7684\u5224\u65ad\u54ea\u4e9b\u8f93\u51fa\u662f\u5e94\u8be5\u8de8\u8d8a\u65f6\u95f4\u6b65\u7684":61,"paddlepaddle\u53ef\u4ee5\u901a\u8fc7\u8be5\u673a\u5236\u5224\u65ad\u662f\u5426\u5df2\u7ecf\u6536\u96c6\u9f50\u6240\u6709\u7684\u68af\u5ea6":67,"paddlepaddle\u5728\u5b9e\u73b0rnn\u7684\u65f6\u5019":61,"paddlepaddle\u5728\u6fc0\u6d3b\u51fd\u6570\u91cc\u5b9e\u73b0dropout":50,"paddlepaddle\u5b58\u7684\u662f\u6709\u503c\u4f4d\u7f6e\u7684\u7d22\u5f15":[2,56],"paddlepaddle\u5b9a\u4e49\u7684\u53c2\u6570":2,"paddlepaddle\u5c06\u4ee5\u8bbe\u7f6e\u53c2\u6570\u7684\u65b9\u5f0f\u6765\u8bbe\u7f6e":89,"paddlepaddle\u5c06\u5728\u89c2\u6d4b\u6570\u636e\u96c6\u4e0a\u8fed\u4ee3\u8bad\u7ec330\u8f6e":52,"paddlepaddle\u5c06\u6839\u636e":58,"paddlepaddle\u5c06\u6bcf\u4e2a\u6a21\u578b\u53c2\u6570\u4f5c\u4e3a\u4e00\u4e2anumpy\u6570\u7ec4\u5355\u72ec\u5b58\u4e3a\u4e00\u4e2a\u6587\u4ef6":52,"paddlepaddle\u5c06\u81ea\u52a8\u4e0b\u8f7d\u548c\u7f16\u8bd1\u6240\u6709\u7b2c\u4e09\u65b9\u4f9d\u8d56\u5e93":[58,59],"paddlepaddle\u5c06train":2,"paddlepaddle\u5f00\u53d1\u8fc7\u7a0b\u4f7f\u7528":42,"paddlepaddle\u5f00\u53d1\u955c\u50cf\u5c31\u88ab\u6784\u5efa\u5b8c\u6bd5\u4e86":54,"paddlepaddle\u63d0\u4f9b\u4e13\u7528\u7684":21,"paddlepaddle\u63d0\u4f9b\u4e86\u5f88\u591a\u4f18\u79c0\u7684\u5b66\u4e60\u7b97\u6cd5":52,"paddlepaddle\u63d0\u4f9bdocker\u955c\u50cf\u6765\u90e8\u7f72\u73af\u5883":55,"paddlepaddle\u652f\u6301\u4e0d\u540c\u7c7b\u578b\u7684\u8f93\u5165\u6570\u636e":56,"paddlepaddle\u652f\u6301\u4ee5\u4e0b\u4efb\u610f\u4e00\u79cdblas\u5e93":53,"paddlepaddle\u652f\u6301\u5927\u91cf\u7684\u8ba1\u7b97\u5355\u5143\u548c\u4efb\u610f\u6df1\u5ea6\u7684\u7f51\u7edc\u8fde\u63a5":52,"paddlepaddle\u652f\u6301\u975e\u5e38\u591a\u7684\u4f18\u5316\u7b97\u6cd5":49,"paddlepaddle\u652f\u6301sparse\u7684\u8bad\u7ec3":49,"paddlepaddle\u6587\u6863\u4e3b\u9898\u5728":70,"paddlepaddle\u6587\u6863\u4f7f\u7528":70,"paddlepaddle\u662f\u6e90\u4e8e\u767e\u5ea6\u7684\u4e00\u4e2a\u6df1\u5ea6\u5b66\u4e60\u5e73\u53f0":[52,56],"paddlepaddle\u6bcf\u6b21\u53d1\u5e03\u65b0\u7248\u672c\u90fd\u4f1a\u53d1\u5e03\u5bf9\u5e94\u7684\u5f00\u53d1\u955c\u50cf\u4f9b\u5f00\u53d1\u8005\u76f4\u63a5\u4f7f\u7528":54,"paddlepaddle\u6bcf\u6b21\u53d1\u65b0\u7684\u7248\u672c":42,"paddlepaddle\u6bcf\u6b21\u53d1\u7248\u672c\u9996\u5148\u8981\u4fdd\u8bc1paddlepaddl":42,"paddlepaddle\u7684\u4e3b\u7248\u672c\u5e93\u9075\u5faa":42,"paddlepaddle\u7684\u5185\u5b58\u5360\u7528\u4e3b\u8981\u5206\u4e3a\u5982\u4e0b\u51e0\u4e2a\u65b9\u9762":49,"paddlepaddle\u7684\u53c2\u6570\u4f7f\u7528\u540d\u5b57":51,"paddlepaddle\u7684\u5df2\u7ecf\u5b89\u88c5\u5b8c\u6210":58,"paddlepaddle\u7684\u6240\u6709layer\u90fd\u6709\u552f\u4e00\u7684nam":50,"paddlepaddle\u7684\u6570\u636e\u5305\u62ec\u56db\u79cd\u4e3b\u8981\u7c7b\u578b":2,"paddlepaddle\u7684\u6587\u6863\u5305\u62ec\u82f1\u6587\u6587\u6863":70,"paddlepaddle\u7684\u6587\u6863\u6784\u5efa\u6709\u4e24\u79cd\u65b9\u5f0f":70,"paddlepaddle\u7684\u6e90\u7801":66,"paddlepaddle\u7684\u7f16\u8bd1\u9009\u9879":55,"paddlepaddle\u7684bas":67,"paddlepaddle\u7684cmake\u7cfb\u7edf\u5c06\u6839\u636e\u8be5\u503c\u81ea\u52a8\u63a8\u5bfc\u548c\u8bbe\u7f6e\u9700\u8981\u4f7f\u7528\u7684\u4ea4\u53c9\u7f16\u8bd1\u5668":58,"paddlepaddle\u7684cmake\u7cfb\u7edf\u5c06\u6839\u636e\u8be5\u503c\u81ea\u52a8\u8bbe\u7f6e\u9700\u8981\u4f7f\u7528\u7684\u4ea4\u53c9\u7f16\u8bd1\u5668":59,"paddlepaddle\u7684cmake\u7cfb\u7edf\u624d\u8ba4\u4e3a\u5728\u662f\u5728\u4ea4\u53c9\u7f16\u8bd1raspberri":59,"paddlepaddle\u7684cmake\u7cfb\u7edf\u624d\u8ba4\u4e3a\u662f\u5728\u4ea4\u53c9\u7f16\u8bd1android\u7cfb\u7edf\u7684\u7248\u672c":58,"paddlepaddle\u7684docker\u5bb9\u5668\u4f7f\u7528\u65b9\u5f0f":55,"paddlepaddle\u7684docker\u955c\u50cf\u5c31\u5305\u542b\u4e86paddlepaddle\u7684python\u5e93\u4ee5\u53ca\u5176\u4f9d\u8d56\u7684\u591a\u4e2apython\u5e93":54,"paddlepaddle\u76ee\u524d\u53ea\u652f\u6301\u5728\u6bcf\u4e2a\u65f6\u95f4\u6b65\u4e2d":61,"paddlepaddle\u76ee\u524d\u552f\u4e00\u5b98\u65b9\u652f\u6301\u7684\u8fd0\u884c\u7684\u65b9\u5f0f\u662fdocker\u5bb9\u5668":54,"paddlepaddle\u76ee\u524d\u63d0\u4f9b\u4e24\u79cd\u53c2\u6570\u521d\u59cb\u5316\u7684\u65b9\u5f0f":51,"paddlepaddle\u76ee\u524d\u652f\u63018\u79cdlearning_rate_schedul":51,"paddlepaddle\u82e5\u68c0\u6d4b\u5230\u7528\u6237\u4f7f\u7528\u7684cmake\u7248\u672c\u4e0d\u4f4e\u4e8e3":58,"paddlepaddle\u8981\u6c42\u4f7f\u7528\u7684\u7f16\u8bd1\u5de5\u5177\u94fe\u6240\u652f\u6301\u7684andoid":58,"paddlepaddle\u8c03\u7528process\u51fd\u6570\u6765\u8bfb\u53d6\u6570\u636e":89,"paddlepaddle\u8d1f\u8d23\u5b8c\u6210\u4fe1\u606f\u548c\u68af\u5ea6\u5728\u65f6\u95f4\u5e8f\u5217\u4e0a\u7684\u4f20\u64ad":63,"paddlepaddle\u8d1f\u8d23\u5b8c\u6210\u4fe1\u606f\u548c\u8bef\u5dee\u5728\u65f6\u95f4\u5e8f\u5217\u4e0a\u7684\u4f20\u64ad":63,"paddlepaddle\u90fd\u53ef\u4ee5\u81ea\u52a8\u5224\u65ad\u786c\u4ef6\u662f\u5426\u652f\u6301avx":54,"paddlepaddle\u955c\u50cf\u9700\u8981\u63d0\u4f9b":80,"paddlepaddle\u9700\u8981\u7528\u6237\u5728\u7f51\u7edc\u914d\u7f6e":1,"paddlepaddle\u9700\u8981\u7684\u6240\u6709\u7f16\u8bd1\u5de5\u5177":54,"pass\u4e2a\u6a21\u578b\u5230\u7b2c":75,"pass\u5230":94,"pass\u5c06\u4e0d\u8d77\u4f5c\u7528":75,"pass\u8f6e\u5f00\u59cb\u8bad\u7ec3":75,"pass\u8f6e\u7684\u6a21\u578b\u7528\u4e8e\u6d4b\u8bd5":75,"passes\u8f6e":75,"patch\u53f7":42,"patch\u53f7\u52a0\u4e00":42,"path\u6307\u5b9a\u6d4b\u8bd5\u7684\u6a21\u578b":77,"period\u4e2a\u6279\u6b21\u5bf9\u6240\u6709\u6d4b\u8bd5\u6570\u636e\u8fdb\u884c\u6d4b\u8bd5":75,"period\u4e2a\u6279\u6b21\u6253\u5370\u65e5\u5fd7\u8fdb\u5ea6":75,"period\u4e2a\u6279\u6b21\u8f93\u51fa\u53c2\u6570\u7edf\u8ba1":75,"period\u4e2a\u6279\u6b21\u8f93\u51fa\u7b26\u53f7":75,"period\u4e2abatch\u5904\u7406\u7684\u5f53\u524d\u635f\u5931":93,"period\u4e2abatch\u7684\u5206\u7c7b\u9519\u8bef":93,"period\u6574\u9664":75,"period\u8f6e\u4fdd\u5b58\u8bad\u7ec3\u53c2\u6570":75,"pfsclient\u9700\u8981\u548cingress\u4e4b\u95f4\u505a\u53cc\u5411\u9a8c\u8bc1":28,"pfsclient\u9700\u8981\u5728\u4f20\u8f93\u5b8c\u6bd5\u6700\u540e\u4e00\u4e2achunk\u7684\u65f6\u5019\u68c0\u67e5destination\u6587\u4ef6\u7684md5\u503c\u662f\u5426\u548csource\u6587\u4ef6\u4e00\u81f4":28,"pfsserver\u63d0\u4f9brest":28,"pi\u5e73\u53f0\u4e0a\u9002\u7528\u7684paddlepaddle\u5e93":59,"pi\u5e73\u53f0\u4e0a\u9002\u7528\u7684paddlepaddle\u7684\u65b9\u6cd5\u548c\u6b65\u9aa4":59,"pi\u5e73\u53f0\u4f7f\u7528\u7684c":59,"pi\u5e73\u53f0\u53ef\u9009\u914d\u7f6e\u53c2\u6570":59,"pi\u7248\u672c\u7684\u5e93":59,"pi\u7248\u672cpaddlepaddle\u5e93\u65f6":59,"pi\u7cfb\u7edf":59,"pi\u7cfb\u7edf\u4e0a":59,"pi\u7cfb\u7edf\u7684\u7248\u672c":59,"pod\u4e2d\u7684\u5bb9\u5668\u5171\u4eabnet":78,"pod\u662fkubernetes\u7684\u6700\u5c0f\u8c03\u5ea6\u5355\u5143":78,"pooling\u5bf9\u7279\u5f81\u56fe\u4e0b\u91c7\u6837":86,"process\u51fd\u6570\u4f1a\u7528yield\u8bed\u53e5\u8f93\u51fa\u8fd9\u6761\u6570\u636e":89,"prod\u7684\u751f\u4ea7\u955c\u50cf":54,"public":[12,17,30,43,44,67,68,69,79,93],"py\u4e2d":42,"py_paddle\u91cc\u9762\u63d0\u4f9b\u4e86\u4e00\u4e2a\u5de5\u5177\u7c7b":4,"pydataprovider2\u4f1a\u5c3d\u53ef\u80fd\u591a\u7684\u4f7f\u7528\u5185\u5b58":2,"pydataprovider2\u63d0\u4f9b\u4e86\u4e24\u79cd\u7b80\u5355\u7684cache\u7b56\u7565":2,"pydataprovider2\u662fpaddlepaddle\u4f7f\u7528python\u63d0\u4f9b\u6570\u636e\u7684\u63a8\u8350\u63a5\u53e3":2,"pydataprovider2\u7684\u4f7f\u7528":[1,3,49,89,91],"pydataprovider\u4f7f\u7528\u7684\u662f\u5f02\u6b65\u52a0\u8f7d":49,"pypi\u4e0a\u7684package\u540d\u79f0\u4e3apaddlepaddle\u548cpaddlepaddl":42,"python\u4ee3\u7801\u5c06\u968f\u673a\u4ea7\u751f2000\u4e2a\u89c2\u6d4b\u70b9":52,"python\u53ef\u4ee5\u89e3\u9664\u6389\u5185\u90e8\u53d8\u91cf\u7684\u5f15\u7528":2,"python\u5b89\u88c5\u5305\u652f\u6301linux":46,"python\u5c01\u88c5\u7684\u5b9e\u73b0\u4f7f\u5f97\u6211\u4eec\u53ef\u4ee5\u5728\u914d\u7f6e\u6587\u4ef6\u4e2d\u4f7f\u7528\u65b0\u5b9e\u73b0\u7684\u7f51\u7edc\u5c42":67,"query\u6539\u5199":94,"rate\u4e3a0":94,"rate\u4e3a5":94,"rate\u88ab\u8bbe\u7f6e\u4e3a0":86,"reader\u7684\u4f7f\u7528\u65b9\u5f0f\u90fd\u662f\u4e00\u81f4\u7684":21,"reader\u8f93\u51fa\u7684data":21,"recommendation\u6587\u4ef6\u5939\u5185\u5b58\u653e\u8bad\u7ec3\u6587\u4ef6":80,"release\u9875\u9762":42,"repo\u7684\u6839\u76ee\u5f55":54,"request\u524d":66,"request\u7684":66,"request\u88ab\u5408\u5e76\u540e":66,"research\u5b9e\u9a8c\u5ba4\u641c\u96c6\u6574\u7406":90,"resnet\u6a21\u578b":88,"return":[2,6,8,9,11,12,14,15,16,17,21,22,24,27,30,32,35,39,43,44,45,51,52,56,61,64,67,69,80,87],"rnn\u5373\u65f6\u95f4\u9012\u5f52\u795e\u7ecf\u7f51\u7edc":61,"rnn\u5bf9\u4e8e\u6bcf\u4e00\u4e2a\u65f6\u95f4\u6b65\u901a\u8fc7\u4e86\u4e00\u4e2alstm\u7f51\u7edc":61,"rnn\u603b\u662f\u5f15\u7528\u4e0a\u4e00\u65f6\u523b\u9884\u6d4b\u51fa\u7684\u8bcd\u7684\u8bcd\u5411\u91cf":63,"rnn\u6a21\u578b":89,"rnn\u76f8\u5173\u6a21\u578b":71,"rnn\u914d\u7f6e":62,"s3\u4e4b\u7c7b\u7684\u5206\u5e03\u5f0f\u5b58\u50a8\u4e4b\u4e0a":21,"search\u7684\u65b9\u6cd5":75,"sentences\u662f\u53cc\u5c42\u65f6\u95f4\u5e8f\u5217\u7684\u6570\u636e":61,"seq\u53c2\u6570\u5fc5\u987b\u4e3afals":63,"server\u4e2a\u6279\u6b21\u6253\u5370\u65e5\u5fd7\u8fdb\u5ea6":75,"server\u4e4b\u4e0a":20,"server\u4e4b\u95f4\u7684\u7f51\u7edc\u5e26\u5bbd":20,"server\u4f1a\u6682\u505c\u53c2\u6570\u66f4\u65b0\u5e76\u7b49\u5f85":20,"server\u4f1a\u83b7\u53d6parameters\u5185\u5b58\u7684":20,"server\u5185\u5b58\u4e2d\u7684\u6a21\u578b\u6570\u636e\u7684\u5b8c\u6574\u955c\u50cf":20,"server\u540c\u6b65\u7684\u4fdd\u5b58\u4e00\u4e2a\u7279\u5b9a\u65f6\u95f4\u70b9\u7684\u5168\u5c40\u68c0\u67e5\u70b9":20,"server\u5728\u96c6\u7fa4\u4e2d\u542f\u52a8\u540e":20,"server\u6545\u969c\u540e\u88abkubernetes\u91cd\u65b0\u542f\u52a8":20,"server\u6b64\u65f6\u8fd8\u9700\u8981\u901a\u8fc7\u7f51\u7edc\u8bbf\u95ee\u5206\u5e03\u5f0f\u5b58\u50a8\u4ee5\u4fdd\u5b58\u5feb\u7167":20,"server\u751f\u6210\u4e00\u4e2auuid":20,"server\u7684\u5355\u70b9\u6216\u591a\u70b9\u540c\u65f6\u6545\u969c":20,"server\u7684\u6570\u636e\u5feb\u7167":20,"server\u7684\u68c0\u67e5\u70b9\u5404\u81ea\u72ec\u7acb\u4fdd\u5b58":20,"server\u7b2c\u4e00\u6b21\u542f\u52a8\u6216\u4efb\u610f\u65f6\u95f4paramet":20,"sh\u6765\u8bad\u7ec3\u6a21\u578b":86,"sh\u8c03\u7528\u4e86":87,"short":[8,9,43],"simd\u6307\u4ee4\u63d0\u9ad8cpu\u6267\u884c\u6548\u7387":46,"size\u4e3a1":94,"size\u4e3a50":94,"size\u4e3a512":75,"size\u53ef\u80fd\u4f1a\u5bf9\u8bad\u7ec3\u7ed3\u679c\u4ea7\u751f\u5f71\u54cd":49,"size\u5927\u5c0f\u4e3a128":93,"size\u662f3":94,"size\u672c\u8eab\u662f\u795e\u7ecf\u7f51\u7edc\u7684\u8d85\u53c2\u6570":49,"size\u7684\u503c":2,"softmax\u5c42":85,"softmax\u6fc0\u6d3b\u7684\u8f93\u51fa\u7684\u548c\u603b\u662f1":67,"sparse\u8bad\u7ec3\u9700\u8981\u8bad\u7ec3\u7279\u5f81\u662f":49,"srl\u4f5c\u4e3a\u5f88\u591a\u81ea\u7136\u8bed\u8a00\u5904\u7406\u4efb\u52a1\u4e2d\u7684\u4e2d\u95f4\u6b65\u9aa4\u662f\u5f88\u6709\u7528\u7684":92,"ssh\u65b9\u5f0f\u7684\u4e00\u4e2a\u4f18\u70b9\u662f\u6211\u4eec\u53ef\u4ee5\u4ece\u591a\u4e2a\u7ec8\u7aef\u8fdb\u5165\u5bb9\u5668":54,"ssh\u8fdb\u5165\u5bb9\u5668":54,"static":[14,24,34,43,44,84],"step\u51fd\u6570\u4e2d\u7684memori":63,"step\u51fd\u6570\u5185\u90e8\u53ef\u4ee5\u81ea\u7531\u7ec4\u5408paddlepaddle\u652f\u6301\u7684\u5404\u79cdlay":63,"subseq\u7684\u6bcf\u4e2a\u5143\u7d20\u662f\u4e00\u4e2a0\u5c42\u5e8f\u5217":60,"super":67,"swig\u652f\u6301\u7684\u8bed\u8a00\u6216\u8005\u89e3\u91ca\u5668\u6709\u5c40\u9650":33,"swig\u66b4\u9732\u7684\u63a5\u53e3\u4fdd\u7559\u4e86c":33,"swig\u751f\u6210\u7684\u4ee3\u7801\u4e0d\u80fd\u4fdd\u8bc1\u591a\u8bed\u8a00\u4ee3\u7801\u98ce\u683c\u7684\u4e00\u81f4\u6027":33,"swig\u76f4\u63a5\u8bfb\u53d6c":33,"swig\u9700\u8981\u5199\u4e00\u4e2ainterface\u6587\u4ef6":33,"swig_paddle\u4e2d\u7684\u9884\u6d4b\u63a5\u53e3\u7684\u53c2\u6570\u662f\u81ea\u5b9a\u4e49\u7684c":4,"switch":[17,34],"t\u6307\u5b9a\u4e86\u751f\u6210\u7684\u955c\u50cf\u7684\u540d\u5b57":54,"tag\u4e3a":42,"tensor\u5230\u5bf9eigentensor\u7684\u8f6c\u6362":69,"tensor\u5230eigentensor":69,"tensor\u5b9a\u4e49\u5728framework\u76ee\u5f55\u4e0b":69,"tensor\u662f\u4e00\u4e2a\u6b63\u5728\u5f00\u53d1\u4e2d\u7684\u6a21\u5757":69,"tensor\u6a21\u5757\u5bf9el":69,"tensor\u6a21\u5757\u6765\u5b9e\u73b0":68,"tensor\u6a21\u5757\u7684\u6587\u6863\u8f83\u5c11":69,"tensor\u6a21\u5757\u7684\u8be6\u7ec6\u4ecb\u7ecd\u8bf7\u53c2\u8003":69,"test\u548cgen\u8fd9\u4e09\u4e2a\u6587\u4ef6\u5939\u662f\u56fa\u5b9a\u7684":94,"tests\u7684paddlepaddl":66,"tflops\u4e86":72,"toolchain\u76ee\u5f55\u751f\u6210\u4e00\u5957\u7f16\u8bd1\u5de5\u5177\u94fe":58,"trainer\u63a5\u6536\u4e09\u4e2a\u53c2\u6570":56,"true":[6,7,8,9,11,12,14,15,17,22,32,38,44,49,51,61,64,67,77,80,87,91,92,93,94],"true\u8868\u793a\u53cd\u5411\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":64,"try":[10,18,19,22,23,24,38,43,46],"type\u5b57\u6bb5\u5747\u4e0d\u5c3d\u76f8\u540c":34,"type\u662fon":91,"ubuntu\u5b89\u88c5\u5305\u7684\u529f\u80fd\u6b63\u786e\u6027":42,"unit\u5728\u4e00\u4e2a\u65f6\u95f4\u6b65\u5185\u8ba1\u7b97\u5f97\u5230\u7684\u4e2d\u95f4\u503c":50,"unsupported\u6a21\u5757":68,"update\u53c2\u6570\u65f6\u624d\u6709\u6548":75,"utf8\u7f16\u7801":85,"uts\u7b49linux":78,"v1\u7248\u672c":46,"v2\u4e4b\u540e\u7684\u4efb\u4f55\u4e00\u4e2a\u7248\u672c\u6765\u7f16\u8bd1\u8fd0\u884c":53,"var":[17,30,31,32,37,41,45],"vector\u662frank\u4e3a1\u7684tensor":69,"vocab\u4e2d\u6bcf\u4e2a\u5207\u5206\u5355\u8bcd\u7684\u9884\u671f\u8bc4\u7ea7":93,"vocab\u505a\u4e3a\u5b57\u5178":93,"void":[17,22,24,30,33,34,37,44,67,68,69],"volume\u6302\u8f7d\u5230\u5bb9\u5668\u4e2d":78,"w0\u548c":87,"wbias\u662f\u9700\u8981\u5b66\u4e60\u7684\u53c2\u6570":87,"wheel\u5305":42,"while":[6,12,17,26,36,38,43,80,94],"wise\u8ba1\u7b97\u63d0\u4f9b\u4e86\u5f3a\u5927\u7684\u652f\u6301":69,"wmt14\u6570\u636e\u7684\u63d0\u4f9b\u6587\u4ef6\u5728":64,"words\u5373\u4e3a\u8fd9\u4e2a\u6570\u636e\u4e2d\u7684\u5355\u5c42\u65f6\u95f4\u5e8f\u5217":61,"words\u662f\u539f\u59cb\u6570\u636e\u4e2d\u7684\u6bcf\u4e00\u53e5\u8bdd":61,"x86_64\u548cmaco":46,"x\u4e0ey\u4e3a\u4e4b\u524d\u63cf\u8ff0\u7684\u8f93\u5165\u5c42":56,"x\u548cwindow":[54,58],"x\u548cwindows\u4e0a\u7684\u786c\u4ef6\u8d44\u6e90":54,"y\u8868\u793a\u8f93\u5165\u6570\u636e\u662f\u4e00\u4e2a\u7ef4\u5ea6\u4e3a1\u7684\u7a20\u5bc6\u5411\u91cf":56,"yaml\u6587\u4ef6\u4e2d\u5404\u4e2a\u5b57\u6bb5\u7684\u5177\u4f53\u542b\u4e49":80,"yaml\u6587\u4ef6\u63cf\u8ff0\u4e86\u8fd9\u6b21\u8bad\u7ec3\u4f7f\u7528\u7684docker\u955c\u50cf":80,"zero\u4e09\u79cd\u64cd\u4f5c":75,AGE:79,AWS:[21,78,81,82],And:[7,8,10,12,14,16,22