resolved conflict

CMakeLists.txt

@@ -16,8 +16,6 @@ cmake_minimum_required(VERSION 3.0)
 set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} "${CMAKE_CURRENT_SOURCE_DIR}/cmake")
 set(PADDLE_SOURCE_DIR ${CMAKE_CURRENT_SOURCE_DIR})
 set(PADDLE_BINARY_DIR ${CMAKE_CURRENT_BINARY_DIR})
-SET(CMAKE_CXX_FLAGS_RELWITHDEBINFO "-O3 -g -DNDEBUG")
-SET(CMAKE_C_FLAGS_RELWITHDEBINFO "-O3 -g -DNDEBUG")
 
 include(system)
 
@@ -201,6 +199,10 @@ if(WITH_GOLANG)
 endif(WITH_GOLANG)
 
 set(PADDLE_PYTHON_BUILD_DIR "${CMAKE_CURRENT_BINARY_DIR}/python/build")
+
+SET(CMAKE_CXX_FLAGS_RELWITHDEBINFO "-O3 -g -DNDEBUG")
+SET(CMAKE_C_FLAGS_RELWITHDEBINFO "-O3 -g -DNDEBUG")
+
 add_subdirectory(paddle)
 if(WITH_PYTHON)
   add_subdirectory(python)

README.md

@@ -61,32 +61,32 @@ Please refer to our [release announcement](https://github.com/PaddlePaddle/Paddl
 ## Installation
 
 It is recommended to check out the
-[Docker installation guide](http://doc.paddlepaddle.org/develop/doc/getstarted/build_and_install/docker_install_en.html)
+[Docker installation guide](http://www.paddlepaddle.org/docs/develop/documentation/en/getstarted/build_and_install/docker_install_en.html)
 before looking into the
-[build from source guide](http://doc.paddlepaddle.org/develop/doc/getstarted/build_and_install/build_from_source_en.html).
+[build from source guide](http://www.paddlepaddle.org/docs/develop/documentation/en/getstarted/build_and_install/build_from_source_en.html).
 
 ## Documentation
 
-We provide [English](http://doc.paddlepaddle.org/develop/doc/) and
-[Chinese](http://doc.paddlepaddle.org/doc_cn/) documentation.
+We provide [English](http://www.paddlepaddle.org/docs/develop/documentation/en/getstarted/index_en.html) and
+[Chinese](http://www.paddlepaddle.org/docs/develop/documentation/zh/getstarted/index_cn.html) documentation.
 
-- [Deep Learning 101](http://book.paddlepaddle.org/index.html)
+- [Deep Learning 101](http://www.paddlepaddle.org/docs/develop/book/01.fit_a_line/index.html)
 
   You might want to start from this online interactive book that can run in a Jupyter Notebook.
 
-- [Distributed Training](http://doc.paddlepaddle.org/develop/doc/howto/usage/cluster/cluster_train_en.html)
+- [Distributed Training](http://www.paddlepaddle.org/docs/develop/documentation/en/howto/usage/cluster/cluster_train_en.html)
 
   You can run distributed training jobs on MPI clusters.
 
-- [Distributed Training on Kubernetes](http://doc.paddlepaddle.org/develop/doc/howto/usage/k8s/k8s_en.html)
+- [Distributed Training on Kubernetes](http://www.paddlepaddle.org/docs/develop/documentation/en/howto/usage/cluster/k8s_en.html)
 
    You can also run distributed training jobs on Kubernetes clusters.
 
-- [Python API](http://doc.paddlepaddle.org/develop/doc/api/index_en.html)
+- [Python API](http://www.paddlepaddle.org/docs/develop/documentation/en/api/index_en.html)
 
    Our new API enables much shorter programs.
 
-- [How to Contribute](http://doc.paddlepaddle.org/develop/doc/howto/dev/contribute_to_paddle_en.html)
+- [How to Contribute](http://www.paddlepaddle.org/docs/develop/documentation/en/howto/dev/contribute_to_paddle_en.html)
 
    We appreciate your contributions!
 

benchmark/IntelOptimizedPaddle.md

@@ -22,6 +22,7 @@ On each machine, we will test and compare the performance of training on single
 
 #### Training
 Test on batch size 64, 128, 256 on Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz
+Pay attetion that the speed below includes forward, backward and parameter update time. So we can not directly compare the data with the benchmark of caffe `time` [command](https://github.com/PaddlePaddle/Paddle/blob/develop/benchmark/caffe/image/run.sh#L9), which only contain forward and backward. The updating time of parameter would become very heavy when the weight size are large, especially on alexnet.
 
 Input image size - 3 * 224 * 224, Time: images/second
 
@@ -55,6 +56,16 @@ Input image size - 3 * 224 * 224, Time: images/second
 
 <img src="figs/googlenet-cpu-train.png" width="500">
 
+- Alexnet
+
+| BatchSize    | 64     | 128    | 256    |
+|--------------|--------| ------ | -------|
+| OpenBLAS     | 2.13   | 2.45   | 2.68   | 
+| MKLML        | 66.37  | 105.60 | 144.04 |
+| MKL-DNN      | 399.00 | 498.94 | 626.53 | 
+
+chart TBD
+
 #### Inference
 Test on batch size 1, 2, 4, 8, 16 on Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz
 - VGG-19

benchmark/paddle/image/alexnet.py

@@ -6,8 +6,18 @@ height = 227
 width = 227
 num_class = 1000
 batch_size = get_config_arg('batch_size', int, 128)
+gp = get_config_arg('layer_num', int, 1)
+is_infer = get_config_arg("is_infer", bool, False)
+num_samples = get_config_arg('num_samples', int, 2560)
 
-args = {'height': height, 'width': width, 'color': True, 'num_class': num_class}
+args = {
+    'height': height,
+    'width': width,
+    'color': True,
+    'num_class': num_class,
+    'is_infer': is_infer,
+    'num_samples': num_samples
+}
 define_py_data_sources2(
     "train.list", None, module="provider", obj="process", args=args)
 
@@ -31,7 +41,7 @@ net = img_pool_layer(input=net, pool_size=3, stride=2)
 
 # conv2
 net = img_conv_layer(
-    input=net, filter_size=5, num_filters=256, stride=1, padding=2, groups=1)
+    input=net, filter_size=5, num_filters=256, stride=1, padding=2, groups=gp)
 net = img_cmrnorm_layer(input=net, size=5, scale=0.0001, power=0.75)
 net = img_pool_layer(input=net, pool_size=3, stride=2)
 
@@ -40,11 +50,11 @@ net = img_conv_layer(
     input=net, filter_size=3, num_filters=384, stride=1, padding=1)
 # conv4
 net = img_conv_layer(
-    input=net, filter_size=3, num_filters=384, stride=1, padding=1, groups=1)
+    input=net, filter_size=3, num_filters=384, stride=1, padding=1, groups=gp)
 
 # conv5
 net = img_conv_layer(
-    input=net, filter_size=3, num_filters=256, stride=1, padding=1, groups=1)
+    input=net, filter_size=3, num_filters=256, stride=1, padding=1, groups=gp)
 net = img_pool_layer(input=net, pool_size=3, stride=2)
 
 net = fc_layer(
@@ -59,6 +69,9 @@ net = fc_layer(
     layer_attr=ExtraAttr(drop_rate=0.5))
 net = fc_layer(input=net, size=1000, act=SoftmaxActivation())
 
-lab = data_layer('label', num_class)
-loss = cross_entropy(input=net, label=lab)
-outputs(loss)
+if is_infer:
+    outputs(net)
+else:
+    lab = data_layer('label', num_class)
+    loss = cross_entropy(input=net, label=lab)
+    outputs(loss)

benchmark/paddle/image/googlenet.py

@@ -7,13 +7,15 @@ num_class = 1000
 batch_size = get_config_arg('batch_size', int, 128)
 use_gpu = get_config_arg('use_gpu', bool, True)
 is_infer = get_config_arg("is_infer", bool, False)
+num_samples = get_config_arg('num_samples', int, 2560)
 
 args = {
     'height': height,
     'width': width,
     'color': True,
     'num_class': num_class,
-    'is_infer': is_infer
+    'is_infer': is_infer,
+    'num_samples': num_samples
 }
 define_py_data_sources2(
     "train.list" if not is_infer else None,

benchmark/paddle/image/provider.py

@@ -14,6 +14,7 @@ def initHook(settings, height, width, color, num_class, **kwargs):
     else:
         settings.data_size = settings.height * settings.width
     settings.is_infer = kwargs.get('is_infer', False)
+    settings.num_samples = kwargs.get('num_samples', 2560)
     if settings.is_infer:
         settings.slots = [dense_vector(settings.data_size)]
     else:
@@ -23,7 +24,7 @@ def initHook(settings, height, width, color, num_class, **kwargs):
 @provider(
     init_hook=initHook, min_pool_size=-1, cache=CacheType.CACHE_PASS_IN_MEM)
 def process(settings, file_list):
-    for i in xrange(2560 if settings.is_infer else 1024):
+    for i in xrange(settings.num_samples):
         img = np.random.rand(1, settings.data_size).reshape(-1, 1).flatten()
         if settings.is_infer:
             yield img.astype('float32')

benchmark/paddle/image/resnet.py

@@ -7,13 +7,15 @@ num_class = 1000
 batch_size = get_config_arg('batch_size', int, 64)
 layer_num = get_config_arg("layer_num", int, 50)
 is_infer = get_config_arg("is_infer", bool, False)
+num_samples = get_config_arg('num_samples', int, 2560)
 
 args = {
     'height': height,
     'width': width,
     'color': True,
     'num_class': num_class,
-    'is_infer': is_infer
+    'is_infer': is_infer,
+    'num_samples': num_samples
 }
 define_py_data_sources2(
     "train.list" if not is_infer else None,

benchmark/paddle/image/run_mkldnn_infer.sh → benchmark/paddle/image/run_mkl_infer.sh

@@ -37,7 +37,7 @@ function infer() {
       --trainer_count=1 \
       --num_passes=1 \
       --save_dir="models/${topology}-${layer_num}" \
-      --config_args="batch_size=128,layer_num=${layer_num}" \
+      --config_args="batch_size=128,layer_num=${layer_num},num_samples=256" \
       > /dev/null 2>&1
     echo "Done"
   fi
@@ -79,8 +79,9 @@ fi
 # inference benchmark
 for use_mkldnn in True False; do
   for batchsize in 1 2 4 8 16; do
-    infer googlenet v1 $batchsize $use_mkldnn
-    infer resnet 50 $batchsize $use_mkldnn
     infer vgg 19 $batchsize $use_mkldnn
+    infer resnet 50 $batchsize $use_mkldnn
+    infer googlenet v1 $batchsize $use_mkldnn
+    infer alexnet 2 $batchsize $use_mkldnn
   done
 done

benchmark/paddle/image/run_mkldnn_train.sh → benchmark/paddle/image/run_mkl_train.sh

@@ -28,6 +28,10 @@ function train() {
     --test_period=100 \
     --config_args=$args \
     2>&1 | tee ${log} 
+
+  avg_time=`tail ${log} -n 1 | awk -F ' ' '{print $8}' | sed 's/avg=//'`
+  fps=`awk 'BEGIN{printf "%.2f",('$bs' / '$avg_time' * 1000)}'`
+  echo "FPS: $fps images/sec" 2>&1 | tee -a ${log}
 }
 
 if [ ! -f "train.list" ]; then
@@ -43,5 +47,6 @@ for use_mkldnn in True False; do
     train vgg 19 $batchsize $use_mkldnn
     train resnet 50 $batchsize $use_mkldnn
     train googlenet v1 $batchsize $use_mkldnn
+    train alexnet 2 $batchsize $use_mkldnn
   done
 done

benchmark/paddle/image/run_openblas_infer.sh

+set -e
+
+function clock_to_seconds() {
+  hours=`echo $1 | awk -F ':' '{print $1}'`
+  mins=`echo $1 | awk -F ':' '{print $2}'`
+  secs=`echo $1 | awk -F ':' '{print $3}'`
+  echo `awk 'BEGIN{printf "%.2f",('$secs' + '$mins' * 60 + '$hours' * 3600)}'`
+}
+
+function infer() {
+  unset OMP_NUM_THREADS MKL_NUM_THREADS OMP_DYNAMIC KMP_AFFINITY
+  topology=$1
+  layer_num=$2
+  bs=$3
+  thread=`nproc`
+  if [ $thread -gt $bs ]; then
+    thread=$bs
+  fi
+  log="logs/infer-${topology}-${layer_num}-${thread}openblas-${bs}.log"
+
+  models_in="models/${topology}-${layer_num}/pass-00000/"
+  if [ ! -d $models_in ]; then
+    echo "./run_mkl_infer.sh to save the model first"
+    exit 0
+  fi
+  log_period=$((32 / bs))
+  paddle train --job=test \
+    --config="${topology}.py" \
+    --use_mkldnn=False \
+    --use_gpu=False \
+    --trainer_count=$thread \
+    --log_period=$log_period \
+    --config_args="batch_size=${bs},layer_num=${layer_num},is_infer=True,num_samples=256" \
+    --init_model_path=$models_in \
+    2>&1 | tee ${log}
+
+  # calculate the last 5 logs period time of 160(=32*5) samples,
+  # the time before are burning time.
+  start=`tail ${log} -n 7 | head -n 1 | awk -F ' ' '{print $2}' | xargs`
+  end=`tail ${log} -n 2 | head -n 1 | awk -F ' ' '{print $2}' | xargs`
+  start_sec=`clock_to_seconds $start`
+  end_sec=`clock_to_seconds $end`
+  fps=`awk 'BEGIN{printf "%.2f",(160 / ('$end_sec' - '$start_sec'))}'`
+  echo "Last 160 samples start: ${start}(${start_sec} sec), end: ${end}(${end_sec} sec;" >> ${log}
+  echo "FPS: $fps images/sec" 2>&1 | tee -a ${log}
+}
+
+if [ ! -f "train.list" ]; then
+  echo " " > train.list
+fi
+if [ ! -f "test.list" ]; then
+  echo " " > test.list
+fi
+if [ ! -d "logs" ]; then
+  mkdir logs
+fi
+
+# inference benchmark
+for batchsize in 1 2 4 8 16; do
+  infer vgg 19 $batchsize
+  infer resnet 50 $batchsize 
+  infer googlenet v1 $batchsize
+  infer alexnet 2 $batchsize
+done

benchmark/paddle/image/run_openblas_train.sh

+set -e
+
+function train() {
+  unset OMP_NUM_THREADS MKL_NUM_THREADS OMP_DYNAMIC KMP_AFFINITY
+  topology=$1
+  layer_num=$2
+  bs=$3
+  thread=`nproc`
+  # each trainer_count use only 1 core to avoid conflict
+  log="logs/train-${topology}-${layer_num}-${thread}openblas-${bs}.log"
+  args="batch_size=${bs},layer_num=${layer_num}"
+  config="${topology}.py"
+  paddle train --job=time \
+    --config=$config \
+    --use_mkldnn=False \
+    --use_gpu=False \
+    --trainer_count=$thread \
+    --log_period=3 \
+    --test_period=30 \
+    --config_args=$args \
+    2>&1 | tee ${log} 
+
+  avg_time=`tail ${log} -n 1 | awk -F ' ' '{print $8}' | sed 's/avg=//'`
+  fps=`awk 'BEGIN{printf "%.2f",('$bs' / '$avg_time' * 1000)}'`
+  echo "FPS: $fps images/sec" 2>&1 | tee -a ${log}
+}
+
+if [ ! -f "train.list" ]; then
+  echo " " > train.list
+fi
+if [ ! -d "logs" ]; then
+  mkdir logs
+fi
+
+# training benchmark
+for batchsize in 64 128 256; do
+  train vgg 19 $batchsize
+  train resnet 50 $batchsize
+  train googlenet v1 $batchsize
+  train alexnet 2 $batchsize
+done

benchmark/paddle/image/vgg.py

@@ -7,13 +7,15 @@ num_class = 1000
 batch_size = get_config_arg('batch_size', int, 64)
 layer_num = get_config_arg('layer_num', int, 19)
 is_infer = get_config_arg("is_infer", bool, False)
+num_samples = get_config_arg('num_samples', int, 2560)
 
 args = {
     'height': height,
     'width': width,
     'color': True,
     'num_class': num_class,
-    'is_infer': is_infer
+    'is_infer': is_infer,
+    'num_samples': num_samples
 }
 define_py_data_sources2(
     "train.list" if not is_infer else None,