,26,27,38,39,43],But:[8,9,12,16,30,43,46,84],EOS:8,For:[4,7,8,10,12,14,15,17,23,24,25,27,30,31,35,36,38,39,44,72,84],IDE:65,IDs:[12,26],IRs:39,Its:32,K8s:84,NFS:78,NMS:8,Not:[15,19,41,84],OPs:[39,40],One:[7,9,14,16,26,43],Ops:[41,44],PFS:28,QoS:79,TLS:[15,28],That:[8,12,38],The:[2,5,6,7,8,9,10,12,14,15,16,18,19,23,25,26,27,29,34,35,37,38,39,40,43,44,45,67,68,69,80,89,92,94],Their:[8,19],Then:[8,30,45],There:[7,8,12,14,15,17,18,19,24,26,27,29,35,36,43,45],These:[7,17,31,37,44],Use:[7,12,15,29,38,41],Used:[9,41],Using:[19,41,43,93],Will:[12,14],With:[8,9,37],YES:27,Yes:54,___embedding_0__:80,___embedding_1__:80,__doc__:44,__file__:27,__gradient_machines__:14,__impl__:44,__init__:[45,67],__main__:[4,87],__meta__:91,__metaclass__:68,__name__:[4,87],__param_conf__:14,__rnn_step__:64,__square_error_cost_0__:80,__tmp_params__:14,_binari:18,_block:45,_librari:18,_link:9,_name:45,_op:68,_proj:8,_recurrent_group:64,_res2_1_branch1_bn:87,_source_language_embed:[64,85],_target_language_embed:[64,85],_test:18,_var:45,_var_desc:45,a_op:68,aaaaa:21,abc:8,abl:[8,15,39,41,45,84],about:[8,9,17,18,27,29,35,39,43,44,92,94],abov:[2,7,8,15,17,18,19,23,30,31,35,37,39,44,72,84],abs:[8,9,16],abs_numerical_grad:16,acceler:20,accept:[6,8,12,15,38,41,92],access:[8,9,15,18,23,26,27,37,39],accord:[7,8,16,24,31,39,40,41,45],accordingli:[7,8],account:[41,84],accrodingli:22,accumul:[19,24],accur:[16,26],accuraci:7,acl:93,aclimdb:93,across:[8,12,35],act:[8,9,17,39,49,52,56,61,64],act_output:44,act_typ:89,activ:[9,13,18,39,44,49,56,64,89],activi:9,actual:[22,44],adadelta:[49,89],adagrad:89,adam:[15,24,51,89,93],adamoptim:[85,89,93,94],adapt:[7,10,14],add:[8,9,12,14,16,17,18,22,26,30,32,39,40,41,43,45,46,66,69],add_depend:18,add_execut:18,add_input:[35,67],add_memori:[17,35],add_output:35,add_test:[18,67],add_to:[8,50],add_two:[17,35],add_unittest_without_exec:67,add_var:45,addattr:[44,68],addbia:67,addcom:[44,68],added:[7,14,17,40,66],addinitialoper:45,addinput:[44,68],addit:[8,9],addop:30,addoutput:68,addr:19,address:[19,24,39,84],addrow:67,addtolay:8,addtyp:44,admin:84,administr:[26,84],advanc:16,advantag:16,adversari:38,affect:[8,17],afford:23,afi:2,aforement:18,after:[8,9,12,17,18,23,24,26,29,36,39,40,66],again:[15,19],age:[12,80,91],agg_level:[8,60,61],aggregatelevel:[60,61],ago:18,aircraft:94,airplan:86,aistat:8,alex:[8,93],alexnet_pass1:77,alexnet_pass2:77,algo_hrnn_demo:61,algorithm:[8,10,23,85,93,94],align:[8,9,12,94],all:[2,6,7,8,10,14,15,17,18,19,22,24,26,27,29,31,34,35,37,39,41,44,49,63,80,84,92,93],alloc:[6,24,27,69],allow:[15,24,39],allow_only_one_model_on_one_gpu:[74,75,77],alpha:18,alreadi:[18,19,39,40,43,46],also:[8,9,12,15,17,18,22,25,30,31,36,38,43,44,72,84,89],altogeth:84,alwai:[8,9,14,18,38,45,80],amazon:79,ambigu:38,amd64:78,amend:66,analysi:[92,93],ancestor:37,android:58,android_abi:58,android_arm_neon:58,android_native_api_level:58,android_standalone_toolchain:58,androideabi:58,ani:[8,9,12,15,18,19,24,26,27,38,39,40],annot:92,annual:92,anoth:[8,14,15,17,27,43,44],anroid_arm_mod:58,anyth:[12,38,92],api:[12,14,15,18,24,25,27,28,30,35,36,41,42,45,72,80,83,84,93],api_shar:18,api_test:18,api_trainer_config:91,api_trainer_config_helpers_data_sourc:91,api_trainer_config_helpers_lay:[64,89],api_trainer_config_helpers_layers_context_project:91,api_trainer_config_helpers_layers_cos_sim:91,api_trainer_config_helpers_layers_data_lay:91,api_trainer_config_helpers_layers_embedding_lay:91,api_trainer_config_helpers_layers_fc_lay:91,api_trainer_config_helpers_layers_pooling_lay:91,apiserv:78,apivers:[78,79,80],apo:94,appear:31,append:[2,14,38,61,64,80],append_gradient_machin:14,appleyard:72,appli:[8,31,43],applic:[31,41,72,79,84],applicaton:41,applyl1:22,approach:[8,36,39,40,84,89],apt:86,arbitrari:[8,39],arch:58,architectur:94,archiv:[12,33,34],arg:[2,7,9,44,51,52,68,80,86,87,89,92,93],arg_nam:8,argpars:80,args_ext:80,argument:[2,8,12,17,22,23,36,80,92],argumentpars:80,argv:87,arm64:58,arm:[58,59],arm_soft_fp_abi:58,armeabi:58,around:[2,8,26,84],arrai:[4,6,8,12,14,24,31,38,41,49,51,56,68,87],articl:[31,66],arxiv:[8,9,93],as_row_vector:8,ask:[19,26],assert:4,assign:[7,8,23,37,84],associ:[36,92],assum:[7,8,17,39],astyp:[38,68],asyc:19,async:[19,74],async_count:[74,75],async_lagged_grad_discard_ratio:75,async_lagged_ratio_default:[74,75],async_lagged_ratio_min:[74,75],asynchron:19,atla:53,atlas_root:53,attenion:9,attent:[8,9,94],attr1:8,attr2:8,attr:[6,8,9,17,37,44,49,50,51,64,68],attr_map:44,attrdesc:37,attribut:[8,9,13,17,37,40,41,43,44],attributemap:68,attrproto:44,attrtyp:[37,44,68],attrvalu:44,auc:74,author:[28,78],authorized_kei:73,auto:[17,22,33,41,43,67,68,69,72],automat:[8,15,24,30,39,40,41,44,94],automaticli:8,automobil:86,avail:[19,24,40,84],averag:[7,8,11,23,92],average_test_period:[74,75,92],averagepool:8,avg:[60,72,89],avgcost:[89,91,93,94],avgpool:[8,89],avoid:[16,17,19,36,72],avx:54,await:79,awar:[15,35],aws:28,axi:49,aync:40,azur:78,b2t:85,b363:79,b8561f5c79193550d64fa47418a9e67ebdd71546186e840f88de5026b8097465:79,back:[14,19,39],background:[7,8],background_id:[7,8],backpropag:16,backward:[5,8,9,16,17,22,24,36,67,68],backward_first:64,backward_op:16,backwardactiv:67,baidu:79,bake:39,balanc:40,balasubramanyan:93,bank:92,bare:[78,79,84],barrierstatset:72,base:[7,8,11,12,15,23,37,39,65],baseactiv:9,baseev:14,basematrix:67,basenam:7,basepoolingtyp:[8,9],basestr:[6,7,8,9,11,14],bash:[54,65,66,79,80],basic:[8,14,37,41],batch:[8,9,10,12,14,15,17,19,21,22,35,52,56,73,79,80,86,87,89,91,93,94],batch_0:87,batch_id:[14,56],batch_norm_lay:9,batch_norm_typ:8,batch_read:[21,38],batch_siz:[12,49,52,56,73,85,86,89,91,93,94],batchsiz:[8,67],bazel:18,bbbbb:21,bbox:[7,8],bcd:8,bcebo:12,bcm2708:59,bdist_wheel:42,beam:[8,64,94],beam_gen:[8,64],beam_search:[14,63,64],beam_siz:[8,64,74,75,77],beamsiz:94,becaus:[7,8,12,15,17,18,19,24,36,38,39,43,45,61],becom:[40,43],been:[18,23,92],befor:[8,9,19,26,29,31,36,38,46,49,68,84],begin:[7,8,22,24,29,31],beginiter:[14,15],beginn:64,beginpass:[14,15],begintrain:15,being:[26,38],belong:[8,39,43],below:[8,17,19,24,36,38,39,40,45],benefit:[9,26,27],bengio:8,bertolami:93,besid:[8,12,39],best:[8,18,91],best_model_path:92,besteffort:79,beta1:10,beta2:10,beta:87,better:[8,9,18,84],between:[7,8,10,14,18,19,24,34,36,40,41,43,94],bgr:87,bi_gru:9,bi_lstm:9,bia:[8,9,10,67,87],bias:8,bias_attr:[8,9,49,51,52,61,64],bias_param_attr:9,biases_:67,biasparameter_:67,biassiz:67,bidi:79,bidirect:[8,9,92,93],bidirectional_lstm:50,bidirectional_lstm_net:93,big:[40,84],bigger:19,bilinear:8,bilinear_interpol:8,bilinearfwdbwd:72,bin:[54,73,78,79,80,91],binari:[7,8,12,18,27,39],bind:[30,43],bird:86,bitext:94,blank:8,bleu:94,bloack:32,block:[8,20,22,24,32,35,37,39,40,45,69],block_i:8,block_x:8,blockdesc:[17,31,37,41],bn_bias_attr:9,bn_layer_attr:9,bn_param_attr:9,bollen:93,book:[12,41,64],bool:[6,7,8,9,11,12,14,17,44,45,67,75,77,93],boot:[8,63,64,84],boot_bia:8,boot_bias_active_typ:8,boot_lay:[61,64],boot_with_const_id:8,bootstrapp:84,bos_id:[8,64],both:[5,6,8,9,15,17,18,19,26,39,40,45],bottom:[14,89],bound:8,boundari:39,boundri:7,box:8,brace:[17,31],brain:26,branch:[8,15,17,18,32,39,42,66],brendan:93,brief:[18,24,69],broadcast:[19,41,84],broken:66,browser:54,bryan:93,bsd:65,buf:22,buf_siz:12,buffer:[12,22,38,43],buffer_s:12,buffered_read:38,bug:66,build:[8,12,18,27,31,39,42,44,54,58,59,65,66,70,80,81,82,94],build_dict:12,build_doc:70,built:[18,39,44,72,84],bunk:93,c11:33,c703c041:66,c99:34,cach:[39,49,92],cache_pass_in_mem:[2,49,92],cachetyp:[2,49,92],cacul:9,caff:17,caffe2:17,caffe_poli:51,calc_batch_s:[2,92],calcul:[7,8,9,10,16,19,24],call:[7,8,9,14,15,16,17,22,23,24,25,27,31,35,37,39,41,43,44,72,80,89],callabl:[6,8,12],callback:67,caller:16,calrnn:61,caltech:86,can:[6,7,8,9,12,14,15,16,17,18,19,22,23,26,27,30,31,32,35,37,38,39,40,41,44,45,72,84],can_over_batch_s:[2,92],cancel:26,candid:8,cannot:[41,43,46],capabl:36,capi:33,capi_prvi:34,caption:94,card:73,care:[9,27,38,84],caret:14,cast_to_op_attr:44,cat:[54,80,86,87,93],categori:[8,12,19],categorig:12,categoryfil:79,caus:[19,29],cc_:18,cc_binari:18,cc_test:18,ccb2_pc30:94,cclient:25,cde:8,cdn:12,cdot:8,ceil:8,ceil_mod:8,cell:[8,9],cento:84,ceph:[21,78],cephf:[21,27,28],certain:43,certif:[15,28,78],cffi:33,cfg:79,cgo:33,chain:[12,31],challeng:[8,19],chanc:15,chang:[8,12,18,23,27,36,37,38,43,66,93],channel:[8,9,72],chapter:35,chapter_data:35,chapter_out:35,check:[2,12,17,18,41,51,66,67,75],check_align:12,check_attr:44,check_eq:67,check_fail_continu:2,check_grad:[16,68],check_l:67,check_sparse_distribution_batch:[74,75],check_sparse_distribution_in_pserv:[74,75],check_sparse_distribution_ratio:[74,75],check_sparse_distribution_unbalance_degre:[74,75],checker:41,checkgrad:75,checkgrad_ep:75,checkmark:84,checkout:66,checkpoint:40,checksum:28,child:17,choic:18,chunk:[23,28],chunk_schem:7,chunktyp:7,cifar:86,cifar_vgg_model:86,circl:31,claimnam:80,clang:[33,54,66],clarifi:7,class1:93,class2:93,class_dim:93,classic:8,classif:[8,31,89,93,94],classifi:[8,87],classification_cost:[39,49,61,86,89],classification_error_evalu:[7,89,93,94],classification_evalu:7,clean:[17,18,36,41,46,66],clear:[7,18,36,43],clearer:36,clearli:43,client:[22,25,41,78],clip:[6,9,75,89],clock:8,clone:[8,54,59,65],close:[2,38,66],cloud:[18,19,27,28,29,41,84],cloud_read:12,cluster:[12,14,15,17,19,24,39,73,78,80],cluster_train:73,cmake:[34,46,53,58,59,65,66,68,70,72],cmake_build_typ:[58,59],cmake_c:[58,59],cmake_system_nam:[58,59],cmakelist:[18,67],cmatrix:[33,34],cna:8,cnn:[8,79,89],coars:30,code:[2,4,8,12,15,18,26,30,31,36,38,39,40,41,44,45,54,67,79,91],codebas:41,coded_stream:51,codedinputstream:51,coeff:8,coeffici:8,collabor:19,collect:[8,12,14],collectbia:67,color:[86,87],colour:12,column:[7,8,31,38],column_evalu:7,com:[8,9,12,18,54,58,59,65,66,78,79,84,87],combin:[7,8,9,12,14,43],comma:[14,24],command:[18,22,27,29,65,67,77,79,80,81,82],commandlin:[72,80],comment:[10,18,44,61,80],commit:[18,79],common:21,common_util:73,commonli:29,commun:[19,24,25,39,40],compar:[16,18,41],compare_grad:68,comparison:18,compat:30,compelet:41,compil:[8,18,37,39,44,45,58,59,65],complaint:18,complet:[8,9,10,12,14,17,19,23,24,28,31,41,79,80,84],complex:[9,26,38,41],complic:[8,30,39],compon:39,compos:[12,15,30,35,41],composenotalign:12,composit:30,compress:23,comput:[8,9,15,16,19,39,40,43,45,68,69,92,93],computaiton:41,computation:8,conat:8,concat:[64,94],concaten:[9,35,49],concentr:41,concept:[7,15,30,35,36,43,45],concern:15,concurr:[19,26,40],cond:[17,32],condit:[8,23,39,79],conf:[4,8,51,61,73,80,85,87,94],conf_paddle_gradient_num:80,conf_paddle_n:80,conf_paddle_port:80,conf_paddle_ports_num:80,conf_paddle_ports_num_spars:80,confid:8,confidence_threshold:8,config:[6,8,21,29,52,56,67,73,74,75,77,78,79,80,84,86,89,91,92,93,94],config_:[22,75],config_arg:[74,75,77,87,89,92,93],config_bas:[7,8,14],config_fil:92,config_gener:[73,91],config_lay:67,config_len:24,config_pars:[4,67],config_proto:24,configprotostr:51,configur:[8,14,22,24,26,27,40,67,84,85,87,94],confirm:29,conflict:[43,66],confront:94,conll05st:92,conll:[12,92],connect:[9,27,39,40,47,79,84,89],connectionist:[8,93],connor:93,consequ:[8,9],consid:[7,8,10,84],consider:9,consist:[7,8,12,23,38,44],consolid:17,const_cast:68,constant:51,constraint:[39,43],construct:[7,15,35,41,43,44,45],constructbackwardgraph:31,constructoptimizationgraph:31,constructor:[41,43,44],consum:19,contact:26,contain:[2,7,8,9,11,12,14,15,17,23,36,41,43,44,65,79,80],content:[24,29],content_len:24,context:[2,8,9,12,43,49,64,68,69,78],context_attr:9,context_len:[8,9,89],context_proj_layer_nam:9,context_proj_param_attr:9,context_project:9,context_start:[8,9,89],continu:[7,19],contrast:8,contributor:41,control:[6,17,79,84,94],conv:9,conv_act:9,conv_batchnorm_drop_r:9,conv_bias_attr:9,conv_filter_s:9,conv_layer_attr:9,conv_num_filt:9,conv_op:8,conv_pad:9,conv_param_attr:9,conv_strid:9,conv_with_batchnorm:9,conveni:[15,44],convent:24,convers:39,convert:[2,4,12,21,38],convlay:8,convolut:[8,9,12],convoper:8,convtran:8,convtranslay:8,cool:66,coordin:[19,24],copi:[14,15,23,26,29,31,35,91],core:[6,34,36],coreo:84,corner:41,corpora:94,corpu:[12,92],correct:[8,16],correctli:[7,12],correspoind:15,correspond:[15,17,18,35,41,44,51],corss_entropi:15,cos:[8,44],cosin:[8,44],cosineop:44,cosineopproto:44,cosineopprotomak:44,cost:[10,14,15,31,37,39,52,56,91,93,94],cost_id:8,cost_val:39,could:[8,12,14,15,16,23,35,36,37,38,39,40,41],count:[7,19,27,38,72,75,77,79,91,92,93,94],counter:[19,23,31],cours:[7,27],cpp:[16,22,30,33,34,36,40,45,51,61,67,72,80,89,91,94],cpu:[2,6,8,16,27,36,39,41,42,54,65,68,69,72,77,79,80,92],cpuinfo:54,cpuplac:[68,69],cpusparsematrix:34,crash:[19,72],creat:[6,8,12,14,15,16,17,19,24,28,29,30,31,35,36,45,51,56,66,67,79,80,84],create_bias_paramet:67,create_doc_str:44,create_ifelseop:[17,32],create_input_paramet:67,create_op:68,create_oper:30,create_python_ops_creatation_funct:44,create_rnn:17,create_rnn_op:35,createfromconfigproto:4,createop:44,createoper:17,createvar:45,createvari:17,creation:30,creator:[12,21,41],credenti:29,crf:92,critic:93,crlf:66,crop:87,crop_siz:87,cross:[8,51,58,59,89],cross_entropi:[8,15],cross_entropy_with_selfnorm:8,crt:[28,78],csc:67,cslm:94,csr:67,csv:51,ctc:7,ctc_evalu:7,ctest:[54,65,66,68],ctrl:[65,73,91],ctx:[68,69,92],ctx_0:92,ctx_0_slot:92,ctx_n1:92,ctx_n1_slot:92,ctx_n2:92,ctx_n2_slot:92,ctx_p1:92,ctx_p1_slot:92,ctx_p2:92,ctx_p2_slot:92,cub:86,cuda:[18,65,72,73,75],cuda_dir:[74,75],cuda_so:[46,54],cuda_visible_devic:49,cudaconfigurecal:72,cudadevicegetattribut:72,cudaeventcr:72,cudaeventcreatewithflag:72,cudafre:72,cudagetdevic:72,cudagetdevicecount:72,cudagetdeviceproperti:72,cudagetlasterror:72,cudahostalloc:72,cudalaunch:72,cudamalloc:72,cudamemcpi:72,cudaprofilerstart:72,cudaprofilerstop:72,cudaprofilestop:72,cudaruntimegetvers:72,cudasetdevic:72,cudasetupargu:72,cudastreamcr:72,cudastreamcreatewithflag:72,cudastreamsynchron:72,cudeviceget:72,cudevicegetattribut:72,cudevicegetcount:72,cudevicegetnam:72,cudevicetotalmem:72,cudnn:[8,11,18],cudnn_batch_norm:8,cudnn_conv:8,cudnn_conv_workspace_limit_in_mb:[74,75],cudnn_convt:8,cudnn_dir:[74,75],cudnnv5:53,cudrivergetvers:72,cuinit:72,cumul:8,curl:78,curli:[17,31],current:[2,8,17,18,19,22,24,35,36,37,40,43,45,78],current_block:37,current_oper:37,current_word:[49,64],currentcost:[89,91,93,94],currentev:[89,91,93,94],curv:15,custom:[15,27,41],custom_batch_read:38,cut:40,cutoff:12,cxx_compil:[58,59],cxx_flags_minsizerel:[58,59],cycl:19,cyclic:8,cython:33,dalla:2,dan:92,dandroid_abi:58,dandroid_arm_mod:58,dandroid_arm_neon:58,dandroid_standalone_toolchain:58,darwin:78,dat:[21,73,91],data:[2,7,10,14,15,16,17,21,22,23,28,31,35,36,37,40,41,43,44,45,49,54,56,61,64,69,73,74,79,80,81,85,86,87,89,91,92,93,94],data_batch:49,data_config:4,data_dir:[85,86,93,94],data_fil:52,data_lay:[2,22,49,52,61,86,89,92],data_read:[12,38],data_reader_creator_random_imag:38,data_typ:[12,45,56,64],databas:12,datacent:[21,29],datacenter1:21,datacenter2:21,datacenter_1:21,datacenter_2:21,datacenter_nam:21,datadim:8,dataprovid:[1,49,51,52,73,80,89,91,92],dataprovider_:89,dataprovider_bow:89,dataprovider_emb:89,dataproviderconvert:4,datasci:8,dataset:[21,27,38,56,64,87,90,93,94],datatyp:45,date:92,db_lstm:92,dcmake_build_typ:70,dcmake_install_prefix:[58,59],dcmake_system_nam:[58,59],dcudnn_root:53,ddim:69,dead:19,deal:84,deb:66,debian:54,debug:[16,29,39,70],decai:10,decar:12,decayr:22,decent:23,decid:[15,26,38],declar:[8,17,35],decod:[8,9,63,64,94],decoder_boot:64,decoder_group_nam:64,decoder_input:[49,64],decoder_mem:64,decoder_prev:9,decoder_s:[49,64],decoder_st:[9,64],deconv:8,deconvolut:8,decor:[2,12],deduc:41,deep:[8,26,31,72,86,87],deer:86,def:[2,4,8,12,15,16,21,27,30,35,38,44,45,49,51,52,56,61,64,67,68,80,87,89,92],defalut:8,default_decor:80,default_devic:77,default_valu:[32,77],defect:36,defer:26,defin:[2,8,9,12,14,15,17,18,19,26,30,35,37,38,40,41,43,44,49,56,68,89],define_py_data_sources2:[2,51,52,86,87,89],defini:94,definit:[17,19,23,39,44,85],definiton:30,degre:8,delet:[27,28,66],delimit:[7,51,91],deliv:84,delta:[8,16],delv:8,demand:19,demo:[4,8,12,41,73,79,81,85,86,87,89,91,93,94],dens:[8,12,24,25,91],dense_arrai:12,dense_vector:[2,4,12,52,56],dense_vector_sequ:12,dep:18,depend:[17,18,19,27,39,40,45,84],deploi:[8,84],deploy:[41,84],deprec:8,depth:17,dequeu:40,deriv:[5,15],desc:[17,44],desc_:17,descent:[8,19],describ:[15,17,18,23,36,37,41,44,45,79],descript:[7,17,18,45,80],deseri:[14,36,41],desgin:31,design:[8,12,22,33,84],desir:[19,79],destin:[24,29],destroi:17,destruct:43,det_output:7,detail:[6,7,8,9,10,16,23,27,29,35,39,43,84],detect:66,detection_evalu:7,detection_output:7,determin:[8,12,17,41],dev:[41,46,54,65,66,84,86,94],dev_ctx:17,develop:[18,36,42,66,94],deviat:6,devic:[6,36,39,41,46,54,69,77],device_context:68,device_context_:68,devicecontext:[17,68],deviceid:77,devid:8,dez:93,dhcp:84,diagram:35,dict:[2,7,12,14,51,61,80,91,93,94],dict_dim:[49,61,93],dict_fil:[7,61,92],dict_siz:[12,22],dictionari:[2,7,8,12,14,15,16,49,89,94],dictrionari:89,dictsiz:94,did:36,diff:66,diff_mat:16,differ:[7,8,14,17,18,19,24,26,39,40,43],differenti:30,difficult:[7,16],dig:41,digit:8,dilat:8,dilation_i:8,dim0:68,dim1:68,dim:[8,12,22,35,41,45,67,68,69,85,93],dim_:69,dimens:[5,8,11,12,41,49,69],dimes:8,dir:[58,73,80,87,91,92,93,94],direct:[8,9],directli:[9,18,25,27,36,39,44],directori:[8,18,21,26,28,29,66,72,79],disabl:51,discard:[12,19,23,66,75],discexp:51,discount:8,discov:19,discuss:[15,17,23,24,25,39],dispatch:[36,39],displai:[27,29],disput:94,dist:[42,46],dist_train:[15,27],distanc:7,distinguish:18,distribut:[8,17,23,24,25,26,40,41,45,75,81,82,84],distribute_test:[74,75],distributedli:39,disucss:15,divid:[10,44,45],diy_beam_search_prob_so:[74,75],dmkl_root:53,do_forward_backward:38,doc:[4,12,35,54,68,70,80],doc_cn:70,doc_them:70,docker:[42,46,54,66,79,80,81,82,84],docker_build:15,docker_push:15,dockerfil:[54,65,66,80],dockerhub:54,document:[9,16,28,31,35,41],documentari:2,doe:[9,19,23,24,26,27,36,38,39,40,45],doesn:[6,8,12,15,17,38],dog:[86,87],doing:[22,26,31,39,41],don:[9,15,18,30,31,38],done:[7,8,9,18,19,23,24,39,66,72,80],dot:[68,94],dot_period:[75,77,80,86,91,93,94],dotmuloper:8,dotmulproject:8,doubl:[8,31,39,68,75],down:72,download:[12,18,19,22,26,28,79,84],download_cifar:86,doxygen:66,dozen:18,dpython_execut:46,dpython_include_dir:46,dpython_librari:46,drive:43,drop:9,drop_fc:50,drop_rat:[6,50],dropout:[6,9,50],dropout_r:[8,50],drpi_arm_neon:59,drpi_toolchain:59,drwxr:79,dst:24,dtoh:72,dtype:[4,51,52,87],dubai:94,due:[23,26],dummi:[14,23],duplic:40,durat:23,dure:[2,8,9,17,19,23,26,27,41,84,94],dwith_c_api:[34,58,59],dwith_doc:70,dwith_gpu:[53,59,70],dwith_mkldnn:70,dwith_mklml:70,dwith_profil:72,dwith_python:[34,59],dwith_swig_pi:[34,58,59],dwith_test:68,dwith_tim:72,dynam:[2,24,34,38],dynamic_cast:67,e2e:84,each:[2,7,8,9,11,12,14,16,17,18,19,22,23,24,26,27,31,35,36,37,38,39,40,41,43,44,84,91],each_feature_vector:5,each_pixel_str:2,each_time_step_output:5,each_word:2,eaqual:8,eas:[12,38],easi:[16,38,41],easier:[15,38,40],easili:[15,38,43],ec2:78,echo:[46,54,91,93],edit:7,edu:[12,79,86],effici:[8,39],effort:39,efg:8,eigen:[41,68],eigen_use_gpu:68,eigenmatrix:69,eigentensor:69,eigenvector:69,either:[8,12,14,15,35],electron:79,elem_dim:8,element:[7,8,9,12,14,16,23,32,38,40,41,68],element_typ:24,elementori:41,elementwis:41,elif:[15,44],els:[15,22,27,39,40,43,44,54,61,67,87],emac:65,emailweixu:18,emb1:[22,61],emb2:[22,61],emb:[49,51,61,79,89],emb_group:61,emb_para:51,emb_param_fil:51,emb_sum:49,embed:[15,17,22,40,51,64,85,91,93],embedding_lay:[22,49,61,89],embedding_nam:[8,64],embedding_s:[8,64],empir:8,emplace_back:67,emploi:44,empti:[7,12,19,52],enabl:[6,8,17,18,23,40,65,72],enable_grad_shar:[74,75],enable_parallel_vector:75,enc_proj:[9,64],enc_seq:9,enc_vec:64,encapsul:24,encod:[9,23,61,94],encoded_proj:[9,64],encoded_sequ:[9,64],encoded_vector:64,encoder1:61,encoder1_expand:61,encoder1_rep:61,encoder2:61,encoder2_rep:61,encoder_last:8,encoder_s:64,encount:22,encourag:39,end2end:84,end:[7,8,14,17,36,38,43,64,66,92,93],end_pass:15,end_po:8,enditer:[14,15,56],endpass:[14,15],endpoint:[12,21],endtrain:15,engin:27,english:[8,94],enlarg:68,enough:17,enqueu:40,ensembl:9,ensur:[19,43],enter:17,enterpris:41,entir:[8,9,24,26],entiti:[7,17,43],entri:[12,23,27],entropi:[8,89],entry_point:27,enueu:40,enumer:[5,51],env:[49,80],environ:[15,46,72,79],eol:66,eos_id:[8,64],epsilon:10,equal:[8,9,19,61,68],equat:[7,8,9,10,68],equival:[8,15,17,44,84],error:[6,7,8,9,10,15,16,23,29,43,46,51,68,75,89,91,93,94],error_clipping_threshold:[6,49],especi:9,essenc:15,essenti:[8,15],estim:[8,15,40],eta:79,etc:[7,12,17,38,39,43,84,94],etcd:[12,14,19,23,24,26],etcd_addr:24,etcd_endpoint:12,eth0:[73,80],eval:[7,17,41,45,89,91,93,94],eval_bleu:94,evalu:[8,13,14,26,39,72,73,91,93,94],evaluate_difficult:7,evaluate_pass:93,even:[15,38],evenli:24,event:[56,79],event_handl:[14,15,56],everi:[7,8,9,12,15,19,23,24,26,31,40,43,44],everyth:[39,40],evid:36,exactli:[8,9,49],exampl:[7,8,9,10,12,14,17,27,29,30,31,35,36,38,40,41,87],exc_path:46,exceed:8,except:[8,12,26,31],exchang:[36,41],excluded_chunk_typ:7,exconv:8,exconvt:8,exdb:12,exe:78,exec:54,execut:[18,19,23,27],executioncontext:[68,69],exist:[15,17,19,29,38,44,69,93],exit:[24,29,79],exp:51,expand:61,expand_a:[8,60,61],expand_lay:61,expand_level:[8,60],expandconvlay:8,expandlevel:60,expans:8,expect:8,expert:18,expir:19,explain:[7,19,30,31],explan:[8,16,27,39,43],explicit:67,explicitli:15,explor:8,exponenti:5,expos:25,express:[15,40],ext:8,extend:[7,40],extens:[26,40],extent:34,extern:[18,33,34,41],external_librari:18,extra:[6,8,9,39,84],extra_lay:14,extraattr:[6,77],extraattribut:8,extraattributenon:8,extract:[7,8,36,39,85,87,92],extract_fea_c:87,extract_fea_pi:87,extralayerattribut:[6,9,49,50],extralayeroutput:9,extrem:8,f1205:51,f120da72:79,fa0wx:79,fabric:73,face:18,facotr:8,factor:[6,8,10],factori:33,fail:[19,23,46,51,75,79],failur:[19,24],fake_imag:38,falloc:28,fals:[6,7,8,9,10,12,16,17,32,35,38,49,52,56,61,64,67,68,77,79,85,89,92,93,94],false_block:[17,32],false_label:38,false_read:38,faq:83,fast:[8,23,72],faster:[8,9,19],fault:[14,23,41],fbd1f2bb71f4:79,fc1:[45,67,77],fc2:[45,77],fc3:77,fc4:77,fc_act:9,fc_attr:9,fc_bias_attr:9,fc_layer:[44,49,51,52,61,77,89,91],fc_layer_nam:9,fc_mat:14,fc_op:44,fc_out:17,fc_output:44,fc_param_attr:9,fc_without_b:17,fclayer:67,fcop:30,fdata:[61,92],fea:87,fea_output:87,feat:93,featur:[2,5,8,12,66,87,91,92],feature_map:91,feed:[9,14,15,31,35,39,56],feeder:12,fernan:93,festiv:2,fetch:[12,19,22,39,46,66],few:[18,19,38],fewer:8,fg0:8,field1:14,field2:14,field:[8,14,17,44,49,91],fifth:31,figur:[15,18,35,40,85,94],file1:94,file2:94,file:[2,7,8,12,14,15,18,19,21,23,24,26,27,28,29,34,38,39,45,66,68,84,87,90,93,94],file_list:2,file_nam:[51,52,61,87,89,92],filenam:[2,21,49,91],fileoffset:28,filer:8,filesystem:[26,27,28,39],fill:[8,19,23],fill_zero_grad:41,filter:[8,9],filter_s:[8,9],filter_size_i:8,find:[8,10,17,19,26,43],find_var:16,findop:17,findvar:[17,43],fine:[6,23,30],finish:[19,23,26,27,44,79],first:[8,14,15,17,19,23,26,27,29,31,35,37,39,41,44,68,69,84,89],first_seq:64,firstli:[7,8],firstn:12,firstseen:79,fit:[12,41],five:37,fix:[6,33,39,66],flag:12,flatten:69,flatten_result:49,flatten_s:45,flexiabl:38,flexibl:[8,9,15,24,31,35,39],flight:94,float32:[4,12,30,38,51,52,68,87],floor:[8,51],flow:[17,35,42],fmt:51,fname:51,folder:[18,21,27,29,94],follow:[7,8,9,10,12,14,15,16,17,18,19,23,27,30,31,35,37,38,40,41,43,44,81,82,84],forbid:15,force_load:33,forest:17,forget:[10,15],fork:8,form:[8,9,10],format:[7,12,14,16,23,66,67],former:[15,18],formula:[8,9,16],formular:8,forward:[5,8,9,16,17,22,24,36,41,67],forward_op:16,forwardactiv:67,forwardtest:4,found:[8,43],four:[7,31],fp16:[41,45],fp32:[41,45],fp64:45,fparam:51,frame:[7,41],framework:[15,17,41,43,44,45,68],free:[12,84],french:94,frequenc:12,frequent:[23,38],fresh:26,friend:43,frog:86,from:[2,4,7,8,9,12,14,16,17,18,19,21,22,23,24,29,30,31,32,35,36,38,39,40,41,43,46,52,54,63,68,69,72,79,84,85,86,89,91,92,93,94],from_no_sequ:[8,60],from_sequ:[8,60],from_tar:14,fromfil:[38,51,52,87],fromstr:51,fulfil:72,full:[8,19,26,35,84],full_matrix_project:[9,61,64],fullfil:45,fulli:[39,40,84,89],fullmatrixproject:8,fullsiz:22,fully_matrix_project:9,fullyconnectedlay:67,func:[2,12,23],funciton:9,functor:30,fundament:[40,41],further:[8,44,84],futur:[8,26,39,41],fvs:44,gain:8,gamma:87,gan:15,gangliao:18,gate:[8,9],gate_act:[8,9,61],gate_attr:8,gate_bias_attr:8,gate_param_attr:8,gate_recurr:8,gather:[8,36,68],gauss:6,gcc:[33,41,54,58,59,65],gce:78,gcepersistentdisk:78,gcreators_:44,gen:[8,94],gen_conf:94,gen_data:94,gen_proto_pi:70,gen_rand_param:51,gen_result:94,gender:[12,80,91],gener:[2,7,8,9,12,14,15,16,17,18,19,21,23,24,26,30,32,37,38,39,40,41,44,72,77,80,85,94],generated_word_embed:8,generatedinput:[8,63,64],genr:[80,91],geq:8,gereat:7,get:[2,7,8,12,14,16,17,18,19,23,24,26,27,28,35,41,43,44,45,67,79,86,91,92,93],get_all_op_proto:44,get_batch_s:92,get_best_pass:93,get_config_arg:[77,93],get_data:[79,89,92],get_default_block:45,get_dict:12,get_dim:16,get_embed:12,get_float_el:16,get_grad:14,get_imdb:93,get_input_lay:67,get_model:87,get_movie_title_dict:12,get_numeric_gradi:16,get_numerical_gradi:16,get_output:16,get_sample_from_lin:49,get_shap:14,get_support:46,get_tensor:16,get_vari:17,get_word_dict:12,getbatchs:67,geteigendevic:69,getenv:[15,27,80],gethostbynam:80,gethostnam:80,getidmap:80,getinput:67,getinputgrad:67,getinputvalu:67,getlayeroutput:14,getmat:22,getoptconfig:22,getoutputgrad:67,getoutputvalu:67,getparam:22,getparameterconfig:22,getparameterptr:67,getparameterspars:22,getparametersremot:22,getplac:[68,69],getpodlist:80,getsiz:67,gettask:23,gettranspos:67,getw:67,getweight:67,getwgrad:67,gildea:92,gist:9,git:[42,46,54,59,65,66],github:[9,18,54,59,65,66,87],give:[2,19,35,41],given:[8,12,14,24,26,30,38,40],glide:18,global:[6,15,17,18,19,36,41,43,44,65,72],global_learning_r:6,globalstat:72,globalstatinfo:72,globe:2,glusterf:78,gnueabihf:59,go_librari:18,go_test:18,goal:[31,41],gob:23,godep:18,godoc:33,goe:[8,9,19,43],going:[30,84],golang:18,good:[8,38,84],goodfellow13:8,googl:[15,41,51,58],googleapi:78,got:43,gprotos_:44,gpu:[6,8,11,16,27,36,39,41,42,46,54,65,68,72,77,84,87,92,93,94],gpu_id:[49,75,77],gpu_per_train:39,gpugpu_id:74,gpukernel:41,gpuplac:68,grab:19,grad:[16,24,75],grad_op_class:41,grad_op_typ:41,grad_share_block_num:[74,75],grad_var_nam:16,gradient:[6,7,8,10,14,19,23,31,36,41,75,80,89],gradient_check:68,gradient_clipping_threshold:[6,49,89,93],gradient_evalu:7,gradient_flat:16,gradient_machin:[14,34],gradientcheck:68,gradientmachin:[4,14,34,36,80,91,94],gradientmachine_:22,grain:30,gram:85,grandient:14,graph:[8,14,17,18,19,35,37,39,45],grave:93,great:[40,84],greater:8,greaterthan:44,grep:[54,93],groudtruth:64,ground:[7,8,94],group:[9,23,84],group_input1:64,group_input2:64,group_input:[61,64],grouplen:[12,90],grpc:84,gru:[8,89],gru_bias_attr:9,gru_decod:64,gru_decoder_with_attent:64,gru_encoder_decod:[85,94],gru_layer_attr:9,gru_memori:9,gru_siz:89,gru_step:64,gru_step_lay:9,grumemori:[9,50,64],gserver:[8,67],gsizex:72,guard:22,guid:[28,79],gzip:[23,79],hadoop:15,hand:41,handi:18,handl:[12,15,27,36,38,39,43],handler:[14,17],handwrit:93,happen:[23,44],hard:[39,40],harvest:89,has:[7,8,9,10,12,15,16,17,18,19,23,24,26,32,39,40,41,44,72,84,92],has_kei:14,hassubseq:61,have:[8,9,12,15,16,17,18,19,23,24,26,27,30,31,32,35,36,38,39,40,43,84],hdf:21,head:[68,93],header:[24,34,51,52,87],heavili:41,height:[8,12,17,33,38,51,67,68],held:19,hello:15,help:[4,17,29,41,45,66],helper:[8,39],henc:43,here:[6,7,8,9,12,15,18,19,25,29,31,38,44,84],heterogen:[39,40],heurist:[8,40,94],hidden:[8,9,39,50,51],hidden_a:51,hidden_b:51,hidden_dim:61,hidden_out:17,hidden_s:9,hierach:63,hierarch:[8,41,61],hierarchi:41,high:[6,84],higher:30,highest:[12,17],highli:12,him:15,hint:[4,40],histor:30,hl_get_sync_flag:67,hold:[15,19,23,25,39,43,44,69],holder_:69,home:[21,29,39,73,79,80],honor:23,hook2:61,hook:[6,61,92],hookattr:6,hookattribut:6,horizont:8,hors:86,horst:93,host:[18,27,73,79],host_c:[58,59],hostnetwork:80,hostpath:[78,79,80],hot:89,hous:[2,12],how:[6,8,15,17,19,23,29,30,36,44],howardjohnson:61,howev:[8,9,16,26,36,38,39,44],howto:80,hpp:33,html:[12,86],htod:72,http:[8,9,12,18,27,54,58,59,65,66,78,79,84,86,87,90,94],huber:8,huge:8,huina:93,human:8,hyper:8,hyperplan:12,i0601:91,i0706:94,i0719:94,i1116:80,i1117:72,ib0:73,ics:12,icwsm:93,id_input:7,idea:[8,18,38],identityoffsetproject:8,identityproject:8,idmap:80,ids:[7,8,49,91],idx:[23,67],ieee:93,ies:29,ifelseop:[32,37],ifop:32,ight:8,ignor:[2,8],iil:51,ijcnlp:93,illustr:[7,19,24,30,35,39],ilsvrc:87,imag:[11,12,15,31,37,38,45,54,65,66,79,80,81,82,84,86,87,94],image_a:38,image_b:38,image_classif:86,image_fil:38,image_lay:38,image_list_provid:87,image_nam:15,image_path:38,image_provid:86,image_reader_cr:38,image_s:87,imagenet:[8,21],imagepullpolici:80,imageri:8,images_reader_cr:38,imagin:31,imdber:93,img:[2,8,9,39,86],img_conv_lay:9,img_pool_lay:9,img_siz:86,imgsiz:72,imgsizei:72,imgsizex:72,immutable_paramet:15,imperfect:41,implement:[8,9,10,12,17,23,24,25,26,27,30,33,34,36,39,40,41,43,44],implemet:22,impli:18,implicitli:37,imposs:84,improv:[8,40,41],inarg:[4,22],inc_path:46,includ:[7,8,9,12,15,17,18,24,27,33,34,35,39,41,44,58,59,66,68,72],incorpor:8,incorrect:8,increas:[19,23,51],increment:31,incupd:67,inde:[12,38],independ:[8,16,24,43,84],index:[7,8,11,12,14,16,17,19,23,37,61,91],indexslot:8,indic:[7,8,17,24,35,37],individu:19,industri:[19,84],ineffici:36,infer:[15,17,19,33,41],inferior:26,infershap:[17,37,41,45,68,69],infershapecontext:[68,69],infiniband:73,info:[7,8,12,35,56,61,67,73,80,84],inform:[8,12,14,17,27,29,35,39,43],inherit:[17,41],ininst:15,init:[6,14,17,35,39,45,56,67,80,92],init_from_tar:14,init_hook:[61,89,92],init_model_path:[74,75,77,85,89,92],initi:[2,6,8,9,12,14,18,23,31,35,39,40,44,45,75,89],initial_max:[6,51],initial_mean:[6,8,51],initial_min:[6,51],initial_std:[6,8,51],initpaddl:4,initrd:84,inlcud:9,inlin:69,inner:[49,61],inner_:61,inner_mem:61,inner_param_attr:9,inner_rnn_output:61,inner_rnn_st:61,inner_rnn_state_:61,inner_step:61,inner_step_impl:61,inproj_attr:8,inproj_bias_attr:8,inproj_param_attr:8,input0:69,input1:[8,9,69],input2:8,input:[2,5,7,8,9,11,12,14,16,17,22,26,30,31,32,35,36,38,39,40,41,43,44,45,49,50,51,52,56,60,61,63,64,67,68,69,77,80,85,86,89,91,92,94],input_conf:8,input_data:67,input_data_target:67,input_featur:5,input_fil:[52,92],input_hassub_sequence_data:67,input_id:8,input_imag:[9,86],input_index:67,input_label:67,input_lay:67,input_loc:8,input_nam:15,input_proj_bias_attr:9,input_proj_layer_attr:9,input_seq:8,input_sequence_data:67,input_sequence_label:67,input_sparse_float_value_data:67,input_sparse_non_value_data:67,input_t:67,input_to_check:16,input_typ:[49,52,61,89],input_valu:16,input_var:16,inputbuff:22,inputdef:67,inputlayers_:67,inputs_to_check:16,inputtyp:12,insert:[41,66],insid:[7,8,19,36,38,40,41],instal:[8,27,42,46,58,59,66,73,79,86,91],instanc:[8,16,17,19,21,25,32,35,39,41],instanti:19,instead:[8,9,11,18,22,27,31,38,39],instruct:[17,31],int16:45,int32:[37,45,75],int64:[28,45],integ:[2,7,8,12,23,27,33,37,56,89],integer_sequ:49,integer_valu:[2,12,49,56,61,89],integer_value_sequ:[2,12,61,64,89,92],integer_value_sub_sequ:61,integr:[7,84,92],inter:[8,39],intercept:8,interchang:31,interfac:[6,8,9,14,17,23,27,29,36,41,73,84],intergr:8,intermedi:[29,39],intern:[8,9,14],internet:[18,19,84],interpret:7,interv:8,introduc:[8,17,19,41,43,44],intuit:[26,41],invalid:[38,43],invoc:[17,18,30,41],invok:[2,8,14,36,39,40,41,44,72],iob:7,ioe:7,ip_str:80,ipc:78,ips:80,ipt:[8,44,51,61,64],ipx:84,ipython:15,is_gener:[85,94],is_kei:91,is_loc:14,is_predict:[89,93],is_seq:[8,64],is_stat:[6,51],is_tensor:44,is_test:[87,92,93],is_train:2,isinst:[4,56],ispodallrun:80,isspars:67,issu:18,issue_numb:66,item:[8,12,14,26,38,80,84],iter:[2,8,9,10,12,14,15,19,38,39],its:[2,8,9,14,15,16,17,19,23,31,35,36,41,43,44,72],itself:[19,26,43,45],ivs:44,java:[17,33,37,41],jeremi:72,jie:[92,93],jmlr:8,job:[12,26,39,41,73,74,75,77,78,80,87,89,92,93,94],job_dispatch_packag:73,job_id:12,job_mod:85,job_nam:[27,80],job_namespac:80,job_path:80,job_path_output:80,job_workspac:73,jobnam:80,jobpath:80,jobselector:80,jobserv:27,johan:93,join:[19,61],joint:94,jointli:[9,94],journal:[92,93],jpg:87,json:[73,79,91],jth:9,jupyt:[27,54],just:[5,7,8,9,12,18,23,24,35,36,39,41,43],jypyt:15,k8s:[80,84],k8s_data:80,k8s_job:15,k8s_token:15,k8s_train:80,k8s_user:15,kafka:21,kaim:8,kaimingh:87,kebilinearinterpbw:72,kebilinearinterpfw:72,keep:[8,12,19,31,43,44,84],keep_top_k:8,kei:[2,12,14,16,17,19,21,23,28,41,44,65,66,72,78,80,91],kept:8,kernel:[8,16,68,69],key1:75,key2:75,keyword:80,kill:19,kind:[15,16,19,25,31,39,78,79,80],kingsburi:92,know:[15,23],known:[17,30,35],kriz:[12,86],ksimonyan:9,kube:78,kube_cluster_tl:15,kube_ctrl_start_job:15,kube_list_containers_in_job_and_return_current_containers_rank:15,kubeadm:78,kubectl:[78,79,80],kuberent:19,kubernet:[15,19,41,71,73,80,81,82,84],kubernetes_service_host:15,kwarg:[2,9,10,12,44,61,89,92],l1_rate:6,l2_rate:6,l2regular:[49,86,89,93],l93:22,label:[2,7,8,10,12,14,31,37,38,39,40,49,52,56,61,79,86,87,89,92,93],label_dict:92,label_dim:[8,61,89],label_fil:[38,92],label_lay:38,label_list:92,label_path:38,label_slot:92,labeledbow:93,labelselector:80,lag:75,lake:2,lambdacost:8,lambdarank:8,languag:[8,12,31,41,43,85],larg:[11,12,39,40,66,94],larger:[6,7,8],last:[7,8,9,35,45,60,61],last_seq:61,last_time_step_output:8,lastseen:79,latenc:[8,39],later:[18,41,69],latest:[8,17,18,19,26,46,79,80],launcher:15,layer1:[8,9,49,60],layer2:[8,49,60],layer3:8,layer:[4,6,7,9,11,12,13,14,17,22,31,32,37,38,39,40,41,44,45,49,51,56,60,63,64,67,87,89,92],layer_0:67,layer_att:50,layer_attr:[8,49,50,64,77],layer_num:[77,87],layer_s:8,layer_typ:8,layerbas:67,layerconfig:67,layergradutil:67,layermap:67,layeroutout:8,layeroutput:9,layers_test:46,lbl:[7,86],ld_library_path:73,learn:[6,7,8,9,10,12,15,24,26,31,38,39,40,54,72,86,87,92,93,94],learnabl:14,learning_method:[52,85,86,89,93,94],learning_r:[6,24,39,49,51,52,85,86,89,93,94],learning_rate_arg:51,learning_rate_decay_a:51,learning_rate_decay_b:51,learning_rate_schedul:51,leas:19,least:[7,8,19],leav:17,lecun:12,left:[8,17],legal:44,leman:94,len:[2,8,24,28,61,67,80,89,92],length:[8,9,12,24,35,41,79],leq:8,less:[8,15,84],less_than:15,let02:79,let:[7,8,15,17,26,30,31,35],level:[6,8,30,35,63],lgtest:18,lgtest_main:18,lib64:[46,54,73,75],lib:[34,53,58,59],lib_path:46,libapi:18,libari:34,libcuda:[46,54],libjpeg:86,libnvidia:[46,54],libpaddl:[33,34,41,66],libpaddle_capi:34,libpaddle_gserv:34,libpaddle_math:34,libprotobuf:51,librari:[8,18,25,34,39,73,75],life:19,lifecycl:84,lifetim:43,lightweight:30,like:[7,8,12,17,18,19,22,27,30,31,37,38,39,41,43,84,87],limit:[8,12,41,51,72],linaro:59,line:[2,7,12,18,22,27,29,31,37,41,49,51,61,66,77,92],line_count:51,linear:[8,49,51,56],linearactiv:52,linguist:92,link:[8,9,18,28,29,41,43,63,84,93],linux:[28,54,58,59,65,78],linux_x86_64:[42,46],lipeng:85,lipton:93,list:[1,2,7,8,9,12,14,15,17,18,23,27,29,32,36,43,49,52,73,77,86,87,89,91,92,93,94],listen:19,littl:24,lium:94,liwicki:93,load:[8,15,19,39,52,80,87,92,93,94],load_data_arg:4,load_featur:87,load_feature_c:87,load_feature_pi:87,load_missing_parameter_strategi:[74,75,77,85,92],load_paramet:51,loadparamet:4,loadsave_parameters_in_pserv:[22,74,75],loc:7,local:[6,14,16,17,19,25,26,31,35,41,53,73,74,75,80],local_scop:16,localhost:[54,78],localip:80,localpath:29,locat:[8,14,18],lock:[18,19,23,24,40],lod:35,lod_level:45,lod_tensor:[35,45],lodtensor:[41,45],lodtesnordesc:45,log:[23,29,39,47,51,67,73,75,79,80,86,91,92,93,94],log_barrier_abstract:[74,75],log_barrier_lowest_nod:[74,75],log_barrier_show_log:[74,75],log_clip:[74,75],log_error_clip:[74,75],log_period:[75,77,79,80,86,89,91,92,93,94],log_period_serv:[74,75],logarithm:5,logger:[2,61],logic:[26,36,39,40,43],logist:89,longer:[19,39],look:[2,7,17,27,31,39],lookahead:8,loop:[16,17,38,43],loss:[8,89],lot:[39,84],low:8,low_rnn:35,lower:[8,35],lower_level_rnn:35,lpaddle_capi_shar:34,lpaddle_capi_whol:34,lstm:[8,61,64,79,89],lstm_bias_attr:9,lstm_cell_attr:9,lstm_group:[9,61],lstm_group_input:61,lstm_input:61,lstm_last:61,lstm_layer_attr:9,lstm_nest_group:61,lstm_output:61,lstm_size:89,lstm_step:9,lstmemori:[9,50,61,64],lstmemory_group:[8,50,61],lstmemory_unit:50,ltr:8,mac:[34,54,58],machin:[8,9,12,14,39,40,49,63,84,93,94],machine_transl:64,maco:65,macro:30,made:[19,24],mai:[8,9,16,17,38,39,40,41,43],main:[4,41],mainli:25,maintain:[8,17,23,41],majel:18,major:[39,94],make:[2,7,8,15,17,18,19,23,24,26,31,35,36,38,39,41,58,59,65,66,67,68,70,72,84,93],make_ddim:69,make_function_oper:30,make_vari:44,maker:41,man:28,manag:[14,19,24,25,29,39,43,45],mandarin:8,mani:[8,9,18,23,36,41,43,44],manner:8,manual:[39,51,84],manufactur:94,manylinux1_x86_64:[42,46],mao:93,map:[7,8,12,14,15,17,23,43,44,56,84],map_read:12,mapper:12,mapreduc:15,marcu:93,mark:[31,35,40,43,45,84,92],mark_slot:92,market:93,martha:92,mask:[6,8],master:[15,26,41,42,78,93],mastermind:18,mat:[33,34],mat_cache_row:22,mat_norm:22,mat_normal_shar:22,mat_param_attr:9,mat_sparse_row:22,mat_sparse_row_auto_grow:22,mat_sparse_row_id:22,mat_sparse_row_prefetch:22,mat_sparse_row_prefetch_full_s:22,mat_value_shar:22,match:[8,18,49],matchbox:84,math:[9,33,41,67,68,72],matirx:8,matmul:[17,35,68],matplotlib:86,matrix:[7,8,9,12,14,22,33,34,67,68],matrixptr:67,matrixtyp:34,mattyp:22,max:[6,8,12,16,51,60,72,77,91],max_diff:16,max_id:[8,14],max_job_id:12,max_length:[8,64],max_movie_id:12,max_relative_error:[16,68],max_sort_s:8,max_user_id:12,maxframe_evalu:7,maxid:7,maxid_evalu:7,maxim:8,maximum:[7,8,12,17,24],maxinum:11,maxpool:8,mayb:[8,17],md5:[12,20],mean:[6,7,8,9,10,11,12,14,18,35,38,39,43,49,75,84,87,91],mean_img_s:86,mean_meta:87,mean_meta_224:87,mean_valu:87,mean_var_nam:8,mechan:[8,9,25],meet:92,mem:[8,17,27,61],member:[12,15,31,36,43],memcpi:[36,72],memori:[9,17,22,23,27,41,64,69,72,79],memory_boot:9,memory_nam:[8,50],memory_test:65,memory_threshold_on_load_data:[74,75],mention:[18,23,40],mere:9,merg:[8,24,26,35,36,66],mergedict:[85,94],messag:[17,31,37,41,45,46,66,79],meta:[73,86,87,91],meta_config:[73,91],meta_gener:[73,91],meta_path:86,metadata:[28,79,80],metal:[78,84],metaphor:31,metaplotlib:15,method:[2,8,10,14,16,17,26,31,40,41,43,45,94],mfs:80,might:[8,17,18,37,40],million:12,min:[6,8,72,77,91],min_block:17,min_count:40,min_desc:17,min_pool_s:[2,49],min_word_freq:12,mini:[8,12,14,17,19,35],mini_batch:38,minibatch:[8,17,31],minikub:78,minim:[10,17,40,41],minimum:8,minsizerel:[58,59],minut:[19,26],mip:58,mirror:18,mislead:24,miss:92,mix:[9,64],mixed_lay:[9,61,92],mixed_layer_attr:9,mixedlayertyp:8,mkdir:[29,70],mkl:[41,53],mkl_root:53,ml_data:[73,91],mnist:[2,4,21,31,37,38,39,41],mnist_model:4,mnist_provid:2,mnist_random_image_batch_read:38,mnist_train:[2,38],mnist_train_batch_read:38,mnt:80,mobil:41,mod:92,mode:[8,14,66,80,93],model:[8,9,12,14,17,19,20,31,39,40,41,77,85,86,89,91,92,93,94],model_config:4,model_list:[75,77,92,93],model_output:93,model_path:77,model_zoo:[85,87],modifi:[8,39,66],modul:[2,9,12,14,30,51,52,68,86,87,89],modulo:8,momentum:[6,43,49,56,89],momentumoptim:[52,86],mon:79,mono:8,month:[18,94],mood:93,moosef:78,more:[7,8,9,12,15,16,18,19,23,26,27,29,30,31,35,38,39,40,41,51,72,84],morin:8,mose:[93,94],moses_bleu:94,mosesdecod:93,most:[8,12,15,18,26,31,37,38,39,45,84],mostli:84,motiv:41,mount:27,mountpath:[79,80],move:[8,19,23,29,84],movi:[2,12,91],movie_categori:12,movie_id:[80,91],movie_info:12,movie_review:12,movieinfo:12,movielen:90,moving_average_fract:8,mpi:73,mse:[31,37],much:[8,19,38],mul:[30,41,67,68],mul_grad:68,mul_op:68,mulgradkernel:68,mulkernel:68,mulop:[30,68],mulopgrad:68,mulopmak:68,multi:[8,36,84,87,94],multi_binary_label_cross_entropi:8,multi_crop:87,multigradientmachin:36,multinomi:8,multipl:[7,8,9,12,14,15,16,23,24,26,30,39,40,45],multipli:[7,8],multiprocess:12,must:[5,7,8,9,12,24,32,38,41,44,49,67,68,69],mutabl:69,mutable_data:[68,69],mxnet:17,my_cost:51,my_standalone_toolchain:58,mypaddl:[79,80],name:[2,6,7,8,9,11,14,15,16,17,19,21,22,24,27,30,34,41,42,44,45,51,52,54,56,61,64,67,72,77,78,79,80,81,82,84,86,89,91,94],name_prefix:21,namespac:[17,33,67,68,78,79,80],nativ:8,natur:[23,26,40],nchw:8,ndarrai:[14,21],ndcg:8,ndcg_num:8,ndk:58,nearli:16,necessari:[8,17,24,26,36,44],need:[7,8,9,12,15,16,18,22,23,24,26,27,29,30,36,37,39,40,41,43,44,46,72,80,84],need_tran:51,neg:[2,7,8,92,93],neg_distribut:8,neg_overlap:8,neg_pos_ratio:8,ner:7,nerual:39,nest:[8,12,17,37],net:[8,9,17,35,43,93],net_conf:93,net_diagram:87,netop:[17,41],network:[4,6,7,8,10,12,13,14,15,16,17,19,22,35,38,39,40,43,44,45,50,56,61,73,77,80,84,85,86,87,92,93,94],network_config:77,neural:[8,9,10,12,14,15,17,19,39,43,45,61,63,85,92,93,94],neuralnetwork:36,never:[12,38,43,79,80],new_stat:35,newest:24,newli:84,newop:17,newremot:39,newvar:[17,43],next:[2,8,12,19,25],nfs:80,nfsdir:80,nginx:54,ngram:12,nic:[73,74,75,80],nil:23,nine:12,nlp:8,nltk:12,nms_threshold:8,nms_top_k:8,nmt:94,nmt_without_attent:49,nnz:67,no_cach:2,no_grad_set:[16,68],no_sequ:[2,8,56,91],noah:93,noavx:54,node0:80,node:[8,18,26,39,40,41,45,78,79,80,84],node_0:80,node_1:80,node_2:80,nodebook:54,nodefil:73,nois:[8,19],non:[8,19],none:[2,4,6,7,8,9,10,11,14,15,16,35,44,45,52,64,87],norm:9,norm_by_tim:8,normal:[8,9,12,79,80,87],notat:8,note:[6,8,9,11,14,15,17,22,23,27,38,41,69,93],notebook:27,noth:[5,14,43,66],notingradi:68,notori:16,novel:93,now:[8,12,18,19,40,41,43,63],np_arrai:12,nproc:65,ntst1213:94,ntst14:94,nullptr:[43,67],num:[8,9,73,75,80,92,93,94],num_channel:[8,9,86],num_chunk_typ:7,num_class:[8,9,86],num_filt:[8,9],num_flatten_dim:45,num_gradient_serv:[74,75],num_group:8,num_neg_sampl:8,num_p:39,num_parameter_serv:15,num_pass:[14,52,56,74,75,77,79,80,89,91,92,93,94],num_repeat:8,num_result:7,num_results_per_sampl:8,num_samples_process:51,num_shard:21,num_train:39,number:[7,8,9,12,17,19,21,38,40,41,44,94],numchunktyp:7,numdevices_:77,numeric_grad:16,numerical_grad:16,numlogicaldevices_:77,numofallsampl:7,numofwrongpredict:7,numpi:[6,12,14,21,38,51,52,56,68,87],numreal:22,numsampl:72,numtagtyp:7,numtimeout:23,nv_:18,nv_gpu:65,nv_librari:18,nv_test:18,nvcc:[18,54],nvidia:[46,54,65],obei:7,obj:[2,51,52,86,87,89,91],object:[2,6,8,9,10,12,14,15,22,33,41,43,45,72],observ:[8,10],obtain:26,obvious:18,occup:[80,91],occur:[12,14],occurr:17,oct:79,odd:8,off:[34,53,54,58,59,65,66,70,84],offer:[17,41,44],offici:[8,18],offlin:[19,21,84],offset:[8,22,91],often:[22,37],ograd:67,old:[16,24,26,41],omit:[49,89],on_init:2,onc:[8,19,23,31,39,40],one:[2,5,7,8,9,10,11,12,14,15,16,17,19,22,23,24,26,27,30,32,35,36,38,39,41,43,84,89,92,93],one_host_dens:91,one_hot_dens:91,onehotcrossentropyopkernel:68,ones:[30,41],onli:[7,8,9,11,14,15,16,18,22,23,24,25,26,27,29,31,32,35,36,39,40,41,44,61,63,84],onlin:[8,10,19,21,38],only_cpu:16,onto:[39,40],op_:68,op_check:68,op_class:[41,44],op_maker_class:[41,44],op_proto:44,op_test_util:68,op_typ:41,opattrcheck:68,opcreat:44,opdesc:[17,31,37,41,44,45],opdescbuild:17,open:[2,8,15,21,38,49,51,52,61,87,92],openbla:53,openblas_root:53,oper:[8,9,10,12,16,17,31,32,35,37,39,43,45,68,69],operartor:69,operator_grad:16,operatorbas:[17,30,41,44,68],operatorwithkernel:68,opinfo:41,opinion:93,opkernel:[68,69],opkernelkei:41,opmak:44,opproto:68,opprotoandcheckermak:68,opprotomak:[44,68],opregist:44,opregistri:44,ops:[16,17,18,31,37,41,45,68,84],ops_:17,ops_test:18,opt:[15,44,53,80],optestmeta:68,optim:[6,13,14,16,36,37,39,40,41,49,51,56],optimzi:49,option:[7,8,15,18,37,44,45,84],optyp:44,order:[8,9,12,14,31,38,80,84],ordereddict:14,org:[7,8,9,12,21,28,54,90],organ:[7,8],orient:44,origin:[8,12,16,43,66],other:[7,8,9,12,17,19,24,29,35,43,44,84],otherchunktyp:7,otherwis:[8,12,15,19,24,26,32,38],our:[15,18,39,40,43,45],out:[8,14,15,17,18,23,26,32,35,45,49,61,63,64,68,69,86],out_dir:80,out_left:8,out_mem:64,out_memori:9,out_right:8,out_size_i:8,out_size_x:8,outer:61,outer_mem:61,outer_rnn_st:61,outer_rnn_state_:61,outer_step:61,outlier:8,outout_lay:14,outout_layer1:14,outout_layer2:14,output:[5,6,7,9,11,12,14,15,16,17,21,26,29,30,31,32,35,38,39,40,41,43,44,45,49,52,61,64,68,69,73,77,79,80,85,86,87,89,91,92,93],output_:8,output_all_step:35,output_dir:87,output_fil:92,output_id:8,output_lay:[14,49,87],output_max_index:11,output_mem:[8,64],output_nam:16,output_num:[17,32,35],output_path:21,output_s:45,outputbuff:22,outputh:8,outputw:8,outsid:[2,8,9,43],outv:67,over:[8,9,14,15],overal:84,overlap:[7,8],overlap_threshold:[7,8],overrid:[17,19,29,68,69],overview:[23,24,25],overwrit:29,own:[24,26,39,44],pack:51,packag:[12,23,27,30,42,46],pad:9,pad_c:8,pad_h:8,pad_w:8,paddelpaddl:37,padding_attr:8,padding_i:8,padding_x:8,paddl:[2,4,5,6,7,8,9,10,11,12,14,15,17,18,19,21,27,29,30,32,33,34,35,36,39,41,44,45,49,51,52,54,56,64,65,66,67,68,70,72,73,77,79,80,84,85,86,89,91,92,93,94],paddle_begin_init_param:24,paddle_dir:68,paddle_doc:70,paddle_docs_cn:70,paddle_element_typ:24,paddle_element_type_float32:24,paddle_element_type_float64:24,paddle_element_type_int32:24,paddle_element_type_int64:24,paddle_element_type_uint32:24,paddle_element_type_uint64:24,paddle_enforc:17,paddle_enforce_eq:[68,69],paddle_error:[33,34],paddle_exampl:27,paddle_finish_init_param:24,paddle_get_param:24,paddle_gradi:24,paddle_init_param:24,paddle_job:27,paddle_matrix:[33,34],paddle_matrix_cr:34,paddle_matrix_get_shap:33,paddle_matrix_shap:33,paddle_n:[73,80],paddle_new_etcd_pserver_cli:24,paddle_new_pserver_cli:24,paddle_on_cloud:27,paddle_output:79,paddle_paramet:24,paddle_port:[73,80],paddle_ports_num:[73,80],paddle_ports_num_for_spars:73,paddle_ports_num_spars:80,paddle_process_by_paddl:80,paddle_pserver2:73,paddle_pserver_cli:24,paddle_pserver_client_releas:24,paddle_root:85,paddle_save_model:24,paddle_send_grad:24,paddle_server_num:80,paddle_source_root:85,paddle_train:[34,42,73,80],paddledev:[46,54,79,80],paddlepaddl:[8,9,12,14,18,19,21,24,25,26,27,28,29,30,31,35,36,37,38,41,43,45,54,64,65,66,72,73,81,82,84,85,92],pair:[7,8,17,31,39,41],pakcag:18,palceholder_just_ignore_the_embed:85,palmer:92,paper:[8,94],para:[22,85],paraconvert:85,paragraph:35,paragraph_data:35,paragraph_out:35,parallel:[39,40,41,72,77,79,80,94],parallel_nn:[6,74,75],parallelli:41,param:[6,8,9,12,16,17,24,36,51,69],param_attr:[8,9,22,49,51,52,64],param_config_proto:24,param_fil:51,paramattr:[6,8,49,51,52,64],paramet:[7,9,10,11,12,13,16,17,18,20,22,26,29,31,35,37,38,39,43,44,49,51,56,75,80,91,92,93,94],parameter_attribut:8,parameter_block_s:[74,75],parameter_block_size_for_spars:[74,75],parameter_learning_r:6,parameter_nam:[14,15],parameter_serv:15,parameter_valu:22,parameterattribut:[6,8,9,22],parameterclient2:80,parameterclient_:22,parametermap:67,parametermutex_:22,parameters_:67,parameterserver2:22,parameterset:15,parameterupdat:36,parameterupdater_:22,parametris:10,paramutil:91,paraphras:[85,94],paraphrase_data:85,paraphrase_model:85,paraphrase_modeldata:85,paraspars:67,parent:[17,37,41],parent_:[17,43],pars:[12,18,65],parse_config:4,parse_known_arg:80,parsefromstr:51,parser:80,part:[7,8,17,26,41,84,93],partial:[8,14],partial_sum:8,particular:[31,41],partit:[19,21,39,40,41],paserv:80,pass:[2,8,14,17,19,36,38,41,43,45,49,52,56,66,72,75,77,79,80,86,89,91,92,93,94],pass_id:[14,56],pass_idx:38,pass_manu:51,passtyp:67,past:15,patch:28,path:[7,12,14,19,23,24,27,38,58,59,73,75,78,79,80,92,93],pattern:[12,19,33,93],paul:92,paus:[19,26],pave:94,pdf:[8,9],peer:47,pem:[15,21],pend:[19,23],penn:92,pep425tag:46,per:[7,8,12,19,24,38],perform:[8,16,24,36,41,72,74],period:[19,26,75,92,93,94],perl:[93,94],peroid:8,persistentvolum:78,persistentvolumeclaim:[78,80],person:[7,15],perspect:41,perturb:16,pex:84,pfs:[21,29],pfsclient:21,pfspath:29,phase:84,physic:84,pid:78,piec:[8,9],pillow:[27,86],pip:[42,46,66,73,86,91],pixel:[2,8,12],pixels_float:2,pixels_str:2,place:[2,19,26,39,40,41,68,69],placehold:69,placement:40,plain:[7,8,27,34],plan:[19,41],platform:[17,58,68,69],pleas:[6,8,9,10,15,19,23,24,25,35,38,39,41,46,80],plot:[15,86],plotcurv:86,plu:[8,16],png:86,pnpairvalid:74,pod:[21,27,78,79,80],podip:80,podlist:80,point:[17,19,27,69,72,84],pointer:[17,24,41,43,69],polar:12,poli:51,poll:93,pollut:26,pool3:67,pool:[9,13],pool_attr:9,pool_bias_attr:9,pool_layer_attr:9,pool_pad:9,pool_siz:[2,8,9],pool_size_i:8,pool_strid:9,pool_typ:[8,9],pooling_lay:[9,49,89],pooling_typ:[8,49,60,89],poolingtyp:11,pop:17,popul:24,popular:18,port:[18,73,74,75,79,80],port_num:74,ports_num:[75,80],ports_num_for_spars:[22,74,75,77,80],pos:[91,93],pose:19,posit:[2,7,8,9],positive_label:7,posix:78,possibl:[15,17,23,40],possiblli:41,post:[27,28],potenti:72,pow:51,power:84,practic:8,pre:[8,9,12,15,17,24,85,93,94],pre_dictandmodel:85,pre_stat:35,precis:7,precision_evalu:7,pred:92,predetermin:8,predic:[12,92],predicate_dict:92,predicate_dict_fil:92,predicate_slot:92,predict:[2,4,7,8,10,14,39,49,73,85,86,87,89,91,92,93],predict_fil:[74,75],predict_lay:14,predict_output_dir:[74,75,89],predict_sampl:4,prediction1:14,prediction2:14,predin:86,prefer:78,prefetch:[22,67],prefix:[7,19,21],pregrad:67,premodel:85,prepar:[16,27,36,81],preprocess:[12,73,85,86,89,91,93,94],present:[15,17,66],preserv:29,prev_batch_st:[74,75],prevent:[10,15,19,23,26],preview:41,previou:[8,9,14,19,29,35,40],previous:8,price:[12,41],primari:31,principl:[15,18],print:[4,6,14,15,17,39,45,46,52,56],print_arrai:17,printallstatu:72,printer:7,printstatu:72,priorbox:8,prioriti:41,prite:7,privat:[17,34,43,44,66,69],prob:[7,14],probabilist:[8,85],probabl:[7,8,14],problem:[8,10,15,16,18,26,31,41],proc:54,proce:[12,19,38],procedur:17,proceed:[8,92],process2:61,process:[2,6,8,9,12,15,17,21,22,23,26,39,44,49,51,52,61,80,89,91,92],process_num:12,processdata:[86,87],processor:72,prod:[45,54,66],produc:[8,9,12,19,32,38,45],product:[8,27],productgraph:79,profil:[29,72],proflier:72,prog:80,program:[12,15,21,24,26,31,37,38,39,41,43,72,80],programm:39,progress:[19,23],proj:8,project:[8,9,27,34],promis:[8,9],prompt:29,prone:15,prop:92,propag:10,properti:37,propos:[17,40],proposit:92,protect:[44,67,68],proto:[11,41,44,45,68],proto_:44,protobuf:[14,17,27,31,37,41,44,51],protocol:[7,84],protomak:68,provid:[8,12,15,17,24,27,44,49,52,61,74,84,89,92],provis:84,prune:[8,17],ps_desir:19,pserver:[14,22,24,25,27,41,73,74,75,80],pserver_addr:24,pserver_cpu:27,pserver_id:20,pserver_mem:27,pserver_num_thread:[22,74,75],pserver_spec:14,pseudo:[15,27],psize:67,ptr:34,pull:[18,42,54],purpos:[8,19,39,40,72],push:[17,80],push_back:67,put:[18,19,22,40,45],pwd:[54,65,66],pxe:84,py_paddl:[4,46],pybind:17,pydataprovid:49,pydataprovider2:[2,4,52,80],pyramid:8,pyramid_height:8,python:[14,15,17,25,30,31,33,36,37,41,42,46,49,54,64,65,66,67,68,73,85,86,87,91,92,93,94],pythonpath:[46,86],pzo:93,queri:[8,94],question:[8,15,40,44],queue:40,quick:79,quick_start:[27,79,80,81,89],quick_start_data:79,quickli:41,quickstart:79,r14b:58,r_t:8,rac:8,rais:12,ramnath:93,ran:[40,72],rand:[51,72,75,77,92],random:[6,8,12,21,36,38,51,52,68],random_imag:21,random_uniform:45,randomli:26,rang:[8,12,21,38,39,44,80,89],rank:[8,15,87,89],rare:2,raspberri:59,raspberrypi:59,raspbian:59,rate:[6,7,9,10,12,24,49,73,80,91],rather:27,ratio:[8,75],raw:8,raw_meta:91,rdma_tcp:[74,75],reach:19,read:[2,12,14,15,19,21,38,39,40,41,51,52,84,87],read_from_realistic_imag:15,read_from_rng:15,read_lock:20,read_mnist_imag:15,read_next_from_fil:49,read_paramet:51,read_ranking_model_data:15,readabl:41,reader:[14,21,31,37,39,56],reader_cr:21,reader_creator_bool:38,reader_creator_random_imag:[12,38],reader_creator_random_image_and_label:[12,38],readi:[19,79,84],readlockguard:22,readm:[34,66,91,93],readwritebuffer_:22,real:[8,22,38],real_process:2,realist:15,realiz:[17,35],reaon:45,reason:[9,15,16,19,79],recal:7,receiv:[19,27,40],recognit:[8,87,93],recommend:[9,15,73,80,91],record:[12,23,44,45,92],recordio:[12,15,21,23,39],recov:19,recover:41,recoveri:23,rectangular:8,rectifi:8,recurr:[43,61,62,92,93],recurrent_group:[9,50,61,63,64],recurrent_lay:9,recurrentgradientmachin:34,recurrentgroup:7,recurs:[17,18,29,41],recv:[39,40],recvparametertyp:22,reduc:[10,40,41],reduce_by_kei:41,ref:91,refactor:[31,36,39,40],refactorizaiton:41,refer:[6,8,9,10,16,17,19,23,24,25,35,43,53],referenc:[8,23],reference_cblas_root:53,refine_unknown_arg:80,reflect:23,reg:44,regard:84,region:[8,43],register_gpu_profil:72,register_lay:67,register_op:[30,41,44,68],register_op_cpu_kernel:68,register_op_gpu_kernel:68,register_op_without_gradi:[41,68],register_tim:22,register_timer_info:72,registerop:44,registr:41,registri:[27,79,84],regress:8,regular:[6,49,86,89,93],rel:[9,16,26,68],relat:[19,26,27,37,43,45,84],releas:[42,58,59,78,92],reli:[16,41],reliabl:19,relu:8,rememb:8,remot:[6,18,22,41,66,73,75,77],remoteparameterupdat:[22,25],remov:[12,29,66],renam:29,reorgan:8,repeat:[17,31,37,44,45],repeatedli:31,replac:[18,23],replic:39,replicaset:27,repo:18,report:[23,39],reportdataset:23,repositori:[8,58],repres:[8,9,17,23,37,40,41,45],represent:[8,24,31,39],request:[18,19,22,26,39,42,79,84,94],requir:[7,8,15,19,24,26,27,29,35,37,39,40,41,44,45,73,84,91],res5_3_branch2c_bn:87,res5_3_branch2c_conv:87,res:92,research:[12,39,86],reserv:29,reserveoutput:67,reset:[8,19,47],reshap:[16,38,51],reshape_s:8,residu:87,resiz:[22,68,69],resnet_101:87,resnet_152:87,resnet_50:87,resolv:[18,66,79],respect:[16,35],respons:[8,22,36,79],rest:[8,17,27,28,84],restart:[19,24,79,84],restartpolici:[79,80],restor:16,restrict:43,resu:38,result:[2,5,7,8,14,16,23,31,36,39,72,89,94],result_fil:7,resum:26,retriev:[17,43],return_seq:9,reuqest:42,reus:[17,26,38,41],rev:65,reveal:15,revers:[8,9,63,64],review:[12,66,79,93],reviews_electronics_5:79,rewrit:[18,94],rgb:8,rgen:93,rho:10,right:[8,16,17,18,27,41],rkt:[27,65],rmsprop:[49,89],rnn:[8,9,17,41,43,63,64,74,93],rnn_bias_attr:64,rnn_layer_attr:64,rnn_out:64,rnn_state:61,rnn_state_:61,rnn_step:8,rnn_use_batch:[74,75],rnnlm:12,robot:86,robust:8,role:[12,15,23,24,39,92,93],roman:93,root:[10,11,54,73,79,80,85],root_dir:73,rot:8,routledg:93,row:[7,8,12,22],row_id:8,rpc:23,rpcserver:23,rpi:59,rpi_arm_neon:59,rpi_toolchain:59,rstrip:80,rtk:84,rtype:[8,12],rule:[7,31,39],run:[15,16,17,18,19,27,30,31,35,37,39,40,41,43,45,46,54,65,66,69,72,73,79,80,81,82,84,91],runinitfunct:[72,80],runnabl:40,running_on_cloud:27,runtim:[2,17,35,41,45,46],runtime_table_:17,safe:27,sai:[8,31],same:[2,7,8,9,14,15,16,23,24,26,30,35,39,41,43,61,69,89],samping_id:8,sampl:[2,7,12,44,89,91,93,94],sample_id:7,sample_num:7,santiago:93,satifi:7,satisfi:18,save:[2,8,12,14,19,21,23,24,27,31,39,45,79,91,92,93,94],save_dir:[52,75,77,79,80,86,89,91,92,93,94],save_only_on:[74,75],savetxt:51,saving_period:[74,75,80],saving_period_by_batch:[74,75,77],saw:2,sbin:54,scalabl:41,scalar:[8,17],scale:[5,39,40,44,68],scaleop:68,scaleopmak:[41,68],scalingproject:8,scan:[23,41],scatter:8,schedul:[23,27,40],scheduler_factor:6,scheme:[7,22],schmidhub:93,schwenk:94,scienc:93,scope:16,score:[7,8,91],script:[12,70],seaplane_s_000978:86,search:[8,19,43,64,94],seat:94,second:[8,15,29,35,38,43,44,68,89],section:40,sed:93,see:[8,9,15,19,39,51],seed:[51,72,75],seem:18,segment:[7,35],sel_fc:8,selcet:8,select:8,selected_indic:8,selectiv:8,selector:79,self:[16,45,67,68],selfnorm:8,semant:[12,15,42,92,93],semat:15,sen_len:92,send:[19,24,39,40,41,44,45],send_back_parameter_typ:22,sendbackparameterspars:22,sendbackparametertyp:22,sendparameterrequest:22,sendparameterrespons:22,sensit:8,sent:[15,24,39,41,44,79],sentanc:49,sentenc:[2,8,12,35,61,64,92],sentence_last_state1:61,sentence_last_state2:61,sentiment:[2,92,93],sentiment_data:93,sentiment_net:93,sentimental_provid:2,sentimental_train:2,separ:[7,24,30,91],seq:[8,12,61,91],seq_pool:[8,60],seq_silc:8,seq_text_print:7,seq_to_seq_data:[85,94],seq_typ:12,seqlastin:61,seqtext_evalu:7,seqtoseq:[8,51,85,94],seqtoseq_net:[8,85,94],sequel:2,sequenc:[2,5,7,8,9,11,12,17,31,49,56,61,63,89,91,93,94],sequence_conv_pool:89,sequence_group:8,sequence_layer_group:61,sequence_nest_group:8,sequence_nest_layer_group:61,sequencegen:61,sequencestartposit:8,sequencetextprint:7,sequencetyp:[2,8,56],sequenti:[8,17,92],seri:[9,61,93],serial:[14,17,23,36,41,45],serv:[39,41],server:[15,18,22,25,26,39,41,47,73,75,78,80,84],serverless:19,servic:84,set:[2,6,7,8,9,12,14,15,19,27,32,35,41,43,49,51,52,61,65,69,72,73,79,85,86,87,89,92,93,94],set_active_typ:67,set_default_parameter_nam:6,set_drop_r:67,set_float_el:16,set_input:[8,17],set_output:[17,32],set_siz:67,set_typ:67,setdefault:68,setq:65,settotalbyteslimit:51,setup:[42,67,68,84],sever:[7,8,16,22,35,36],sexstant:84,sgd:[10,14,15,19,27,36,39,40,52,56,73,74,93],shall:18,shape:[8,12,14,16,17,35,41,56],shard:[19,20,21,22,23,24,26,39,40],share:[8,18,34,36,41,92],shared_bia:9,shared_bias:8,shared_librari:18,shared_ptr:[33,34,43,69],ship:86,shold:93,shorten:8,should:[6,7,8,12,14,15,16,17,24,27,30,32,35,36,37,38,39,41,44,63,68],should_be_fals:15,should_be_tru:15,should_shuffl:[2,61,92],show:[7,10,17,19,29,35,65,92,93,94],show_check_sparse_distribution_log:[74,75],show_layer_stat:[74,75],show_parameter_stats_period:[74,75,77,79,92,93,94],shown:[8,15,39],shuf:[49,91],shuffl:[12,49],side:[8,14,36],sigint:73,sigmod:44,sigmod_op:44,sigmod_output:44,sigmoid:[8,17,44],sigmoidactiv:[9,61],sign:28,similar:[8,38,40,41,84],similarli:[8,12],simpl:[5,7,8,9,12,14,31,40,43,44,80],simple_attent:64,simple_gru:[64,89],simple_lstm:[8,50,89],simple_rnn:[8,64],simpler:36,simpli:[8,15,24,35],simplifi:[15,44],sinc:[8,19,23,25,26,38,39,45,84],singl:[7,9,12,19,39,40,41,43],sinlg:14,site:18,size:[2,7,8,9,10,12,14,19,21,22,24,38,39,44,49,51,52,56,61,64,67,68,69,86,87,89,93,94],size_a:8,size_b:8,size_t:[22,67],sizeof:17,skip:[38,51,52,66,87],sleep:80,sliceproject:8,slide:[8,10,12,19],slope:8,slopeinterceptlay:8,slot:[91,92],small:[12,16],small_messag:[74,75],small_vgg:86,smaller:[8,16,19],smart:43,smartli:8,smith:93,smooth:8,snap:79,snapshot:[20,26],snippet:30,sock:27,sock_recv_buf_s:[74,75],sock_send_buf_s:[74,75],socket:80,softmax:[8,9,15,17,32,39,40,49,64,67,89],softmax_param:51,softmax_param_attr:9,softmax_selfnorm_alpha:8,softmaxactiv:[61,89],softwar:84,solid:41,solut:[37,84],solv:[15,39,41],some:[6,8,12,14,15,17,18,22,23,24,26,27,30,31,35,37,39,40,41,43,45,84],some_c_api_funct:34,some_inst:34,some_op:35,some_other_valu:17,some_python_class:33,some_valu:17,somecppclass:33,somedata:14,somegotyp:33,someth:[8,22],sometim:[8,10,38],somewhat:24,somewher:43,soon:19,sort:[8,12,80,93],sourc:[8,12,16,18,29,34,38,94],source_dict_dim:64,source_language_word:64,space:[7,8,40],space_seperated_tokens_from_dictionary_according_to_seq:7,space_seperated_tokens_from_dictionary_according_to_sub_seq:7,spars:[6,8,10,12,22,49,67,73,75,77,80],sparse_binary_vector:[2,12,49,56],sparse_binary_vector_sequ:12,sparse_float_vector:[2,56],sparse_non_value_slot:12,sparse_remot:22,sparse_upd:[6,22,49],sparse_value_slot:12,sparse_vector:[12,49],sparse_vector_sequ:12,sparseparam:67,sparseprefetchrowcpumatrix:67,spatial:8,spec:[79,80],special:[8,24,32,39],specif:[14,18,19,29,41,43],specifi:[7,8,15,16,22,23,24,27,29,39,43,44],speech:8,speed:[9,84],sphinx:[33,54,70],split:[2,8,26,41,61,73,91,92],split_count:80,sqrt:8,squar:[8,10,11],square_error_cost:[52,56],squarerootnpool:8,squash:66,srand:75,src:[18,80,94],src_backward:64,src_dict:51,src_dict_path:51,src_embed:64,src_forward:64,src_root:4,src_word_id:64,srl:[12,92],ssd:8,ssh:[54,73],sshd:54,sstabl:15,stabil:16,stabl:[42,78],stack:41,stacked_lstm_net:93,stacked_num:93,stackexchang:8,stage:[18,25,45,66],stake:94,stale:19,standalon:58,standard:[6,41],stanford:[12,16,79],star:18,start:[8,9,14,18,19,22,23,24,26,27,36,39,46,66,75,79,80],start_paddl:80,start_pass:[74,75],start_po:8,start_pserv:[74,75],startpaddl:80,startup:[19,27],stat:[72,75,92,93,94],state:[8,9,19,35,43,50,63,79],state_act:[8,9,61],statement:31,statfulset:80,static_cast:69,staticinput:[8,63,64],statist:8,statset:72,statu:[27,66,72,79,80],status:79,std:[14,18,22,33,34,41,43,44,67,68,69,75],stderr:73,stdout:73,step:[8,9,11,16,17,19,24,31,36,39,40,41,44,61,63,64,84],step_net:17,stepnet:[17,35,41,43],stepout:61,still:[23,26,39],stochast:[10,19,23,26],stock:93,stop:8,storag:[28,78],store:[7,8,12,14,16,17,18,22,36,39,41,43],str:[14,27,77,80],strategi:[11,19,40,75,92],street:[8,92],strict:38,stride:[8,9],stride_i:8,stride_x:8,string:[2,7,8,14,17,23,29,41,43,44,45,67,68,75],strip:[51,61,92],struct:[23,24,28,34,44,51],structur:[17,23,37,41],stuff:[17,66],stun:2,style:[8,41,44],sub:[7,8,12,15,26,35,36,39],sub_nest_seq:8,sub_sequ:[2,8,56],subcommand:29,subgradi:10,subgraph:40,submiss:39,subnet:15,subobjectpath:79,subseq:[60,63],subsequ:8,subsequenceinput:[8,61],succeed:[23,79],success:[8,24,79],successfulcr:79,sudo:[65,86],suffer:16,suffic:38,suffix:27,suggest:[8,18],suit:84,sum:[8,10,17,20],sum_:8,sumpool:[8,49],support:[6,7,8,11,12,16,17,19,26,27,30,36,38,39,40,41,61,84],suppos:[18,30],suppress:[8,29],sure:[41,93],surpass:8,svs:44,swagger:28,swap_channel:87,swig:[25,33,34],swig_paddl:4,switchop:17,symbol:[8,17,34],symbols_ready_:17,symbolt:[17,41],symlink:66,sync:19,syncflag:67,synchron:[10,19,23],syntax:38,sys:87,sysroot:58,system:[17,18,19,24,26,28,30,39,40,49,93],t2b:85,tab:89,tabl:[7,8,17,45],tableproject:8,tag:[7,12,46,54],tagtyp:7,tainer_id:80,take:[2,7,8,9,15,17,18,19,26,31,32,37,41,45],talk:24,tanh:[8,9,39,64,67],tanhactiv:[9,61],tar:14,target:[8,12,14,17,18,39,41,45,94],target_dict_dim:64,target_dictionary_dim:8,target_language_embed:8,target_language_word:64,target_link_librari:18,targetinlink:[8,61],task:[7,8,39,44,92],task_queu:23,taskentri:23,taskqueu:23,tbd:[25,61],tconf:93,tcp:75,tear:72,technic:19,tee:[79,86,92,93,94],tell:[19,23,24,44],tellig:93,templat:[30,44,68,69,79,80,84],tempor:8,temporari:27,tensor:[16,17,18,35,40,45,68],tensor_desc:45,tensor_s:16,tensor_test:18,tensor_to_check:16,tensorflow:[17,39,40],term:[8,9,19],termin:79,tesh:92,test100:12,test10:12,test1:21,test:[1,2,8,12,14,15,16,18,34,38,42,66,67,68,69,72,73,75,77,86,87,89,91,92,93,94],test_:68,test_all_data_in_one_period:[79,86,91,92,93],test_compar:46,test_comparespars:46,test_comparetwonet:46,test_comparetwoopt:46,test_config_pars:46,test_cpu_gpu_compar:68,test_data:[4,94],test_fcgrad:67,test_gpuprofil:72,test_ignore_i:68,test_ignore_x:68,test_layergrad:67,test_list:[2,51,52,86],test_mul_op:68,test_networkcompar:46,test_norm:68,test_part_000:93,test_pass:[74,75,77,94],test_period:[74,75,77],test_predict:46,test_pydataprovid:46,test_pydataprovider2:46,test_pydataproviderwrapp:46,test_ratio:91,test_recurrent_machine_gener:46,test_recurrentgradientmachin:[46,61],test_swig_api:46,test_train:46,test_traineronepass:46,test_wait:[74,75],testa:15,testb:15,testbilinearfwdbwd:72,testcas:68,testconfig:67,tester:[91,94],testfcgrad:67,testfclay:67,testlayergrad:67,testmulgradop:68,testmulop:68,testq:15,testresult:14,testutil:67,text1:29,text:[2,7,9,12,15,35,85,93,94],text_conv:89,text_conv_pool:91,text_fil:[12,93],tflop:72,tftp:84,tgz:12,than:[6,7,8,9,19,27,30,41,51,84],thei:[8,15,18,19,24,26,29,31,40,41,44,72],them:[7,8,9,15,16,18,19,22,27,30,38,40,41,43,44,45,72],themselv:18,therein:[8,17],thi:[2,6,7,8,9,10,12,14,15,16,17,18,19,22,23,24,25,26,27,30,31,35,36,37,38,39,40,41,44,45,72,84,93],thing:2,think:[15,18],third:[8,19],third_parti:[8,58,59],those:[17,18,19,30,31,92],though:84,thought:18,thread:72,thread_local_rand_use_global_se:[74,75],threadid:77,threadloc:72,three:[7,8,16,19,31,36,38,87],threshold:[6,7,8,19,23,75],through:[8,18,19,23,25,32],throughput:72,thrust:41,thu:[8,26],tier:79,time:[8,9,11,12,15,16,18,19,23,26,30,35,38,39,40,41,44,45,61,72,75,79,80,84,93],timelin:[8,41],timeout:[19,23],timeout_sec:12,timer:72,timestamp:[8,20],timestep:[8,43],titl:[12,80,91],tls:28,tmp:2,to_no_sequ:[8,60],to_proto:45,to_sequ:[8,60,61],to_your_paddle_clone_path:70,todo:[7,10,12,17,19,23,26,44],toend:8,togeth:[8,9,12,14],token:[7,8,15,64,93],toler:[14,16],too:[12,16,39],took:84,tool:[58,59,70,80],toolchain:[58,59],top:[7,14,35,87],top_k:7,top_level_rnn:35,toplevel:65,topolog:[15,19,36,39],topolopi:14,torch:17,toronto:[12,86],tostr:51,total:[14,19,38,40,72,79,84,94],total_pass:38,touch:93,tourist:94,trace:[17,37,45],track:[19,23,66],tractabl:8,tradit:[8,17],traffic:39,trail:12,train100:12,train10:12,train:[1,2,6,7,8,10,12,17,21,23,24,26,31,36,37,41,45,47,49,52,56,64,73,75,77,79,80,81,82,85,86,87,89,91,92,93,94],train_arg:80,train_args_dict:80,train_args_list:80,train_conf:[85,94],train_config_dir:80,train_data:94,train_i:56,train_list:[2,51,52,86,87],train_part_000:93,train_read:56,train_x:56,trainabl:8,trainer:[2,4,15,20,21,22,23,25,36,39,40,41,52,56,67,75,77,80,92,93,94],trainer_config:[1,2,4,52,73,79,80,89,91,93],trainer_config_help:[2,52,67,86],trainer_count:[49,74,75,77,79,80,91,92,93,94],trainer_cpu:27,trainer_cr:27,trainer_gpu:27,trainer_id:[75,80],trainer_intern:22,trainer_mem:27,trainer_packag:27,trainerconfighelp:51,trainerid:[26,80],trainerintern:[89,91,94],trainonebatch:22,tran:67,tranform:41,transact:[19,23,93],transform:[8,41],transform_param_attr:9,translat:[8,9,49,85,94],transpos:8,transposedfullmatrixproject:8,travel:2,travers:31,travi:66,treat:[8,24],treatment:24,tree:[8,17,80],trg:94,trg_embed:64,tricki:33,trigger:[26,36],trim:8,truck:86,true_block:[17,32],true_imag:38,true_label:38,true_read:38,truth:[7,8,94],tune:[6,8,74],tupl:[8,9,12,14,38],ture:8,turn:[8,38,63],tutori:[79,80,81,82,85,94],tweet:93,twice:40,twine:42,twitter:93,two:[7,8,9,15,17,24,25,26,27,29,31,38,39,43,44,45,68,69,72],txt:[2,18,27,29,67,73,78,89,91,93],type:[2,7,8,9,11,12,15,17,19,22,23,26,27,28,29,33,34,35,37,38,41,44,45,52,56,61,64,67,68,69,77,79,87,89,91,92],type_nam:44,typedef:[24,33,34],typeid:44,typenam:[30,44,68,69],typic:[7,39],ubuntu:42,ubyt:38,uci:12,ufldl:8,uid:79,uint32:28,uint64:33,uint64_t:33,unawar:24,unclear:26,unconstrain:93,under:[18,23,40],understand:84,understand_senti:64,undeterminist:72,unidirect:8,uniform:[6,8,12,21,38],uninstal:46,uniqu:[15,17,19,26,27,43],unique_ptr:[43,67],unit:[8,9,18],unittest:[34,46,68],univ:94,unix:73,unk:[45,85,94],unk_idx:92,unknown:8,unlik:8,unordered_map:43,unrol:35,unseg:8,unsign:24,unsup:93,unsupbow:93,unsupport:68,until:[19,24,40,43,80],untrack:66,unzip:58,updat:[6,8,10,17,19,23,24,28,35,36,39,43,66,73,77],update_equ:[14,56],update_hook:6,updatecallback:67,upgrad:46,upload:[19,27,28,42],upon:19,upper:8,upstream:[46,66],url:[12,93],urls_neg:93,urls_po:93,urls_unsup:93,usag:[7,8,9,14,36,80],use:[6,7,8,9,11,12,14,15,16,17,18,19,25,35,36,39,40,43,44,45,66,68,72,80,87,92,93,94],use_etcd:14,use_global_stat:8,use_gpu:[4,49,56,74,75,77,79,80,86,87,89,91,92,93,94],use_jpeg:86,use_old_updat:[22,74,75],use_seq:[52,91],use_sparse_remote_updat:22,used:[2,7,8,9,10,11,12,14,15,16,17,18,19,25,26,35,36,38,39,41,43,44,45,72,93],useful:[8,9,16,43],usegpu:67,user:[6,8,9,12,14,15,16,17,18,21,23,26,27,29,30,31,38,39,40,41,43,44,78,84,91],user_id:[80,91],user_info:12,user_nam:21,usercert:21,userinfo:12,userkei:21,usernam:[21,66],uses:[19,26,36,39],using:[6,8,9,12,14,15,17,18,19,23,24,26,27,29,30,35,38,39,43,44,68,92],usr:[46,53,54,73,75,80],usrdict:85,usrmodel:85,usual:[8,14,27,72],utc:90,util:[72,80,84,86,91],uuid:[20,26],v28:8,v7a:58,valid:[8,38,41,43],valu:[2,4,6,7,8,10,11,12,14,16,17,19,32,35,36,37,39,41,43,44,45,49,67,77,80,87,89,92],value1:75,value2:75,value_evalu:7,value_rang:12,valueerror:49,vanilla:64,var1:45,var2:45,varabl:40,varaibl:17,vardesc:[17,31,37,41],vardescbuild:17,variabl:[8,12,15,16,17,31,32,35,39,40,44,79],variablenamemap:68,varianc:8,variant:8,variou:17,varproto:44,vars_:[17,43],vec:51,vector:[8,9,12,15,17,22,24,32,35,41,89,91],vendor:18,verb:12,verbos:29,veri:[8,11,18,23,30,36,40,43,86],verifi:17,version:[8,9,18,27,29,42,54,72,74,75],versu:15,vertic:8,vgg:[9,86],vgg_16_cifar:86,via:[19,35,38,66,84],view:8,virtualenv:65,visibl:[26,43],vision:86,visipedia:86,visit:14,visual:8,vlog:22,vocab:93,volum:[54,78,79,80],volumemount:[79,80],vutbr:12,wai:[7,8,9,15,24,26,39,94],wait:[10,19,24,80],wall:92,wangkuiyi:18,want:[2,8,15,27,38,43],warn:[29,46,51,80],warp:8,warpctc:8,watch:19,wbia:87,wei:[92,93],weight:[7,8,9,10,67,87],weight_act:9,weightlist:67,weights_:67,weights_t:67,welcom:18,well:[27,30,39,41],were:[7,18],wether:8,wget:58,what:[6,8,10,17,18,40,44,66,84],when:[2,6,7,8,12,14,16,17,18,19,22,23,24,27,29,36,37,39,40,41,45,72,84],where:[8,9,10,15,17,19,26,31,32,35,37,41],wherea:[17,23,30],whether:[7,8,14,16,17,38,93],which:[6,7,8,9,10,12,14,15,16,17,18,19,21,23,24,26,27,30,35,36,37,38,39,40,41,43,44,45,84,91],whileop:[17,37],who:[30,31],whoever:24,whole:[2,7,12,33,34,35,44,84],whose:[8,12,16,19,26,32,35,41,44],why:[9,16,34,72],wide:18,width:[7,8,12,22,33,38,51,67,68,94],wiki:[8,18],wikipedia:[8,12],wilder:2,window:[8,11,12,65,78],wise:[8,40,41],with_avx:[53,54,58,66],with_bia:44,with_c_api:[58,59],with_doc:53,with_doubl:[53,67],with_dso:53,with_gpu:[53,54,58,65,66],with_profil:72,with_python:[53,58],with_rdma:[53,58],with_style_check:[53,66],with_swig_pi:[53,58],with_test:[53,54,66,68],with_tim:[53,72],within:[8,23],without:[7,8,19,24,38,40,41],wmt14:[64,94],wmt14_data:94,wmt14_model:94,wmt_shrinked_data:12,woboq:54,won:61,wonder:2,word2vec:[27,49],word:[2,7,8,12,31,35,40,41,44,49,61,63,89,92],word_dict:[61,89,92],word_dim:[51,61,89],word_id:[2,49],word_idx:12,word_slot:92,word_vector_dim:[8,64,85],words_freq_sort:12,work:[12,15,17,18,19,38,54,61,66,79,80,84],worker:[40,45],workflow:41,workspac:73,would:[14,17,18,19,26,30,31,36,38,40,84,92],wouldn:31,wrap:[30,84],wrapper:[9,18,30,39,72],write:[12,15,19,26,30,38,39,40,41,51,92],write_lock:20,writelin:52,writer:15,written:[17,39,41],wrong:38,wrote:40,wsj:92,wuyi:78,www:[8,12,86,94],x64:59,x86:[58,59],x86_64:58,x_1:8,x_n:8,x_neg:16,x_po:16,xarg:[7,46,54,67],xgbe0:75,xgbe1:75,xiaojun:93,xmap_read:12,xrang:[16,38,45,52,56,67],xxbow:93,xxx:[15,87,94],xxxx:20,y_neg:16,y_po:16,y_predict:56,yaml:[18,79,80,84],yancey1989:27,yann:12,yapf:66,yeild:14,yet:84,yield:[2,12,15,21,38,49,52,56,61,92],yin:8,you:[2,6,8,9,14,16,27,43,84,87,93],your:[8,14,15,18,22,27,29,41,46,58,59,84],your_param_nam:51,your_repo:80,your_source_root:34,yourself:8,yuyang18:[10,12],zachari:93,zeng:93,zero:[6,8,9,10,12,16,19,36,37,75],zhou:[92,93],zip:[12,58,80,90],zoo:85},titles:["API","DataProvider\u7684\u4ecb\u7ecd","PyDataProvider2\u7684\u4f7f\u7528","API\u4e2d\u6587\u624b\u518c","\u57fa\u4e8ePython\u7684\u9884\u6d4b","Activation","Parameter Attribute","Evaluators","Layers","Networks","Optimizer","Pooling","Data Reader Interface and DataSets","Model Configuration","Training and Inference","PaddlePaddle Design Doc","Auto Gradient Checker Design","Design Doc: Block and Scope","Required CMake Function","Design Doc: Distributed Training","\u6a21\u578b\u53c2\u6570\u68c0\u67e5\u70b9\uff08Checkpointing\uff09","\u8bad\u7ec3\u6570\u636e\u7684\u5b58\u50a8\u548c\u5206\u53d1","Alalysis of large model distributed training in Paddle","Design Doc: Master Server","Design Doc: The Client Library of Parameter Server","Design Doc: Remote Parameter Updater for Cluster Train","Design Doc: Save Model","Submit a Distributed Training Job","FileManager\u8bbe\u8ba1\u6587\u6863","PFSClient","Design Doc: Functions, Operators, and Layers","Design Doc: Computations as a Graph","&lt;no title&gt;","Paddle\u591a\u8bed\u8a00\u63a5\u53e3\u5b9e\u73b0","C-API \u6a21\u578b\u63a8\u65ad\u5b9e\u73b0\u6587\u6863","RNNOp design","Design Doc: The C++ Class <code class=\"docutils literal\"><span class=\"pre\">Parameters</span></code>","Design Doc: ProgramDesc","Python Data Reader Design Doc","Design Doc: Distributed Training Architecture","Design Doc: Operation Graph Based Parameter Server","Design Doc: Refactorization Overview","PaddlePaddle\u53d1\u884c\u89c4\u8303","Design of Scope in Paddle","Interaction between C++ and Python","Background","\u7f16\u8bd1\u5b89\u88c5\u4e0e\u5355\u5143\u6d4b\u8bd5","\u96c6\u7fa4\u8bad\u7ec3\u4e0e\u9884\u6d4b","FAQ","\u672c\u5730\u8bad\u7ec3\u4e0e\u9884\u6d4b","\u6a21\u578b\u914d\u7f6e","\u53c2\u6570\u8bbe\u7f6e","\u7ecf\u5178\u7684\u7ebf\u6027\u56de\u5f52\u4efb\u52a1","PaddlePaddle\u7684\u7f16\u8bd1\u9009\u9879","PaddlePaddle\u7684Docker\u5bb9\u5668\u4f7f\u7528\u65b9\u5f0f","\u5b89\u88c5\u4e0e\u7f16\u8bd1","\u57fa\u672c\u4f7f\u7528\u6982\u5ff5","\u65b0\u624b\u5165\u95e8","\u6784\u5efaAndroid\u5e73\u53f0\u4e0a\u7684PaddlePaddle\u5e93","\u6784\u5efaRaspberry