cmake/external/protobuf.cmake

@@ -253,9 +253,9 @@ IF(NOT PROTOBUF_FOUND)
     IF(WITH_C_API)
         INSTALL(DIRECTORY ${PROTOBUF_INCLUDE_DIR} DESTINATION third_party/protobuf)
         IF(ANDROID)
-            INSTALL(FILES ${PROTOBUF_LIBRARY} DESTINATION third_party/protobuf/lib/${ANDROID_ABI})
+            INSTALL(FILES ${PROTOBUF_LITE_LIBRARY} DESTINATION third_party/protobuf/lib/${ANDROID_ABI})
         ELSE()
-            INSTALL(FILES ${PROTOBUF_LIBRARY} DESTINATION third_party/protobuf/lib)
+            INSTALL(FILES ${PROTOBUF_LITE_LIBRARY} DESTINATION third_party/protobuf/lib)
         ENDIF()
     ENDIF()
 

doc/api/index_cn.rst

@@ -7,3 +7,4 @@ API
     模型配置 <v2/model_configs.rst>
     数据访问 <v2/data.rst>
     训练与应用 <v2/run_logic.rst>
+    v2/fluid.rst

doc/api/v2/config/layer.rst

@@ -252,6 +252,11 @@ first_seq
 ..  autoclass:: paddle.v2.layer.first_seq
     :noindex:
 
+sub_seq
+---------
+..  autoclass:: paddle.v2.layer.sub_seq
+    :noindex:
+
 concat
 ------
 ..  autoclass:: paddle.v2.layer.concat
@@ -467,7 +472,7 @@ lambda_cost
     :noindex:
 
 square_error_cost
---------
+-----------------
 ..  autoclass:: paddle.v2.layer.square_error_cost
     :noindex:
 
@@ -533,7 +538,7 @@ Miscs
 =====
 
 dropout
---------------
+--------
 ..  autoclass:: paddle.v2.layer.dropout
     :noindex:
 

doc/api/v2/fluid/layers.rst

@@ -19,17 +19,17 @@ dynamic_lstm
     :noindex:
 
 data
----------
+----
 ..  autofunction:: paddle.v2.fluid.layers.data
     :noindex:
 
 mean
----------
+----
 ..  autofunction:: paddle.v2.fluid.layers.mean
     :noindex:
 
 mul
----------
+---
 ..  autofunction:: paddle.v2.fluid.layers.mul
     :noindex:
 
@@ -45,13 +45,13 @@ elementwise_div
 
 
 dropout
----------
+-------
 ..  autofunction:: paddle.v2.fluid.layers.dropout
     :noindex:
 
 
 reshape
----------
+--------
 ..  autofunction:: paddle.v2.fluid.layers.reshape
     :noindex:
 
@@ -68,12 +68,6 @@ scale
     :noindex:
 
 
-reshape
----------
-..  autofunction:: paddle.v2.fluid.layers.reshape
-    :noindex:
-
-
 transpose
 ---------
 ..  autofunction:: paddle.v2.fluid.layers.transpose
@@ -81,67 +75,67 @@ transpose
 
 
 sigmoid_cross_entropy_with_logits
----------
+---------------------------------
 ..  autofunction:: paddle.v2.fluid.layers.esigmoid_cross_entropy_with_logits
     :noindex:
 
 
 cast
----------
+----
 ..  autofunction:: paddle.v2.fluid.layers.cast
     :noindex:
 
 
 concat
----------
+-------
 ..  autofunction:: paddle.v2.fluid.layers.concat
     :noindex:
 
 
 sums
----------
+----
 ..  autofunction:: paddle.v2.fluid.layers.sums
     :noindex:
 
 
 linear_chain_crf
----------
+----------------
 ..  autofunction:: paddle.v2.fluid.layers.linear_chain_crf
     :noindex:
 
 
 assign
----------
+-------
 ..  autofunction:: paddle.v2.fluid.layers.embedding
     :noindex:
 
 
 split_lod_tensor
----------
+----------------
 ..  autofunction:: paddle.v2.fluid.layers.split_lod_tensor
     :noindex:
 
 
 merge_lod_tensor
----------
+----------------
 ..  autofunction:: paddle.v2.fluid.layers.merge_lod_tensor
     :noindex:
 
 cos_sim
----------
+--------
 ..  autofunction:: paddle.v2.fluid.layers.cos_sim
     :noindex:
 
 
 cross_entropy
----------
+-------------
 ..  autofunction:: paddle.v2.fluid.layers.cross_entropy
     :noindex:
 
 
 
 square_error_cost
----------
+-----------------
 ..  autofunction:: paddle.v2.fluid.layers.square_error_cost
     :noindex:
 
@@ -153,74 +147,80 @@ accuracy
 
 
 sequence_conv
----------
+-------------
 ..  autofunction:: paddle.v2.fluid.layers.sequence_conv
     :noindex:
 
 
 conv2d
----------
+------
 ..  autofunction:: paddle.v2.fluid.layers.conv2d
     :noindex:
 
 
 sequence_pool
----------
+-------------
 ..  autofunction:: paddle.v2.fluid.layers.sequence_pool
     :noindex:
 
 
+sequence_first_step
+-------------------
+..  autofunction:: paddle.v2.fluid.layers.sequence_first_step
+    :noindex:
+
+
+sequence_last_step
+------------------
+..  autofunction:: paddle.v2.fluid.layers.sequence_last_step
+    :noindex:
+
+
 pool2d
----------
+------
 ..  autofunction:: paddle.v2.fluid.layers.pool2d
     :noindex:
 
 
 batch_norm
----------
+----------
 ..  autofunction:: paddle.v2.fluid.layers.batch_norm
     :noindex:
 
 
 beam_search_decode
----------
+------------------
 ..  autofunction:: paddle.v2.fluid.layers.beam_search_decode
     :noindex:
 
 
-lstm
----------
-..  autofunction:: paddle.v2.fluid.layers.lstm
-    :noindex:
-
-
 lod_rank_table
----------
+--------------
 ..  autofunction:: paddle.v2.fluid.layers.lod_rank_table
     :noindex:
 
 
 max_sequence_len
----------
+----------------
 ..  autofunction:: paddle.v2.fluid.layers.max_sequence_len
     :noindex:
 
 
 topk
----------
+-----
 ..  autofunction:: paddle.v2.fluid.layers.topk
     :noindex:
 
 
 lod_tensor_to_array
----------
+-------------------
 ..  autofunction:: paddle.v2.fluid.layers.lod_tensor_to_array
     :noindex:
 
 
 
 array_to_lod_tensor
----------
+-------------------
 ..  autofunction:: paddle.v2.fluid.layers.array_to_lod_tensor
     :noindex:
 
@@ -228,26 +228,26 @@ array_to_lod_tensor
 
 
 fill_constant
----------
+-------------
 ..  autofunction:: paddle.v2.fluid.layers.fill_constant
     :noindex:
 
 
 
 fill_constant_batch_size_like
----------
+-----------------------------
 ..  autofunction:: paddle.v2.fluid.layers.fill_constant_batch_size_like
     :noindex:
 
 
 ones
----------
+----
 ..  autofunction:: paddle.v2.fluid.layers.ones
     :noindex:
 
 
 zeros
----------
+-----
 ..  autofunction:: paddle.v2.fluid.layers.zeros
     :noindex:
 
@@ -259,14 +259,14 @@ increment
 
 
 array_write
----------
+-----------
 ..  autofunction:: paddle.v2.fluid.layers.array_write
     :noindex:
 
 
 
 create_array
----------
+------------
 ..  autofunction:: paddle.v2.fluid.layers.create_array
     :noindex:
 
@@ -278,25 +278,67 @@ less_than
 
 
 array_read
----------
+----------
 ..  autofunction:: paddle.v2.fluid.layers.array_read
     :noindex:
 
 
 shrink_memory
----------
+--------------
 ..  autofunction:: paddle.v2.fluid.layers.shrink_memory
     :noindex:
 
 
 array_length
----------
+-------------
 ..  autofunction:: paddle.v2.fluid.layers.array_length
     :noindex:
 
 
 conv2d_transpose
----------
+----------------
 ..  autofunction:: paddle.v2.fluid.layers.conv2d_transpose
     :noindex:
 
+
+sequence_expand
+---------------
+..  autofunction:: paddle.v2.fluid.layers.sequence_expand
+    :noindex:
+
+
+lstm_unit
+---------
+..  autofunction:: paddle.v2.fluid.layers.lstm_unit
+    :noindex:
+
+
+sequence_softmax
+----------------
+..  autofunction:: paddle.v2.fluid.layers.sequence_softmax
+    :noindex:
+
+
+reduce_sum
+----------
+..  autofunction:: paddle.v2.fluid.layers.reduce_sum
+    :noindex:
+
+
+reduce_mean
+-----------
+..  autofunction:: paddle.v2.fluid.layers.reduce_mean
+    :noindex:
+
+
+reduce_max
+----------
+..  autofunction:: paddle.v2.fluid.layers.reduce_max
+    :noindex:
+
+
+reduce_min
+----------
+..  autofunction:: paddle.v2.fluid.layers.reduce_min
+    :noindex:
+

doc/api/v2/fluid/nets.rst

@@ -3,19 +3,19 @@ Nets
 ===========
 
 simple_img_conv_pool
------------
+--------------------
 ..  autofunction:: paddle.v2.fluid.nets.simple_img_conv_pool
     :noindex:
 
 
 img_conv_group
------------
+---------------
 ..  autofunction:: paddle.v2.fluid.nets.img_conv_group
     :noindex:
 
 
 sequence_conv_pool
------------
+------------------
 ..  autofunction:: paddle.v2.fluid.nets.sequence_conv_pool
     :noindex:
 

doc/api/v2/fluid/optimizer.rst

@@ -18,7 +18,7 @@ SGDOptimizer
 
 
 MomentumOptimizer
------------
+-----------------
 ..  automodule:: paddle.v2.fluid.optimizer
     :members: MomentumOptimizer
     :noindex:
@@ -26,14 +26,14 @@ MomentumOptimizer
 
 
 AdagradOptimizer
------------
+----------------
 ..  automodule:: paddle.v2.fluid.optimizer
     :members: AdagradOptimizer
     :noindex:
 
 
 AdamOptimizer
------------
+-------------
 ..  automodule:: paddle.v2.fluid.optimizer
     :members: AdamOptimizer
     :noindex:
@@ -47,7 +47,7 @@ AdamaxOptimizer
 
 
 DecayedAdagradOptimizer
------------
+-----------------------
 ..  automodule:: paddle.v2.fluid.optimizer
     :members: DecayedAdagradOptimizer
     :noindex:

doc/api/v2/fluid/regularizer.rst

@@ -3,14 +3,14 @@ Regularizer
 ===========
 
 WeightDecayRegularizer
------------
+----------------------
 ..  automodule:: paddle.v2.fluid.regularizer
     :members: WeightDecayRegularizer
     :noindex:
 
 
 L2DecayRegularizer
------------
+------------------
 ..  automodule:: paddle.v2.fluid.regularizer
     :members: L2DecayRegularizer
     :noindex:
@@ -18,7 +18,7 @@ L2DecayRegularizer
 
 
 L1DecayRegularizer
------------
+-------------------
 ..  automodule:: paddle.v2.fluid.regularizer
     :members: L1DecayRegularizer
 

doc/design/block.md

@@ -291,10 +291,10 @@ public:
   }
 
   void Run(const framework::Scope& scope,
-           const platform::DeviceContext& dev_ctx) const override {
+           const platform::Place& place) const override {
     PADDLE_ENFORCE(symbols_ready_, "operators and variables should be created first.");
     for (auto& op : runtime_table_.ops()) {
-      op->Run(scope, dev_ctx);
+      op->Run(scope, place);
     }
   }
 

doc/design/executor.md

 # Executor Design Doc
 
 ## Motivation
+In [fluid](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/fluid.md), we encourage the user to use deep learning programming paradigms to describe the training process. When the user-written Python program is executed, it will first create a protobuf message
+[`ProgramDesc`](https://github.com/PaddlePaddle/Paddle/blob/a91efdde6910ce92a78e3aa7157412c4c88d9ee8/paddle/framework/framework.proto#L145) that describes the process and is conceptually like an [abstract syntax tree](https://en.wikipedia.org/wiki/Abstract_syntax_tree).
 
-We use executor to do the runtime evaluation of a `ProgramDesc`.
+The executor runs the `ProgramDesc` like an interpreter. `ProgramDesc` contains the intrinsics (operators in this case) and variables which will be used, executor explicitly executes the stored precompiled code.
 
 ## Overview
 
-An executor takes a `ProgramDesc`, a `block_id` and a `Scope`.  The `ProgramDesc` is a list of blocks and each block contains the protobuf definition of all the parameters and operators. The `block_id` specifies the entrance block. And the `Scope` is the container of all the variable instance, which is persistent throughout different runs.
+An executor takes a `ProgramDesc`, a `block_id` and a `Scope`.  The `ProgramDesc` is a list of blocks and each block contains the protobuf definition of all the parameters and operators in the block. The `block_id` specifies the entrance block. And the `Scope` is the container of all the variable instances, which is persistent throughout different runs.
 
-### What does executor do?
+## Executor
 
-It evaluates all the operators in the `block_id`th block of a `ProgramDesc`.
+The `Executor` explicitly executes all the intrinsics (operators here) in the `block_id`th block of a `ProgramDesc`. Essentially, it instantiates Variables and Operators, then runs all the operators in sequence one-by-one.
+It is very similar to how a push stack frame works when entering a block, following which it cleans up all the temporary variables when a mini-batch is finished. It does not however, have the stack frame pop process.
 
-### What does executor NOT do?
+### The interface
+```c++
+  Executor(places);
+```
+A executor does not own any computing resources, a user can only construct an executor using the specified places.
 
-It does not do runtime optimization, meaning intelligently parse the dependency of each op a choose which one to be run and in which order they should be run.
+### Running an Executor
 
-It does not do graph partitioning, meaning dividing the `ProgramDesc` into several small pieces and executing them on different devices.
-
-## Implementation
-
-`Executor` evaluates a `ProgramDesc`. Essentially, it instantiates Variables and Operators, then run all the operators in sequence. [[code]](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/framework/executor.cc)
+```
+  void Run(ProgramDesc, Scope, block_id, create_local_scope);
+```
+An `Executor` only provides a unified way to execute `ProgramDesc`. `ProgramDesc` is the target that will be executed, the `Scope` specifies the variable container, the `block_id` indicates the entrance block and `create_local_scope` is a boolean that states whether it will destroy the temporary variables after the execution is finished.

doc/design/kernel_hint_design.md

+## Problem
+In PaddlePaddle's [Design](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/switch_kernel.md), one Operator may have multiple kernels. Users may have some personal preference to choose a certain type of kernel for an operator, such as `force_cpu` to choose a CPU kernel, `use_cudnn` to choose a CUDNN kernel, we need to provide a way for users to do this.
+
+In the current design, we use KernelType to describe one kernel.
+
+```cpp
+struct KernelType {
+  Place place_;
+  DataType data_type_;
+  LayoutType layout_;
+};
+```
+ `place_` `data_type_` and `layout_` can be got from the input tensors of the operator, `GetActualKernelType(inputs)` use inputs to infer the proper kernel key that fit the incoming data, but users can not directly configure it.
+
+The [design](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/switch_kernel.md) also provides a virtual method `GetExpectedKernelType` that user can overload and use to choose the KernelType they want to use.
+
+So we should send the information user defined in proto to `GetExpectedKernelType` for choosing a kernel.
+
+The problem is, how should we define and send the information for `GetExpectedKernelType` to use?
+
+## Solution
+
+### Potential choice
+1. Do nothing, let the user add the information they want to operator‘s attribute and get them inside `GetExpectedKernelType`, this can work properly. But there is a little problem that users may define many kinds of hints for the same purpose, such as `force_cpu`, `use_cpu`, `cpu_kernel` to choose CPU kernel, and `use_cudnn`, `force_cudnn`, `cudnn_kernel` to choose CUDNN kernel.
+
+2. Pre-define all the needed option and use a single attr key such as `kernel_hint` for the user, this is not so flexible if the user wants to define some more kind of hint.
+
+### Final choice
+To provide enough flexibility while avoiding confusion definition, we can define some global constants for these attribute names, such as `force_cpu`, `use_cudnn`, `use_mkldnn` for a user to choose.
+
+In C++
+
+```cpp
+const std::string kForceCPU = "force_cpu";
+const std::string kUseCUDNN = "use_cudnn";
+const std::string kUseMKLDNN = "use_mkldnn";
+
+KernelType GetExpectedKernelType() {
+  if (Attr<bool>(kForceCPU)) {
+    return KernelType(CPUPlace, ...)
+  } else {
+    ...
+  }
+}
+```
+
+In Python code
+
+```python
+FORCE_CPU = core.kForceCPU()
+
+def xx_layer(..., force_cpu=false):
+  layer_helper = LayerHelper(...)
+  layer_helper.append_op(
+    type="xx",
+    attr={FORCE_CPU: force_cpu})
+```