Pi\u5e73\u53f0\u4e0a\u7684PaddlePaddle\u5e93","\u652f\u6301\u53cc\u5c42\u5e8f\u5217\u4f5c\u4e3a\u8f93\u5165\u7684Layer","\u5355\u53cc\u5c42RNN API\u5bf9\u6bd4\u4ecb\u7ecd","RNN\u76f8\u5173\u6a21\u578b","Recurrent Group\u6559\u7a0b","RNN\u914d\u7f6e","\u7f16\u8bd1PaddlePaddle\u548c\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5","\u5982\u4f55\u8d21\u732e\u4ee3\u7801","\u5b9e\u73b0\u65b0\u7684\u7f51\u7edc\u5c42","\u5982\u4f55\u5199\u65b0\u7684Operator","\u5728Paddle\u4e2d\u5982\u4f55\u4f7f\u7528Eigen","\u5982\u4f55\u8d21\u732e/\u4fee\u6539\u6587\u6863","\u8fdb\u9636\u6307\u5357","GPU\u6027\u80fd\u5206\u6790\u4e0e\u8c03\u4f18","\u8fd0\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3","\u53c2\u6570\u6982\u8ff0","\u7ec6\u8282\u63cf\u8ff0","\u8bbe\u7f6e\u547d\u4ee4\u884c\u53c2\u6570","\u4f7f\u7528\u6848\u4f8b","Kubernetes \u7b80\u4ecb","Kubernetes\u5355\u673a\u8bad\u7ec3","Kubernetes\u5206\u5e03\u5f0f\u8bad\u7ec3","&lt;no title&gt;","&lt;no title&gt;","PaddlePaddle \u6587\u6863","Cluster bootstrapping tool survey","\u4e2d\u6587\u8bcd\u5411\u91cf\u6a21\u578b\u7684\u4f7f\u7528","\u56fe\u50cf\u5206\u7c7b\u6559\u7a0b","Model Zoo - ImageNet","\u5b8c\u6574\u6559\u7a0b","\u5feb\u901f\u5165\u95e8\u6559\u7a0b","MovieLens\u6570\u636e\u96c6","MovieLens\u6570\u636e\u96c6\u8bc4\u5206\u56de\u5f52\u6a21\u578b","\u8bed\u4e49\u89d2\u8272\u6807\u6ce8\u6559\u7a0b","\u60c5\u611f\u5206\u6790\u6559\u7a0b","\u6587\u672c\u751f\u6210\u6559\u7a0b"],titleterms:{"\u4e00\u4e9b\u7ec6\u8282\u7684\u8865\u5145":80,"\u4e0a\u4f20\u8bad\u7ec3\u6587\u4ef6":21,"\u4e0b\u8f7d\u4e0e\u89e3\u538b\u7f29":94,"\u4e0b\u8f7d\u548c\u6570\u636e\u62bd\u53d6":85,"\u4e0b\u8f7d\u5e76\u89e3\u538b\u6570\u636e\u96c6":91,"\u4e0b\u8f7d\u6570\u636e":79,"\u4e0d\u4f7f\u7528":33,"\u4e0d\u4f7f\u7528swig\u8fd9\u79cd\u4ee3\u7801\u751f\u6210\u5668":33,"\u4e0d\u540c\u7684":50,"\u4e0d\u5bfc\u51fapaddle\u5185\u90e8\u7684\u7ed3\u6784\u4f53":33,"\u4e0d\u5f15\u7528\u5176\u4ed6\u52a8\u6001\u5e93":33,"\u4e24\u79cd\u4f7f\u7528":50,"\u4e2d\u6587\u5b57\u5178":85,"\u4e2d\u6587\u77ed\u8bed\u6539\u5199\u7684\u4f8b\u5b50":85,"\u4e2d\u6587\u8bcd\u5411\u91cf\u6a21\u578b\u7684\u4f7f\u7528":85,"\u4e2d\u6587\u8bcd\u5411\u91cf\u7684\u9884\u8bad\u7ec3\u6a21\u578b":85,"\u4e3a\u4ec0\u4e48\u8981":65,"\u4e3a\u4ec0\u4e48\u9700\u8981\u6027\u80fd\u5206\u6790":72,"\u4ec0\u4e48\u662f\u6027\u80fd\u5206\u6790":72,"\u4ec5\u4ec5\u4f7f\u7528void":33,"\u4ecb\u7ecd":[85,87],"\u4ece\u5feb\u7167\u6062\u590d":20,"\u4ee3\u7801\u8981\u6c42":66,"\u4efb\u52a1\u7b80\u4ecb":52,"\u4f18\u5316\u7b97\u6cd5":89,"\u4f7f\u7528":[66,79],"\u4f7f\u7528\u52a8\u6001\u5e93\u6765\u5206\u53d1paddl":33,"\u4f7f\u7528\u6848\u4f8b":77,"\u4f7f\u7528\u6982\u8ff0":89,"\u4f7f\u7528\u6a21\u578b\u521d\u59cb\u5316\u7f51\u7edc":77,"\u4f7f\u7528\u73af\u5883\u53d8\u91cf":80,"\u4f7f\u7528\u7528\u6237\u6307\u5b9a\u7684\u8bcd\u5411\u91cf\u5b57\u5178":85,"\u4f7f\u7528\u8bf4\u660e":71,"\u4f7f\u7528\u8f6c\u6362\u5e93":21,"\u4f7f\u7528docker\u6784\u5efa":70,"\u4f7f\u7528paddlepaddle\u751f\u6210\u6a21\u578b":94,"\u4f7f\u7528paddlepaddle\u8bad\u7ec3\u6a21\u578b":94,"\u4fdd\u6301\u672c\u5730\u4ed3\u5e93\u6700\u65b0":66,"\u4fee\u6539\u542f\u52a8\u811a\u672c":79,"\u4fee\u6539\u6587\u6863":70,"\u514b\u9686":66,"\u5177\u4f53\u67d0\u79cd\u7c7b\u578b\u7684\u5934\u6587\u4ef6":34,"\u5177\u4f53\u67d0\u79cd\u7c7b\u578b\u7684\u5b9e\u73b0\u6587\u4ef6":34,"\u5185\u5b58\u4e0d\u591f\u7528\u7684\u60c5\u51b5":2,"\u5185\u7f6e\u5b9a\u65f6\u5668":72,"\u5199\u68af\u5ea6\u68c0\u67e5\u5355\u5143\u6d4b\u8bd5":67,"\u51c6\u5907\u4ea4\u53c9\u7f16\u8bd1\u73af\u5883":[58,59],"\u51c6\u5907\u5de5\u4f5c\u7a7a\u95f4":73,"\u51c6\u5907\u6570\u636e":[52,91],"\u51c6\u5907\u8bad\u7ec3\u6570\u636e":80,"\u51c6\u5907\u96c6\u7fa4\u4f5c\u4e1a\u914d\u7f6e":73,"\u51cf\u5c11\u6570\u636e\u8f7d\u5165\u7684\u8017\u65f6":49,"\u51cf\u5c11dataprovider\u7f13\u51b2\u6c60\u5185\u5b58":49,"\u51fa\u73b0":50,"\u5206\u5272\u8bad\u7ec3":91,"\u5206\u5757\u6587\u4ef6\u4f20\u8f93":28,"\u5206\u652f\u89c4\u8303":42,"\u521b\u5efa\u672c\u5730\u5206\u652f":66,"\u521b\u5efajob":80,"\u521b\u5efapaddl":79,"\u5220\u9664\u672c\u5730\u5206\u652f":66,"\u5220\u9664\u8fdc\u7a0b\u5206\u652f":66,"\u5229\u7528\u66f4\u591a\u7684\u8ba1\u7b97\u8d44\u6e90":49,"\u5230\u8fdc\u7a0b\u4ed3\u5e93":66,"\u5236\u4f5c\u955c\u50cf":80,"\u5236\u4f5cdocker\u955c\u50cf":79,"\u524d\u5411operator\u5355\u5143\u6d4b\u8bd5":68,"\u524d\u63d0\u6761\u4ef6":73,"\u52a0\u8f7dpaddlepaddl":56,"\u52a0\u901f\u6267\u884c":20,"\u52a0\u901f\u8bad\u7ec3\u901f\u5ea6":49,"\u52a8\u6001\u5e93\u4e2d\u4e0d\u5d4c\u5165\u4efb\u4f55\u5176\u4ed6\u8bed\u8a00\u7684\u89e3\u91ca\u5668":33,"\u52a8\u6001\u6269\u5bb9":20,"\u5355\u5143\u6d4b\u8bd5":75,"\u5355\u53cc\u5c42rnn":61,"\u5377\u79ef\u6a21\u578b":89,"\u5377\u79ef\u795e\u7ecf\u7f51\u7edc":86,"\u539f\u56e0":33,"\u539f\u56e0\u5217\u8868":33,"\u53c2\u6570\u4fe1\u606f":87,"\u53c2\u6570\u5185\u5b58":49,"\u53c2\u6570\u670d\u52a1\u5668\u548c\u5206\u5e03\u5f0f\u901a\u4fe1":75,"\u53c2\u6570\u6982\u8ff0":74,"\u53c2\u6570\u8bbe\u7f6e":51,"\u53c2\u6570\u8bfb\u53d6":87,"\u53c2\u8003":2,"\u53c2\u8003\u6587\u6863":[28,93],"\u53c2\u8003\u8d44\u6599":72,"\u53cc\u5411lstm":93,"\u53cc\u5c42rnn":61,"\u53cc\u5c42rnn\u4ecb\u7ecd":63,"\u53cc\u5c42rnn\u7684\u4f7f\u7528":63,"\u53cd\u5411operator\u5355\u5143\u6d4b\u8bd5":68,"\u53ef\u80fd\u7684\u5185\u5b58\u6cc4\u9732\u95ee\u9898":2,"\u53ef\u80fd\u78b0\u5230\u7684\u95ee\u9898":65,"\u53ef\u9009\u529f\u80fd":85,"\u540d\u8bcd\u89e3\u91ca":28,"\u5411\u7cfb\u7edf\u4f20\u9001\u6570\u636e":89,"\u5411\u91cf":75,"\u542f\u52a8\u4efb\u52a1":80,"\u542f\u52a8\u96c6\u7fa4\u4f5c\u4e1a":73,"\u5440":65,"\u547d\u4ee4\u884c\u53c2\u6570":89,"\u548c":60,"\u56fe\u50cf\u5206\u7c7b\u6559\u7a0b":86,"\u5728\u4e0d\u540c\u8bbe\u5907\u4e0a\u6307\u5b9a\u5c42":77,"\u5728paddle\u4e2d\u5982\u4f55\u4f7f\u7528eigen":69,"\u5728paddlepaddle\u5e73\u53f0\u8bad\u7ec3\u6a21\u578b":85,"\u57fa\u4e8epython\u7684\u9884\u6d4b":4,"\u57fa\u672c\u4f7f\u7528\u6982\u5ff5":56,"\u57fa\u672c\u539f\u7406":63,"\u57fa\u672c\u8981\u6c42":33,"\u5982\u4f55\u4e66\u5199\u6587\u6863":70,"\u5982\u4f55\u4f7f\u7528":50,"\u5982\u4f55\u5171\u4eab\u53c2\u6570":51,"\u5982\u4f55\u5199\u65b0\u7684oper":68,"\u5982\u4f55\u51cf\u5c11\u5185\u5b58\u5360\u7528":49,"\u5982\u4f55\u521d\u59cb\u5316\u53c2\u6570":51,"\u5982\u4f55\u52a0\u8f7d\u9884\u8bad\u7ec3\u53c2\u6570":51,"\u5982\u4f55\u52a0\u901f\u8bad\u7ec3\u901f\u5ea6":49,"\u5982\u4f55\u548c\u660e\u6587\u8fdb\u884c\u76f8\u4e92\u8f6c\u5316":51,"\u5982\u4f55\u6307\u5b9agpu\u8bbe\u5907":49,"\u5982\u4f55\u66f4\u65b0\u6587\u6863\u4e3b\u9898":70,"\u5982\u4f55\u66f4\u65b0doc":70,"\u5982\u4f55\u6784\u5efa\u6587\u6863":70,"\u5982\u4f55\u8bbe\u7f6e\u5b66\u4e60\u7387\u9000\u706b":51,"\u5982\u4f55\u8c03\u7528":49,"\u5982\u4f55\u8d21\u732e":70,"\u5982\u4f55\u8d21\u732e\u4ee3\u7801":66,"\u5982\u4f55\u8fdb\u884c\u6027\u80fd\u5206\u6790":72,"\u5982\u4f55\u9009\u62e9sgd\u7b97\u6cd5\u7684\u5b66\u4e60\u7387":51,"\u5b50\u5e8f\u5217\u95f4\u65e0memori":61,"\u5b50\u5e8f\u5217\u95f4\u6709memori":61,"\u5b57\u6bb5\u914d\u7f6e\u6587\u4ef6":91,"\u5b58\u50a8\u7684\u53c2\u6570\u683c\u5f0f\u662f\u4ec0\u4e48":51,"\u5b89\u88c5":89,"\u5b89\u88c5\u4e0e\u7f16\u8bd1":55,"\u5b89\u88c5\u6d41\u7a0b":55,"\u5b89\u88c5kubectl":78,"\u5b8c\u6574\u6559\u7a0b":88,"\u5b9a\u4e49operator\u7c7b":68,"\u5b9a\u4e49opkernel\u7c7b":68,"\u5b9a\u4e49protomaker\u7c7b":68,"\u5b9e\u73b0":33,"\u5b9e\u73b0\u5355\u5143\u6d4b\u8bd5":68,"\u5b9e\u73b0\u65b0\u7684\u7f51\u7edc\u5c42":67,"\u5b9e\u73b0\u65b9\u5f0f":34,"\u5b9e\u73b0\u8ba1\u7b97":69,"\u5b9e\u73b0c":[67,68],"\u5b9e\u73b0python\u5c01\u88c5":67,"\u5bfc\u51fac":33,"\u5c06\u547d\u4ee4\u53c2\u6570\u4f20\u7ed9\u7f51\u7edc\u914d\u7f6e":77,"\u5c0f\u7ed3":2,"\u5de5\u5177":72,"\u5e38\u7528\u6a21\u578b":88,"\u5e76\u5b8c\u6210":66,"\u5efa\u7acb":66,"\u5f00\u53d1\u6807\u51c6":71,"\u5f00\u59cb\u5f00\u53d1":66,"\u5f02\u6b65\u968f\u673a\u68af\u5ea6\u4e0b\u964d":75,"\u5f15\u7528":92,"\u5feb\u7167\u4fdd\u5b58\u7684\u8bbe\u8ba1\u5982\u4e0b":20,"\u5feb\u901f\u5165\u95e8\u6559\u7a0b":89,"\u6027\u80fd\u4f18\u5316":71,"\u6027\u80fd\u5206\u6790\u5c0f\u6280\u5de7":72,"\u6027\u80fd\u5206\u6790\u5de5\u5177\u4ecb\u7ecd":72,"\u6027\u80fd\u8c03\u4f18":75,"\u603b\u4f53\u6548\u679c\u603b\u7ed3":89,"\u603b\u4f53\u6d41\u7a0b":65,"\u60c5\u611f\u5206\u6790\u6559\u7a0b":93,"\u6216\u8005\u662f":46,"\u627e\u5230\u7684pythonlibs\u548cpythoninterp\u7248\u672c\u4e0d\u4e00\u81f4":46,"\u62a5importerror":46,"\u6307\u9488\u4f5c\u4e3a\u7c7b\u578b\u7684\u53e5\u67c4":33,"\u63a5\u53e3":87,"\u63a5\u53e3\u8f93\u51fa\u591a\u4e2alayer\u7684\u9884\u6d4b\u7ed3\u679c":49,"\u63a8\u5bfc\u65b9\u7a0b":67,"\u63a8\u6d4b\u6267\u884c":20,"\u63d0\u4ea4":66,"\u63d0\u4ea4\u4ee3\u7801\u7684\u4e00\u4e9b\u7ea6\u5b9a":66,"\u63d0\u4ea4\u955c\u50cf":79,"\u63d0\u53d6\u7535\u5f71\u6216\u7528\u6237\u7684\u7279\u5f81\u5e76\u751f\u6210python\u5bf9\u8c61":91,"\u642d\u5efa\u795e\u7ecf\u7f51\u7edc":56,"\u652f\u6301\u53cc\u5c42\u5e8f\u5217\u4f5c\u4e3a\u8f93\u5165\u7684layer":60,"\u652f\u6301\u7528\u6237\u81ea\u5b9a\u4e49\u7684\u6570\u636e\u9884\u5904\u7406job":21,"\u6570\u636e\u51c6\u5907":[86,91,94],"\u6570\u636e\u63cf\u8ff0":92,"\u6570\u636e\u63d0\u4f9b":92,"\u6570\u636e\u63d0\u4f9b\u811a\u672c":91,"\u6570\u636e\u652f\u6301":75,"\u6570\u636e\u683c\u5f0f\u51c6\u5907":89,"\u6570\u636e\u7684\u51c6\u5907\u548c\u9884\u5904\u7406":85,"\u6570\u636e\u96c6\u7279\u5f81":90,"\u6570\u636e\u9884\u5904\u7406":94,"\u6570\u6910\u4ecb\u7ecd":93,"\u6570\u6910\u51c6\u5907":93,"\u6574\u4f53\u65b9\u6848":80,"\u6587\u4ef6\u4f20\u8f93\u4f18\u5316":28,"\u6587\u4ef6\u8bbf\u95ee\u65b9\u5f0f":21,"\u6587\u4ef6\u8bbf\u95ee\u7684\u6743\u9650":21,"\u6587\u4ef6\u9884\u5904\u7406":21,"\u6587\u672c\u751f\u6210":94,"\u6587\u672c\u751f\u6210\u6559\u7a0b":94,"\u6587\u6863":[54,83],"\u65b0\u624b\u5165\u95e8":57,"\u65e5\u5fd7\u4e2d\u4fdd\u5b58\u5747\u4e3a\u7f51\u7edc\u901a\u4fe1\u7c7b\u9519\u8bef":47,"\u65f6\u5e8f\u6a21\u578b":89,"\u65f6\u5e8f\u6a21\u578b\u7684\u4f7f\u7528\u573a\u666f":2,"\u65f6\u95f4\u5e8f\u5217":61,"\u65f6\u95f4\u6b65":61,"\u66b4\u9732\u63a5\u53e3\u539f\u5219":34,"\u672c\u5730\u6d4b\u8bd5":77,"\u672c\u5730\u8bad\u7ec3":77,"\u672c\u5730\u8bad\u7ec3\u4e0e\u9884\u6d4b":49,"\u672f\u8bed":20,"\u6784\u5efa\u548c\u6d4b\u8bd5":66,"\u6784\u5efaandroid\u5e73\u53f0\u4e0a\u7684paddlepaddle\u5e93":58,"\u6784\u5efaraspberri":59,"\u67b6\u6784\u56fe":28,"\u67e5\u770b\u8bad\u7ec3\u7ed3\u679c":79,"\u67e5\u770b\u8f93\u51fa":80,"\u6837\u4f8b\u6570\u636e":2,"\u6846\u67b6\u751f\u6210":28,"\u6848\u4f8b\u4e00":77,"\u6848\u4f8b\u4e8c":77,"\u68c0\u67e5\u6a21\u578b\u8f93\u51fa":73,"\u68c0\u67e5\u96c6\u7fa4\u8bad\u7ec3\u7ed3\u679c":73,"\u6982\u5ff5\u7b80\u4ecb":68,"\u6982\u5ff5\u89e3\u91ca":21,"\u6982\u8ff0":[60,63],"\u6a21\u5757":28,"\u6a21\u578b":87,"\u6a21\u578b\u4e0b\u8f7d":87,"\u6a21\u578b\u53c2\u6570\u68c0\u67e5\u70b9":20,"\u6a21\u578b\u63a8\u65ad\u5b9e\u73b0\u6587\u6863":34,"\u6a21\u578b\u68c0\u9a8c":52,"\u6a21\u578b\u7f51\u7edc\u7ed3\u6784":89,"\u6a21\u578b\u8bad\u7ec3":[86,94],"\u6a21\u578b\u8bc4\u4f30\u548c\u9884\u6d4b":91,"\u6a21\u578b\u914d\u7f6e":[50,61,71],"\u6a21\u578b\u914d\u7f6e\u7684\u6a21\u578b\u914d\u7f6e":61,"\u6ce8\u518coper":68,"\u6ce8\u610f\u4e8b\u9879":[2,68],"\u6d41\u7a0b\u4ecb\u7ecd":21,"\u6d4b\u8bd5":[75,92],"\u6d4b\u8bd5\u6587\u4ef6":91,"\u6d4b\u8bd5\u6a21\u578b":93,"\u7279\u5f81":92,"\u7279\u5f81\u63d0\u53d6":87,"\u751f\u6210\u5e8f\u5217":64,"\u751f\u6210\u6a21\u578b\u7684\u547d\u4ee4\u4e0e\u7ed3\u679c":94,"\u751f\u6210\u6d41\u7a0b\u7684\u4f7f\u7528\u65b9\u6cd5":63,"\u751f\u6210sparse\u6587\u4ef6":28,"\u7528\u6237\u4f7f\u7528\u6d41\u7a0b":28,"\u7528\u6237\u6587\u4ef6\u63cf\u8ff0":90,"\u7528\u6237\u81ea\u5b9a\u4e49\u6570\u636e\u96c6":94,"\u7528\u6237\u81ea\u5b9a\u4e49\u6570\u6910\u9884\u5904\u7406":93,"\u7535\u5f71\u6587\u4ef6\u63cf\u8ff0":90,"\u7684\u533a\u522b":50,"\u7684\u53c2\u6570":50,"\u7684\u65b9\u6cd5\u6709\u4f55\u533a\u522b":50,"\u76ee\u5f55\u7ed3\u6784":34,"\u76ee\u6807":28,"\u76f4\u63a5\u6784\u5efa":70,"\u76f8\u5173\u6982\u5ff5":63,"\u77e9\u9635":75,"\u793a\u4f8b1":61,"\u793a\u4f8b2":61,"\u793a\u4f8b3":61,"\u793a\u4f8b4":61,"\u793a\u4f8b\u7a0b\u5e8f":21,"\u795e\u7ecf\u5143\u6fc0\u6d3b\u5185\u5b58":49,"\u795e\u7ecf\u7f51\u7edc\u7ed3\u6784\u914d\u7f6e":91,"\u795e\u7ecf\u7f51\u7edc\u914d\u7f6e":92,"\u7a00\u758f\u8bad\u7ec3":77,"\u7b26\u53f7":33,"\u7b80\u4ecb":[78,94],"\u7b80\u5355\u95e8\u63a7\u5faa\u73af\u795e\u7ecf\u7f51\u7edc":64,"\u7c7b":[33,67,68],"\u7ebf\u6027\u56de\u5f52\u5b8c\u6574\u793a\u4f8b":56,"\u7ec3\u4e60":86,"\u7ec6\u8282\u63a2\u7a76":86,"\u7ec6\u8282\u63cf\u8ff0":75,"\u7ec8\u6b62\u96c6\u7fa4\u4f5c\u4e1a":73,"\u7ecf\u5178\u7684\u7ebf\u6027\u56de\u5f52\u4efb\u52a1":52,"\u7ed1\u5b9apython":68,"\u7f16\u5199yaml\u6587\u4ef6":79,"\u7f16\u8bd1":68,"\u7f16\u8bd1\u548c\u5b89\u88c5":[58,59],"\u7f16\u8bd1\u548c\u6267\u884c\u5355\u5143\u6d4b\u8bd5":68,"\u7f16\u8bd1\u5b89\u88c5\u4e0e\u5355\u5143\u6d4b\u8bd5":46,"\u7f16\u8bd1\u5b89\u88c5\u540e\u6267\u884c":46,"\u7f16\u8bd1\u6d41\u7a0b":55,"\u7f16\u8bd1\u9009\u9879":34,"\u7f16\u8bd1\u9009\u9879\u7684\u8bbe\u7f6e":53,"\u7f16\u8bd1paddlepaddle\u548c\u8fd0\u884c\u5355\u5143\u6d4b\u8bd5":65,"\u7f29\u5bb9":20,"\u7f51\u7edc\u53ef\u89c6\u5316":87,"\u7f51\u7edc\u914d\u7f6e\u4e2d\u7684\u8c03\u7528":2,"\u800c\u662f\u624b\u5199\u591a\u8bed\u8a00\u7ed1\u5b9a":33,"\u80cc\u666f":33,"\u81ea\u7136\u8bed\u8a00\u5904\u7406":75,"\u8986\u76d6\u4e0d\u4e00\u81f4\u7684\u90e8\u5206":28,"\u89c2\u6d4b\u8bcd\u5411\u91cf":85,"\u8bad\u7ec3":[75,91,92],"\u8bad\u7ec3\u5668\u914d\u7f6e\u6587\u4ef6":91,"\u8bad\u7ec3\u56e0\u6b64\u9000\u51fa\u600e\u4e48\u529e":49,"\u8bad\u7ec3\u6570\u636e\u5b58\u50a8":21,"\u8bad\u7ec3\u6570\u636e\u7684\u5b58\u50a8\u548c\u5206\u53d1":21,"\u8bad\u7ec3\u6a21\u578b":[52,56,89,93],"\u8bad\u7ec3\u6a21\u578b\u7684\u547d\u4ee4\u4e0e\u7ed3\u679c":94,"\u8bad\u7ec3\u6d41\u7a0b\u7684\u4f7f\u7528\u65b9\u6cd5":63,"\u8bad\u7ec3\u8fc7\u7a0b\u4e2d\u51fa\u73b0":49,"\u8bbe\u7f6e\u547d\u4ee4\u884c\u53c2\u6570":76,"\u8bc4\u5206\u6587\u4ef6\u63cf\u8ff0":90,"\u8bcd\u5411\u91cf\u6a21\u578b":89,"\u8bcd\u5411\u91cf\u6a21\u578b\u7684\u4fee\u6b63":85,"\u8bcd\u6c47\u8868":61,"\u8be6\u7ec6\u6559\u7a0b":72,"\u8bed\u4e49\u89d2\u8272\u6807\u6ce8\u6559\u7a0b":92,"\u8bfb\u53d6\u53cc\u5c42\u5e8f\u5217\u6570\u636e":61,"\u8f6c\u6362\u5e93":21,"\u8f93\u5165":63,"\u8f93\u5165\u4e0d\u7b49\u957f":61,"\u8f93\u5165