doc/design/mkl/mkl_packed.md

+# Intel® MKL Packed on PaddlePaddle: Design Doc
+
+
+## Contents
+
+- [Overview](#overview)
+- [Key Points](#key-points) 
+   - [Background](#background)
+   - [Solution](#solution)
+- [Actions](#actions)
+    - [CMake](#cmake)
+	- [Layers](#layers)
+	- [Unit Tests](#unit-tests)
+	- [Python API](#python-api)
+	- [Benchmarking](#benchmarking)
+
+
+## Overview
+我们计划将 Intel® MKL 中引入的 GEMM Packed APIs\[[1](#references)\] 集成到 PaddlePaddle 中，充分发挥英特尔平台的优势，有效提升PaddlePaddle在英特尔架构上的性能。
+现阶段的优化主要针对 Recurrent Neural Network（以下简称RNN）相关层（包括`RecurrentLayer`, `GatedRecurrentLayer`和`LstmLayer`）， 以及 PaddlePaddle V1 API。
+
+## Key Points
+
+### Background
+目前PaddlePaddle采用了 Intel® MKL库的[cblas_?gemm](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm)函数，这个函数本身会在计算前将原数据转换为更适合英特尔平台的内部格式。
+
+1. 转换耗时 \
+这一数据格式的转换操作（Packing），在问题本身的计算量比较小的时候，显得相对来说较为耗时。例如在DeepSpeech2 \[[2](#references)\] 的Vanilla RNN部分中，矩阵大小是`batch_size * 2048`。
+2. 转换冗余 \
+由于在现有的某些情况下（例如RNN），多次调用 cblas_?gemm 会使用相同的原数据，因此，每次调用时对原数据的重复Packing便成为了冗余。
+
+为了最大程度减少多次调用 cblas_?gemm 在Packing上的耗时，Intel® MKL 引入了以下四个API:
+   * [cblas_?gemm_alloc](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-alloc)
+   * [cblas_?gemm_pack](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-pack)
+   * [cblas_?gemm_compute](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-compute)
+   * [cblas_?gemm_free](https://software.intel.com/en-us/mkl-developer-reference-c-cblas-gemm-free)
+
+通过使用这些API，我们可以先完成对原数据的Packing操作，再把已转换为Packed格式的数据传递给那些复用同一数据的gemm_compute函数，从而避免了Packing冗余。
+
+### Solution
+在RNN的情况下，同一次前向、后向（forward/backward）过程中所有时间步（time step）共享同一个权重（weight）。当只做推断（inference）时，各次前向之间也都使用了相同的权重，没有必要在每次前向中每个时间步的计算时对权重进行重复的Packing操作。
+
+我们通过使用新引入的GEMM Packed APIs，在层初始化的时候，先完成对权重的Packing操作，然后在前向，后向时复用已经转换过的权重，并在每次权重更新后，对新的权重进行转换用于下次迭代。
+
+* 优化前，对于序列长度（sequence length）为`T`的网络模型（model）, `N`次迭代执行的转换次数为：
+  - `inference`： `N * T`  
+  - `training`： `2 * N * T`
+* 优化后，对于同样设置的网络模型，其转换次数减少至：
+  - `inference`： `1`    
+  - `training`： `2 * N`
+
+## Actions
+
+添加的相关文件和目录结构如下：
+
+```txt
+PaddlePaddle/Paddle
+├── ...
+└── paddle/
+    ├── ...
+    └── gserver/
+        ├── ...
+        ├── layers/
+        │   ├── ...
+        │   ├── MKLPackedRecurrentLayer.*
+        |   ├── MKLPackedGatedRecurrentLayer.*
+        |   ├── MKLPackedLstmLayer.*
+        |   └── MKLPackedGemm.h
+        └── tests/
+            ├── ...
+            └── test_MKLPacked.cpp
+```
+
+### CMake
+在对应的`CMakeLists.txt`中根据`WITH_MKL`是否打开，来决定是否开启MKL Packed相关功能。
+
+### Layers
+所有的`MKLPacked*Layer`都继承于PaddlePaddle的基类`Layer`, 并添加头文件 `MKLPackedGemm.h`，该文件对相关GEMM Packed APIs做了封装。
+
+### Unit Tests
+我们会添加`test_MKLPacked.cpp`用于MKL Packed优化后layer的测试。
+对于每一个新加的RNN layer，我们会对比如下2个方面：
+1. 对比优化后layer自身，sequence mode（`rnn_use_batch=false`）与batch mode(`rnn_use_batch=true`)的结果。
+2. 对比优化后layer与相对应的PaddlePaddle原有layer, 在batch mode下的结果。
+
+### Python API
+计划在`paddle/utils.Flags`中添加`use_mkl_packed`的flag，用于选择是否使用相关功能，并且当编译时`WITH_MKL=ON`的情况下，默认设置为`true`。
+
+同时，在`python/paddle/trainer/config_parser.py`中对应的layer处，添加`use_mkl_packed`这个选择，方便用户在Python端选择是否启用这个功能。
+
+具体实现方式比如：
+
+```python
+use_mkl_packed = bool(int(g_command_config_args.get("use_mkl_packed", 0)))
+if use_mkl_packed:
+    self.layer_type = mkl_packed_*
+```
+
+所有相关的`layer_type`会以*mkl_packed_*开头，这些会在`MKLPacked*Layer`注册layer的时候保证，以示区分。 
+
+
+### Benchmarking
+会添加相应的脚本用于测试和对比在使用MKL Packed recurrent layers 前后的网络性能。
+
+## References 
+1. [Introducing the new Packed APIs for GEMM](https://software.intel.com/en-us/articles/introducing-the-new-packed-apis-for-gemm)
+2. [DeepSpeech2 on PaddlePaddle](https://github.com/PaddlePaddle/DeepSpeech#deepspeech2-on-paddlepaddle)
+

doc/design/mkldnn/README.MD → doc/design/mkl/mkldnn.md

@@ -208,4 +208,3 @@ if use_mkldnn
 但是在PaddlePaddle中，无论是重构前的layer还是重构后的op，都不会想要知道next layer/op的信息。
 4. MKL-DNN的高性能格式与PaddlePaddle原有的`NCHW`不同(PaddlePaddle中的cuDNN部分使用的也是`NCHW`，所以不存在这个问题)。
 所以需要引入一个转换方法，并且只需要在必要的时候转换这种格式，才能更好的发挥MKL-DNN的性能。
-

doc/design/mkl/mkldnn_fluid.md

+# Design Doc: Add MKLDNN Kernel in Fluid Operator
+
+## Principles
+
+First of all, we should follow some basical principles like:
+1.  [How to write a new operator](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/howto/dev/new_op_en.md). We are trying to add a new kind of kernel into operators, so basically we should follow this doc.
+2.  [Supporting new Device/Library](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/support_new_device.md). Since MKLDNN is a new library to fluid, we should add `MKLDNNDeviceContext` and maybe `mkldnn_helper.h`, just like [cudnn_helper.h](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/cudnn_helper.h).
+3.  [Switch Kernel](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/switch_kernel.md). Another important point is that we should ensure the data synchronization between different kernel types, which is this [topic](https://github.com/PaddlePaddle/Paddle/issues/6549). So basically we should override `GetExpectedKernelType` and `trans` functions to support switching kernels.
+4.  [The Keys of Operator Kernel Type](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/operator_kernel_type.md). Kernel Type is a pivotal conception which can record the `Place`, `Library`, `DataType` and `Layout`.
+
+## Sulution
+
+In general, there are four parts we should follow to run a MKL-DNN primitive.
+-  Create a primitive descriptor that describe this operator
+-  Create a primitive itself by primitive descriptor and the engine
+-  Create all memory buffers that primitive needed
+-  Launch a stream to execute the primitive created
+More details can refer to [here](http://01org.github.io/mkl-dnn).
+
+It's better to avoid reinitialization of primitives and memory handles in the first three stages in every iteration. \
+So we plan to create a map to record all the `primitive` and `memory`, which should not take too much memories as discussed [here](https://github.com/PaddlePaddle/Paddle/issues/6822).
+
+It's assumed that following three conditions should be satisfied.
+1. there is a unique key for each operator instance. May be the actual name of `Output Tensor`.
+2. the `Input Tensor` inside `Compute` function is the one after converted.
+3. we can get the phase(eg. `is_test`) inside `Compute` function, otherwise we need to expose this attribue to user.
+
+### Compute
+The algorithm of `Compute` would be described as follow, let's take conv like an example.
+
+```c++
+
+  PADDLE_ENFORCE(platform::is_cpu_place(ctx.GetPlace()), "It must use CPUPlace.");
+  PADDLE_ENFORCE(platform::is_mkldnn_library(ctx.GetLibrary()), "It must use MKLDNN Library.");
+
+  auto& dev_ctx = ctx.template device_context<platform::MKLDNNDeviceContext>();
+
+  // find primitive by unique key from mkldnn context
+  // the op_key should be a unique name of this op instance
+  auto& p = dev_ctx.findPrimitive(op_key + "_fwd");
+
+  // assuming the input tensor inside this compute function is the one after converted
+  // this point should be guarantee by another mechanism
+  auto& i = dev_ctx.findMemory(op_key + "_input");
+  
+  if (p == nullptr || i == nullptr || inputSizeChanged(p, i))  {
+    auto fwd_primitive_desc = createPrimitiveDesc(ctx);
+    auto* input = ctx.Input<Tensor>("Input");
+    auto* filter = ctx.Input<Tensor>("Filter");
+    auto* output = ctx.Output<Tensor>("Output");
+    shared_ptr<mkldnn::memory> in(new mkldnn::memory(fwd_primitive_desc->src_primitive_desc(), input->data<T>()));
+    shared_ptr<mkldnn::memory> wgt(new mkldnn::memory(fwd_primitive_desc->weights_primitive_desc(), filter->data<T>()));
+    shared_ptr<mkldnn::memory> out(new mkldnn::memory(fwd_primitive_desc->dst_primitive_desc(), output->mutable_data<T>(ctx.GetPlace())));
+    shared_ptr<mkldnn::conv_fwd> fwd_primitive(new mkldnn::conv_fwd(*fwd_primitive_desc, *in, *wgt, *out));
+
+    dev_ctx.addMemory(op_key+"_input", in);
+    dev_ctx.addMemory(op_key+"_output", out);
+    dev_ctx.addMemory(op_key+"_filer", wgt);
+    dev_ctx.addPrimitive(op_key+"_fwd", fwd_primitive);
+    dev_ctx.addPrimitiveDesc(op_key+"_fwd_PD", fwd_primitive_desc);
+  }
+
+  p = dev_ctx.findPrimitive(op_key + "_fwd");
+
+  PADDLE_ENFORCE(p, "Should have forward Primitive");
+  PADDLE_ENFORCE(dev_ctx.findMemory(op_unique_key+"_input"), "Should have input memory");
+  PADDLE_ENFORCE(dev_ctx.findMemory(op_unique_key+"_output"), "Should have output memory");
+  PADDLE_ENFORCE(dev_ctx.findMemory(op_unique_key+"_filter"), "Should have filter memory");
+  PADDLE_ENFORCE(dev_ctx.findPrimitiveDesc(op_unique_key+"_fwd_PD"), "Should have forward PrimitiveDesc");
+  dev_ctx.submit(p);
+  dev_ctx.execute();  // the convert primitive should have already contained.
+
+```
+
+The `createPrimitiveDesc` returns the primitive descripotor of this operator, would be like this:
+```c++
+  auto* input = ctx.Input<Tensor>("Input");
+  auto* filter = ctx.Input<Tensor>("Filter");
+  auto* output = ctx.Output<Tensor>("Output");
+  std::vector<int> strides = ctx.Attr<std::vector<int>>("strides");
+  std::vector<int> paddings = ctx.Attr<std::vector<int>>("paddings");
+  std::vector<int> dilations = ctx.Attr<std::vector<int>>("dilations");
+  int groups = ctx.Attr<int>("groups");
+  algorithm algo = static_cast<algorithm>(ctx.Attr<int>("convolution_algorithm_option"));
+  prop_kind pk = ctx.Attr<bool>("is_test") ? prop_kind::forward_inference : prop_kind::forward_training;
+    
+  auto fwd_desc = mkldnn::conv_fwd::desc(/* all the setting above*/);
+  shared_ptr<mkldnn::conv_fwd::primitive_desc> fwd_primitive_desc(new mkldnn::conv_fwd::primitive_desc(fwd_desc, ctx.getEngine()));
+
+  return fwd_primitive_desc;
+  }
+```
+
+### MKLDNNDeviceContext
+`MKLDNNDeviceContext`, which is very straightforward, should contain some base information like: `stream`, `engine` and the map needed.
+
+
+### mkldnn_helper
+Some functions would be put in `paddle/platform/mkldnn_helper.h`.
+- create MKLDNN memories
+- create MKLDNN primitives
+- error check function
+- etc
+
+
+### Kernel Switch
+We should `reorder` the different Layout from other device or to other device. `GetExpectedKernelType` and `trans` functions can help us to implement it.
+
+`GetExpectedKernelType` should get the context, and this operator can return the best `KernelType`. 
+`trans` would be like this:
+
+```c++
+void trans(inputs, ctx) override {
+  if (NoNeedTrans()) {
+    return;
+  }
+  // find reorder primitive by op_key from context
+  auto& dev_ctx = ctx.template device_context<platform::MKLDNNDeviceContext>();
+  auto& p = dev_ctx.findPrimitive(op_key + "_reorder_input");
+  auto& i = dev_ctx.findMemory(op_key + "_src_input");
+
+  if (p == nullptr || i == nullptr || changeSized(i, input)) {
+    auto prim = createPrimitiveDesc(ctx);
+    auto src = createMemory(memoryDesc(input->dims(), actual_layout), input->data);
+    auto newbuffer = paddle::memory::Alloc(ctx.GetPlace(), input->size_in_bytes());
+    auto dst = createMemory(p->expected_desc(), newbuffer->data);
+    auto reorder_primitive(new mkldnn::reorder(src, dst));
+
+    dev_ctx.addMemory(op_key+"_src_input", src);
+    dev_ctx.addMemory(op_key+"_input", dst);
+    dev_ctx.addPrimitive(op_key+"_reorder_input", reorder_primitive);
+  }
+
+  p = dev_ctx.findPrimitive(op_key + "_reorder_input");
+  PADDLE_ENFORCE(p, "Should have Reorder Primitive");
+  dev_ctx.submit(p);
+  if (! this->isMKLDNNKernel()) {
+    // execute immediately only if this is not mkldnn kernel function.
+    // otherwise, it can be executed with the operator primitive in Compute
+    dev_ctx.stream();
+  }
+  // after submit, the input tensor in ExecutionContext should be changed as the converted one
+  // there should be another mechanism to ensure this
+}
+```
+
+### Unit Test
+All the functions should be tested corresponding.
+TBD