\u793a\u4f8b":63,"\u8f93\u51fa":63,"\u8f93\u51fa\u65e5\u5fd7":89,"\u8fd0\u884c\u5206\u5e03\u5f0f\u8bad\u7ec3":73,"\u8fd0\u884c\u5bb9\u5668":79,"\u8fd0\u884cdocker":46,"\u8fd0\u884cpaddlepaddl":54,"\u8fd9\u4e2a\u52a8\u6001\u5e93\u4f7f\u7528c99\u6807\u51c6\u7684\u5934\u6587\u4ef6\u5bfc\u51fa\u4e00\u4e9b\u51fd\u6570":33,"\u8fdb\u884c\u8bad\u7ec3":[21,79],"\u8fdb\u9636\u6307\u5357":71,"\u9009\u62e9\u5b58\u50a8\u65b9\u6848":78,"\u901a\u7528":75,"\u901a\u8fc7docker\u5bb9\u5668\u5f00\u53d1paddlepaddl":54,"\u903b\u8f91\u56de\u5f52\u6a21\u578b":89,"\u9047\u5230":46,"\u90e8\u7f72kubernetes\u96c6\u7fa4":78,"\u914d\u7f6e\u4e2d\u7684\u6570\u636e\u52a0\u8f7d\u5b9a\u4e49":89,"\u914d\u7f6e\u4ea4\u53c9\u7f16\u8bd1\u53c2\u6570":[58,59],"\u914d\u7f6e\u5faa\u73af\u795e\u7ecf\u7f51\u7edc\u67b6\u6784":64,"\u914d\u7f6e\u7f51\u7edc":56,"\u914d\u7f6ekubectl":78,"\u914d\u7f6ekubectl\u8bbf\u95ee\u4f60\u7684kubernetes\u96c6\u7fa4":78,"\u94a9\u5b50":66,"\u9519\u8bef\u600e\u4e48\u529e":50,"\u9644\u5f55":89,"\u968f\u673a\u6570":75,"\u96c6\u7fa4\u591a\u8282\u70b9\u8bad\u7ec3":47,"\u96c6\u7fa4\u8bad\u7ec3":77,"\u96c6\u7fa4\u8bad\u7ec3\u4e0e\u9884\u6d4b":47,"\u9700\u8981\u7684\u8f6f\u786c\u4ef6":65,"\u975e\u6cd5\u6307\u4ee4":46,"\u9884\u5904\u7406":86,"\u9884\u5904\u7406\u547d\u4ee4\u548c\u7ed3\u679c":94,"\u9884\u5904\u7406\u5de5\u4f5c\u6d41\u7a0b":94,"\u9884\u6d4b":[86,87,89,92,93],"\u9884\u6d4b\u6d41\u7a0b":4,"\u9884\u6d4bdemo":4,"\u9884\u8bad\u7ec3\u7684\u6a21\u578b":94,"abstract":[39,40,84],"api\u4e2d\u6587\u624b\u518c":3,"api\u5bf9\u6bd4\u4ecb\u7ecd":61,"beam_search\u7684\u751f\u6210":61,"blas\u8def\u5f84\u76f8\u5173\u7684\u7f16\u8bd1\u9009\u9879":53,"bleu\u8bc4\u4f30":94,"book\u4e2d\u6240\u6709\u7ae0\u8282":42,"bool\u578b\u7684\u7f16\u8bd1\u9009\u9879":53,"class":[36,45],"cmake\u6e90\u7801\u7f16\u8bd1":46,"cudnn\u7684\u7f16\u8bd1\u9009\u9879":53,"dataprovider\u7684\u4ecb\u7ecd":1,"dataprovider\u7684\u4f7f\u7528":2,"docker\u4f7f\u7528\u5165\u95e8":54,"filemanager\u8bbe\u8ba1\u6587\u6863":28,"float":49,"function":[18,30],"gpu\u548ccpu\u6df7\u5408\u4f7f\u7528":77,"gpu\u6027\u80fd\u5206\u6790\u4e0e\u8c03\u4f18":72,"gpu\u955c\u50cf\u51fa\u73b0":46,"group\u6559\u7a0b":63,"import":46,"kubernetes\u5206\u5e03\u5f0f\u8bad\u7ec3":80,"kubernetes\u5355\u673a\u8bad\u7ec3":79,"meta\u6587\u4ef6":91,"meta\u914d\u7f6e\u6587\u4ef6":91,"mnist\u7684\u4f7f\u7528\u573a\u666f":2,"movielens\u6570\u636e\u96c6":90,"movielens\u6570\u636e\u96c6\u8bc4\u5206\u56de\u5f52\u6a21\u578b":91,"paddle\u52a8\u6001\u5e93\u4e2d":33,"paddle\u591a\u8bed\u8a00\u63a5\u53e3\u5b9e\u73b0":33,"paddle\u7248\u672c\u53f7\u4e3a0":46,"paddlepaddle\u53d1\u5e03\u7684docker\u955c\u50cf\u4f7f\u7528\u8bf4\u660e":54,"paddlepaddle\u53d1\u884c\u89c4\u8303":42,"paddlepaddle\u56de\u5f52\u6d4b\u8bd5\u5217\u8868":42,"paddlepaddle\u7684\u7f16\u8bd1\u9009\u9879":53,"paddlepaddle\u7684docker\u5bb9\u5668\u4f7f\u7528\u65b9\u5f0f":54,"pi\u5e73\u53f0\u4e0a\u7684paddlepaddle\u5e93":59,"pod\u95f4\u901a\u4fe1":80,"pydataprovider2\u7684\u4f7f\u7528":2,"python\u63a5\u53e3":87,"python\u76f8\u5173\u7684\u5355\u5143\u6d4b\u8bd5\u90fd\u8fc7\u4e0d\u4e86":46,"python\u811a\u672c\u8bfb\u53d6\u6570\u636e":89,"return":38,"rnn\u76f8\u5173\u6a21\u578b":62,"rnn\u914d\u7f6e":64,"tensor\u4f7f\u7528\u6837\u4f8b":69,"tensor\u5230eigentensor\u7684\u8f6c\u6362":69,"tensor\u6a21\u5757":69,Abs:5,For:18,The:[17,24,31,36,41],USE:41,Use:17,Using:[18,24],With:27,activ:[5,8],adadelta:10,adagrad:10,adam:10,adamax:10,addto:8,aggreg:8,aggregatelevel:8,alalysi:22,algorithm:[16,19,35,39],all:43,analysi:39,anneal:51,api:[0,3,34,44],appendix:84,applic:3,architectur:39,argument:[29,38],arrai:16,associ:43,assumpt:84,async:75,attent:64,attribut:6,auc:7,auto:16,avg:11,backgraound:16,background:[40,45],backward:[31,41],base:[27,40],basepool:11,basic:84,batch:38,batch_norm:8,batch_siz:38,beam_search:8,becaus:51,benefit:[40,41],between:[15,44],bidirectional_gru:9,bidirectional_lstm:9,big:51,bilinear_interp:8,binari:17,bla:53,block:[17,31,41],block_expand:8,book:54,bootstrap:84,brelu:5,bring:84,build:41,cach:2,can:43,capi:34,capi_priv:34,challeng:40,check:[8,16],checker:16,checkpoint:[19,20,26],choos:18,chunk:7,cifar:12,classif:7,classification_error:7,classification_error_print:7,client:24,clip:8,clone:66,close:16,cluster:[25,84],cmake:18,code:27,column_sum:7,commit:66,compar:84,compil:[17,41],compos:38,comput:[17,31,41],con:84,concat:8,concept:41,conclus:[26,84],config:3,configur:13,conll05:12,connect:8,construct:31,content:[2,34,46,47,49,50,51,60,72],context_project:8,control:41,conv:8,conv_oper:8,conv_project:8,conv_shift:8,convert:[26,39,40],core:16,cos_sim:8,cost:8,creat:[38,41,43],creation:23,creator:38,crf:8,crf_decod:8,cross_channel_norm:8,cross_entropy_cost:8,cross_entropy_with_selfnorm_cost:8,ctc:8,ctc_error:7,cuda:[46,53],cudnn:53,cudnnavg:11,cudnnmax:11,custom:38,dat:90,data:[8,12,19,38,39],datafeed:12,dataprovid:[3,75],dataset:[12,19,23],datatyp:12,decayedadagrad:10,decor:38,deep:17,definit:45,dens:26,deploi:27,descript:[29,41],design:[15,16,17,19,23,24,25,26,30,31,35,36,37,38,39,40,41,43],destroi:43,detail:22,detect:[7,8],detection_map:7,detection_output:8,develop:41,dictionari:38,differ:41,discuss:40,dispatch:[19,23],distribut:[15,19,22,27,39],doc:[15,17,19,23,24,25,26,30,31,36,37,38,39,40,41],docker:[27,65],dotmul_oper:8,dotmul_project:8,driver:46,drop_out:50,dropout:8,duplic:50,dylib:34,dynam:19,eigen:69,elect:26,els:17,embed:8,enough:16,entri:38,environ:27,eos:8,eval:39,evalu:7,event:[14,15],exampl:[15,18,34],except:49,execut:[17,41],exp:5,expand:[8,60],expandlevel:8,explain:16,faq:48,fault:19,file:17,first_seq:[8,60],fork:66,format:[17,19],forward:31,frame:17,framework:[16,69],from:[15,26,44],full_matrix_project:8,fulli:8,gate:64,gated_unit:8,gener:84,get_output:8,gpu:75,gradient:[16,24],gradient_print:7,graident:16,graph:[31,40,41],group:8,gru:[9,75],gru_group:9,gru_step:8,gru_unit:9,grumemori:8,handler:[15,33],happen:26,have:45,hierarchi:17,high:44,how:[16,22,38,41],hsigmoid:8,huber_classification_cost:8,huber_regression_cost:8,ident:5,identity_project:8,ifelseop:17,illeg:46,imag:[8,9,27],imagenet:87,imdb:[12,93],img_cmrnorm:8,img_conv:8,img_conv_bn_pool:9,img_conv_group:9,img_pool:8,imikolov:12,implement:[16,18,22,35,38],infer:[14,49],ingredi:15,ingress:28,init_hook:2,initi:24,input_typ:2,insid:43,instal:84,instruct:46,insuffici:46,interact:44,interfac:[12,16,19,24,25,38,43],intermedi:41,interpol:8,isn:38,issu:66,job:[19,27,79],join:8,kernel:41,kmax_sequence_scor:8,kubernet:[27,78,79],lambda_cost:8,languag:17,larg:22,last_seq:[8,60],layer:[8,15,30,50],learn:[17,51],learnabl:8,leval:44,level:44,libpaddle_capi_shar:34,libpaddle_capi_whol:34,librari:[24,41],limit:39,linear:5,linear_comb:8,list:[20,38],local:[39,43],lodtensor:35,lodtensordesc:45,log:5,logic:23,low:44,lstm:[9,75,92,93],lstm_step:8,lstmemori:8,lstmemory_group:9,lstmemory_unit:9,macro:41,manag:18,map:[38,41],master:[19,23,27],math:8,mathemat:16,max:11,maxframe_print:7,maxid:8,maxid_print:7,maxout:8,memori:[8,35,50,61,63],messag:[44,51],migrat:41,mileston:41,mini:38,minibatch:12,misc:8,mix:8,mnist:12,model:[3,13,15,22,24,26,64,87],modul:[41,46],momentum:10,movi:90,movielen:12,multi_binary_label_cross_entropy_cost:8,multibox_loss:8,multipl:38,multiplex:8,name:[43,46,50],nce:8,necessari:41,need:38,nest:35,network:[9,41,64],neural:64,nlp:[9,75],norm:8,note:16,numer:16,numpi:16,nvprof:72,nvvp:72,object:19,onli:[38,43],op_mak:41,oper:[30,40,41],opinfomap:41,opkernel:41,opproto:44,optim:[10,19,24,31],option:29,opwithkernel:41,order:29,org:70,origin:41,orthogon:43,output:8,overview:[26,41,43],packag:18,pad:8,paddl:[22,38,43,46,50,69],paddlejob:27,paddlepaddl:[15,17,39,42,46,70,83],parallel_nn:77,paramet:[6,8,14,15,19,24,25,27,36,40],parent:43,part:31,partit:24,path:[26,29],perform:75,persist:23,pfsclient:[28,29],pfsserver:28,placement:39,platform:46,pnpair:7,point:49,pool:[8,11,60],power:8,pre:66,precision_recal:7,prefetch:38,prelu:8,print:7,pro:84,procedur:84,process:[19,24,27,41],program:17,programdesc:37,project:18,protocol:51,provid:[2,38],pserver:26,pull:66,push:66,python:[16,27,35,38,39,44,45],qualiti:41,queue:[19,23],rank:7,rank_cost:8,rate:[51,90],reader:[12,15,38],realiz:41,recoveri:19,recurr:[8,9,50,63,64],recurrent_group:8,ref:16,refactor:41,refer:[2,39,40],regist:[41,44],registri:41,regular:24,reject:51,relat:41,relu:5,remot:25,repeat:8,represent:[17,41],request:66,requir:18,reshap:8,resnet:87,retri:23,rmsprop:10,rnn:[35,61,75],rnnop:[17,35,41],rotat:8,row_conv:8,row_l2_norm:8,runtim:[27,39],sampl:8,sampling_id:8,save:26,scale:[8,19],scale_shift:8,scaling_project:8,scope:[17,35,41,43],select:24,selective_fc:8,sentiment:12,separ:41,seq_concat:8,seq_reshap:8,seq_slic:8,seqtext_print:7,sequenc:64,sequence_conv_pool:9,sequencesoftmax:5,server:[19,23,24,27,40],session:39,sextant:84,sgd:75,share:[15,43],should:[43,45],shuffl:38,sigmoid:5,simple_attent:9,simple_gru2:9,simple_gru:9,simple_img_conv_pool:9,simple_lstm:9,singl:38,slice:8,slice_project:8,slope_intercept:8,small_vgg:9,smooth_l1_cost:8,softmax:5,softrelu:5,spars:[24,25,26],spp:8,squar:5,square_error_cost:8,squarerootn:11,stack:[17,93],stanh:5,start:15,step:35,store:19,sub_nested_seq:8,subcommond:29,submit:27,suffici:38,suitabl:18,sum:[7,11],sum_cost:8,sum_to_one_norm:8,summar:15,support:46,survei:84,synopsi:29,tabl:34,table_project:8,take:38,tanh:5,task:[19,23],tecton:84,tensor:[8,41,69],text_conv_pool:9,theori:16,thi:[43,46],timelin:26,todo:[20,21],togeth:43,toler:19,too:51,tool:[18,84],train:[14,15,19,22,25,27,38,39],trainer:[14,19,24,26,27],tran:8,trans_full_matrix_project:8,tune:75,two:16,uci_h:12,updat:[15,25,26],usag:[35,38],use:[22,38],user:[19,90],util:7,value_print:7,vardesc:45,variabl:[41,43,45],version:46,vgg_16_network:9,warp_ctc:8,what:[22,26,45],wheel:46,when:[26,43],whl:46,why:[38,41],wmt14:12,zoo:87}})
\ No newline at end of file
diff --git a/develop/doc_cn/survey/cluster_bootstrapping_tools.html b/develop/doc_cn/survey/cluster_bootstrapping_tools.html
index dbbf2277be24d5e9217097ad30be76541ae78902..e6a8970eda90dfeb3f67696742bb36b451b5641f 100644
--- a/develop/doc_cn/survey/cluster_bootstrapping_tools.html
+++ b/develop/doc_cn/survey/cluster_bootstrapping_tools.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/tutorials/embedding_model/index_cn.html b/develop/doc_cn/tutorials/embedding_model/index_cn.html
index b175f9c7177bc797d0a5ff6d42d989c21ca85d3c..2eebaab520335d98dfd1cde6270cf04343db652c 100644
--- a/develop/doc_cn/tutorials/embedding_model/index_cn.html
+++ b/develop/doc_cn/tutorials/embedding_model/index_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/tutorials/image_classification/index_cn.html b/develop/doc_cn/tutorials/image_classification/index_cn.html
index 0494baa0e82f417770d53aae63e2321e6401f364..fdb7628c5ea7981541d9c9e8a01828ca288fd3a5 100644
--- a/develop/doc_cn/tutorials/image_classification/index_cn.html
+++ b/develop/doc_cn/tutorials/image_classification/index_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/tutorials/imagenet_model/resnet_model_cn.html b/develop/doc_cn/tutorials/imagenet_model/resnet_model_cn.html
index 3a45613d4ff131d61317d3f84b87990051ddc964..041d8a8050e71c98f24d47e4450cbbf9e54c8d7d 100644
--- a/develop/doc_cn/tutorials/imagenet_model/resnet_model_cn.html
+++ b/develop/doc_cn/tutorials/imagenet_model/resnet_model_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/tutorials/index_cn.html b/develop/doc_cn/tutorials/index_cn.html
index bcf2bc7059b8a79236788309abfdbdf5fcaa7c6e..5cad032c70c6b029983fb254d653ddf69ddeed42 100644
--- a/develop/doc_cn/tutorials/index_cn.html
+++ b/develop/doc_cn/tutorials/index_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/tutorials/quick_start/index_cn.html b/develop/doc_cn/tutorials/quick_start/index_cn.html
index 677d0c82e5548e36e5ad7b953ed1a04f8e69272b..4ef8761a047c2cdeb54e074b29970e190f6cde86 100644
--- a/develop/doc_cn/tutorials/quick_start/index_cn.html
+++ b/develop/doc_cn/tutorials/quick_start/index_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/tutorials/rec/ml_dataset_cn.html b/develop/doc_cn/tutorials/rec/ml_dataset_cn.html
index f3ae2013a18ba4aca142fabed0d8447ec2ade5a5..a8985b4d8f3e02b2c543f886f6c42e59c78b4ffd 100644
--- a/develop/doc_cn/tutorials/rec/ml_dataset_cn.html
+++ b/develop/doc_cn/tutorials/rec/ml_dataset_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/tutorials/rec/ml_regression_cn.html b/develop/doc_cn/tutorials/rec/ml_regression_cn.html
index bb5de1619fc1ab5c2ed0ce172de99ae1decf1f6b..ae9aae91ddd442a9c5b5c65c69106c86d9dc9d7b 100644
--- a/develop/doc_cn/tutorials/rec/ml_regression_cn.html
+++ b/develop/doc_cn/tutorials/rec/ml_regression_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/tutorials/semantic_role_labeling/index_cn.html b/develop/doc_cn/tutorials/semantic_role_labeling/index_cn.html
index 0f648a14362be2dbf318fa291f04c29f0cb9aeb9..c0fb4b2a8c9ed2e1767871d86d2855f680b094fa 100644
--- a/develop/doc_cn/tutorials/semantic_role_labeling/index_cn.html
+++ b/develop/doc_cn/tutorials/semantic_role_labeling/index_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/tutorials/sentiment_analysis/index_cn.html b/develop/doc_cn/tutorials/sentiment_analysis/index_cn.html
index 8ab2025a00ca0a3def3c3cd6919eb169469fa9da..c4aa814bea1cab3dc3829547924517967fe4fb7a 100644
--- a/develop/doc_cn/tutorials/sentiment_analysis/index_cn.html
+++ b/develop/doc_cn/tutorials/sentiment_analysis/index_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>
 
         
diff --git a/develop/doc_cn/tutorials/text_generation/index_cn.html b/develop/doc_cn/tutorials/text_generation/index_cn.html
index 6c8105a99a98ae8d52c7453b35ddb4cc0012de6d..11a85f42618e33132d3e9a450ce9f91590bff288 100644
--- a/develop/doc_cn/tutorials/text_generation/index_cn.html
+++ b/develop/doc_cn/tutorials/text_generation/index_cn.html
@@ -154,7 +154,14 @@
 <li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
 </ul>
 </li>
-<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
+<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a><ul>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/model/index_cn.html">模型配置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/parameter/index_cn.html">参数设置</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/local/index_cn.html">本地训练与预测</a></li>
+<li class="toctree-l2"><a class="reference internal" href="../../faq/cluster/index_cn.html">集群训练与预测</a></li>
+</ul>
+</li>
 </ul>