doc/design/operator_kernel_type.md

+# Design Doc: The Keys of Operator Kernel Type
+## Problem
+An operator can have different kernel implementations, and each operator will have a map to store the related kernels. Fluid uses `OpKernelType` as a key to identify a unique Kernel. Before an operator runs, an certain kernel must be chosen by a key of `OpKernelType`. Currently, `OpKernelType` is defined as follows:
+
+```cpp
+struct OpKernelType {
+  platform::Place place_;
+  proto::DataType data_type_;
+};
+```
+For more details, please refer to [codes](https://github.com/PaddlePaddle/Paddle/blob/2d5ec16bc8a09fb8e0f62c89b116b0cd1d333907/paddle/framework/operator.h#L348-L374) in github.
+
+It contains two keys, `Place` and `DataType`. And these two keys will be hashed to a unique key to represent a certain type of kernel. However, these two keys are not enough. We need a more complete representation of `OpKernelType`. 
+
+We often implement a kernel of an operator with some computing library in certain device(place). Please remind that computing library and device are not one-to-one corresponding. A device can have a lot of computing libraries and a computing library can also support several devices. 
+
+For example, Eigen library can support Nvidia GPU/AMD GPU/CPU. And MKLDNN library can support Intel CPU/Intel FPGA. Both `Place` and `Library` should be a key of `OpKernelType`.
+
+It's obvious that different DataTypes, like fp64/fp32/int8 will have different kernels. But the data layout of a Tensor will also lead to different implementation. Please refer to the batch norm operator [kernels](https://github.com/PaddlePaddle/Paddle/blob/a948fac4d0ad7e0412d373b8aabeb711c2899563/paddle/operators/batch_norm_op.cc#L180-L209). Data Layout should also be taken into consideration.
+
+## Solution
+
+There are four keys to determine a kernel type of an operator: `Place`/`Library`/`DataType`/`Layout`.
+
+```cpp
+struct OpKernelType {
+  platform::Place place_;
+  platform::Library library_;
+  proto::DataType data_type_;
+  framework::Layout layout_;
+};
+```
+
+Following is the details:
+
+### Place
+
+`Place` is defined as follows:
+
+```cpp
+typedef boost::variant<CUDAPlace, ROCmPlace, FPGAPlace, CPUPlace> Place;
+```
+
+`Place` is to represent the device memory where data is locating.
+
+
+### Library
+
+One operator kernel is usually implemented based on one library. `Library` is defined as a enum variable:
+
+```cpp
+enum Library { Plain, MKLDNN, CUDNN };
+```
+
+We use `Plain` enumerator to represent default library. Since most operators in Fluid are implemented based on `Eigen` library, we take `Eigen` library as the `Plain` enumerator.
+A library usually has a corresponding `DeviceContext` which contains some handles needed by computation. Fluid now have two default DeviceContexts in CPU and CUDA, `CPUDeviceContext` and `CUDADeviceContext`. `CPUDeviceContext` contains a Eigen library handle and `CDUADeviceContext` contains a Eigen library handle and cuBLAS handle.
+
+If we want to support new Library, a new enumerator need to be added to `Library` and a new corresponding `LibraryDeviceContext` will be created.
+
+
+### DataType
+
+
+`DataType` is defined in [framework.proto](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/framework/framework.proto). Currently, int32/int64/fp32/fp64 are supported.
+
+### Layout
+
+Actually, a Tensor is a view of a block of memory. Besides a pointer to the memory, we also have to get some other descriptions of this block of memory, such as shape(ddim), stride, and layout.
+
+Different layout leads to different implementation of operator kernel. There are mainly 4 principles we have to follow to support layout in our fluid framework.
+
+- We take layout as a data member of Tensor. Layout is actually a enum variable. If fluid is built with MKLDNN, then, the memory format in MKLDNN will be added into this enum variable too.
+
+- Users have to set layout for input data. And some operators like fill_constant/random, also have to set layout of generating data. Of course, we can have some default layout, like NCHW.
+
+- The inference of Layout is at run-time, not compile-time.
+
+- Every operator have to implement different kernels for different layouts. Let's take MKLDNN as an example, if we want to implement a MKLDNN convolution operator, we have to realize all the kernels for different layout, list at [here](http://01org.github.io/mkl-dnn/structmkldnn_1_1memory.html). And we will have a special macro to do registering kernels for MKLDNN operators.
+
+`Layout` is also defined as a enum variable:
+
+```cpp
+enum Layout {
+  kNCHW,
+  kNHWC,
+#ifdef PADDLE_WITH_MKLDNN
+  knChw8c
+  ...
+#endif
+};
+```

doc/design/optimizer.md

@@ -79,7 +79,7 @@ class Optimizer(object):
     def minimize(self, loss, parameter_list):
         """Add operations to minimize `loss` by updating `parameter_list`.
 
-        This method combines interface `append_backward_ops()` and
+        This method combines interface `append_backward()` and
         `create_optimization_pass()` into one.
         """
         params_grads = self.create_backward_pass(loss, parameter_list)

doc/design/refactor/multi_cpu.md

+# Design Doc: Execute the Program with Multi CPU
+
+## Abstract
+
+This Design Doc propose an approach to make the user-defined Op graph
+running with multi-CPU, we will use an auto transpiler to convert the user-defined
+Op graph to a multi-CPU Op graph, and run `ParallelDo` Op to run the graph.
+
+## Transpiler
+
+<img src="src/multi-threads/single-thread@3x.png" width="300">
+
+After converted:
+
+<img src="src/multi-threads/multi-threads@3x.png" width="1000">
+
+## Implement
+
+- `Multi-CPU Transpiler` will convert the graph to a multi-CPU graph
+  which would be executed with multi-threads.
+- `BlockingCounter` will `Init/Decrement` an atomic counter, and Blocking `Wait`
+  for the atomic counter become `0`:
+  ```cpp
+  BlockingCounter bc(thread_count);
+  for (int i = 0; i < thread_count; ++i) {
+    thread_pool->Start([&bc] {bc.DecrementCount(); })
+  }
+  bc.Wait();
+  ```
+- `ParallelDo` Operator
+  - Initialize a thread pool which is a Singleton.
+  - Use a block id as the input, and create run the specify Block on independent scope
+    with multi-threads.
+  - Initialize a `BlockingCounter` instance and wait until all threads are done.
+- `Split` Operator will split the Input Tensor into a TensorArray.
+- `Merge` merge all the gradients which calculated in different threads
+  with `mean/sum/max/min...` method, and then run the Optimizer Op to optimize `W`.
+
+## TODO
+
+- Improve the optimizer stage with multi-threads, since we could
+  assign the parameters to the different threads and execute
+  optimizer with multi-threads.

doc/design/support_new_device.md

@@ -25,13 +25,14 @@ There are mainly three parts that we have to consider while integrating a new de
 
 ### Place and DeviceContext
 
+Please remind that device and computing library are not one-to-one corresponding. A device can have a lot of computing libraries and a computing library can also support several devices.
 
 #### Place
-Fluid uses class [Place](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/place.h#L55) to represent different devices and computing libraries. There are inheritance relationships between different kinds of `Place`.
+Fluid uses class [Place](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/place.h#L55) to represent the device memory where data is located. If we add another device, we have to add corresponding `DevicePlace`.
 
 ```
-        |   CPUPlace   --> MKLDNNPlace
-Place --|   CUDAPlace  --> CUDNNPlace
+        |   CPUPlace
+Place --|   CUDAPlace
         |   FPGAPlace
 ```
 
@@ -43,7 +44,7 @@ typedef boost::variant<CUDAPlace, CPUPlace, FPGAPlace> Place;
 
 #### DeviceContext
 
-Fluid uses class [DeviceContext](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/device_context.h#L30) to manage the resources in different hardwares, such as CUDA stream in `CDUADeviceContext`. There are also inheritance relationships between different kinds of `DeviceContext`.
+Fluid uses class [DeviceContext](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/device_context.h#L30) to manage the resources in different libraries, such as CUDA stream in `CDUADeviceContext`. There are also inheritance relationships between different kinds of `DeviceContext`.
 
 
 ```
@@ -106,7 +107,7 @@ template <typename Place>
 size_t Used(Place place);
 ```
 
-To implementing these interfaces, we have to implement MemoryAllocator for different Devices
+To implement these interfaces, we have to implement MemoryAllocator for different Devices.
 
 
 #### Tensor
@@ -243,6 +244,7 @@ REGISTER_OP_CUDA_KERNEL(
 Generally, we will impelement OpKernel for all Device/Library of an Operator. We can easily train a Convolutional Neural Network in GPU. However, some OpKernel is not sutibale on a specific Device. For example, crf operator can only run on CPU, whereas most other operators can run at GPU. To achieve high performance in such circumstance, we have to switch between different Device/Library.
 
 
-We will discuss how to implement an efficient OpKernel switch policy. 
+For more details, please refer to following docs:
 
-- TBD
+- operator kernel type [doc](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/operator_kernel_type.md)
+- switch kernel [doc](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/switch_kernel.md)

doc/design/switch_kernel.md

+## Background
+Every operator has many kernels because there are multiple data types, places, data layout that Fluid supports. We use the `KernelType` to describe kernel types that operators can hold. 
+
+The `KernelType` is as follows.
+
+```
+struct KernelType {
+  Place place_;
+  DataType data_type_;
+  LayoutType layout_;
+};
+```
+
+The `place_` is a descriptor of the device and the computational library, e.g., `MKLDNNPlace`, `CUDAPlace`.
+
+The `data_type_` is the data type that this kernel performs on, e.g., `FP32`, `INT64`. Note that one kernel may have inputs with different data types. However, it will be a major `data_type`. For example, the `cross_entropy` takes `int64` as it label, and `double`/`float` as its input logit and output cost. The major `data_type` of `cross_entropy` is `float`/`double`.
+
+The `layout` is useful for some computational library. One example is that MKLDNN uses many kinds of layout, such as `nChw8c`. Each kind of layout will invoke the different kernel.
+
+## Problem
+
+We register a kernel for every operator and every kernel type ideally. However, it is impracticable for the following situations.
+
+1. Some operators, like CRF, are complicated and inefficient to be implemented on GPU. The CRF operator will only have a CPU kernel.
+2. Some operators will take too many memory. It is better to force them into CPU. However, the rest of operators in this neural network will be performed on GPU, i.e., model parallel problem.
+3. Some layout and place are particular. One example is that MKLDNN uses `nChw8` and there is no other library uses `nChw8c`.
+
+Problems under these situations are similar. We can formalise this problem as follow.
+
+We register kernels with types $KT = \{kt_1, kt_2, kt_3, ...\}$ for one operator. The inputs of this operator should be run on kernel type $kt_{?}$, which the $kt_{?} \notin KT$. How to cast the input of this operator from $kt_{?}$ to any of kernel type in $KT$.
+
+## Solution
+
+It is clearly that transforming inputs of an operator toadapt another kernel type is not related to the particular operator. So we should register these transformation methods as global methods.
+
+We can infer a kernel type from the inputs of an operators. We let this kernel type as `actual kernel type`, which means this kernel type is the actually kernel type that operator should be performed.
+
+We can get a kernel type by 1) The configuration of operator description. (Users may want to force use `MKL` for `conv` operator). 2) The place of the current executor. (Executor is running on GPU). This kernel type is what we expect the operator will be performed on. We let this kernel type as `expect kernel type`.
+
+We transform the input data from `actual` to `expect` if the expect kernel type is not as same as actual kernel type.
+
+The algorithm is described as follow
+
+```cpp
+using DataTransformationFN = std::function<void(const Tensor& in, Tensor* out)>;
+using KernelTypePair = std::pair<KernelType, KernelType>;
+
+map<KernelTypePair, DataTransformationFN> g_data_transformation_;
+
+void OpWithKernel::Run() {
+  vec<Tensor> inputs = ...
+  auto actual_kernel_type = GetActualKernelType(inputs);
+  
+  // The expected kernel type is related to actual kernel type.
+  // For the most operators, the expected kernel type is as same as
+  // actual kernel type.
+  //
+  // So we pass `actual_kernel_type` as a parameter of 
+  // GetExpectedKernelType
+  auto expect_kernel_type = GetExpectedKernelType(actual_kernel_type);
+  
+  auto trans = g_data_transformation_[{actual_kernel_type, expect_kernel_type}];
+  
+  kernel.run(trans(inputs));
+}
+```

doc/faq/build_and_install/index_cn.rst

@@ -109,3 +109,31 @@ PaddlePaddle使用avx SIMD指令提高cpu执行效率，因此错误的使用二
 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。
+
+8. 下载MKLML库失败
+------------------
+
+..  code-block:: bash
+
+    make[2]: *** [third_party/mklml/src/extern_mklml-stamp/extern_mklml-download] 错误 4
+    make[1]: *** [CMakeFiles/extern_mklml.dir/all] 错误 2
+    make[1]: *** 正在等待未完成的任务....
+
+原因：网速或SSL链接原因，导致MKLML库下载不成功。
+
+解决办法是：手动下载并安装，具体步骤如下。
+
+..  code-block:: bash
+
+    // 1. 进入对应的目录
+    cd build/third_party/mklml/src/extern_mklml
+
+    // 2. 查看包的大小， 正常情况下是75M，如果小于75M，即下载失败：
+    du -sh mklml_lnx_2018.0.1.20171007.tgz
+
+    // 3. 手动下载且解压缩，并手动生成download成功标签：
+    wget --no-check-certificate https://github.com/01org/mkl-dnn/releases/download/v0.11/mklml_lnx_2018.0.1.20171007.tgz -c -O mklml_lnx_2018.0.1.20171007.tgz 
+    tar zxf mklml_lnx_2018.0.1.20171007.tgz
+    touch ../extern_mklml-stamp/extern_mklml-download
+
+    // 4. 接着编译即可

doc/getstarted/build_and_install/build_from_source_cn.rst

@@ -70,13 +70,13 @@ PaddlePaddle编译需要使用到下面的依赖（包含但不限于），其
    :header: "依赖", "版本", "说明"
    :widths: 10, 15, 30
 
-   "CMake", ">=3.5", ""
+   "CMake", ">=3.2", ""
    "GCC", "4.8.2", "推荐使用CentOS的devtools2"
-   "Python", "2.7.x", "依赖libpython2.7.so"
-   "pip", ">=9.0", ""
-   "numpy", "", ""
+   "Python", "2.7.x", "依赖libpython2.7.so"
+   "pip", ">=9.0", ""
+   "numpy", "", ""
    "SWIG", ">=2.0", ""
-   "Go", ">=1.8", "可选"
+   "Go", ">=1.8", "可选"
 
 
 .. _build_options:

doc/getstarted/build_and_install/build_from_source_en.rst

@@ -76,13 +76,13 @@ will be downloaded automatically.
    :header: "Dependency", "Version", "Description"
    :widths: 10, 15, 30
 
-   "CMake", ">=3.5", ""
+   "CMake", ">=3.2", ""
    "GCC", "4.8.2", "Recommend devtools2 for CentOS"
-   "Python", "2.7.x", "Need libpython2.7.so"
-   "pip", ">=9.0", ""
-   "numpy", "", ""
+   "Python", "2.7.x", "Need libpython2.7.so"
+   "pip", ">=9.0", ""
+   "numpy", "", ""
    "SWIG", ">=2.0", ""
-   "Go", ">=1.8", "Optional"
+   "Go", ">=1.8", "Optional"
 
 
 .. _build_options:

doc/getstarted/build_and_install/docker_install_cn.rst

@@ -128,7 +128,7 @@ PaddlePaddle Book是为用户和开发者制作的一个交互式的Jupyter Note
 
 AVX是一种CPU指令集，可以加速PaddlePaddle的计算。最新的PaddlePaddle Docker镜像默认
 是开启AVX编译的，所以，如果您的电脑不支持AVX，需要单独
-`编译 <./build_from_source_cn.rst>`_ PaddlePaddle为no-avx版本。
+`编译 <./build_from_source_cn.html>`_ PaddlePaddle为no-avx版本。
 
 以下指令能检查Linux电脑是否支持AVX：
 

doc/getstarted/build_and_install/docker_install_en.rst

@@ -137,7 +137,7 @@ GPU driver installed before move on.
 AVX is a kind of CPU instruction can accelerate PaddlePaddle's calculations.
 The latest PaddlePaddle Docker image turns AVX on by default, so, if your
 computer doesn't support AVX, you'll probably need to
-`build <./build_from_source_en.rst>`_ with :code:`WITH_AVX=OFF`.
+`build <./build_from_source_en.html>`_ with :code:`WITH_AVX=OFF`.
 
 The following command will tell you whether your computer supports AVX.
 

doc/getstarted/build_and_install/pip_install_cn.rst

@@ -37,11 +37,11 @@ PaddlePaddle可以使用常用的Python包管理工具
     :header: "版本说明", "cp27-cp27mu", "cp27-cp27m", "C-API"
     :widths: 1, 3, 3, 3
 
-    "cpu_avx_mkl", "`paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddle.tgz>`_"
-    "cpu_avx_openblas", "`paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxOpenblas/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxOpenblas/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl>`_", "暂无"
-    "cuda7.5_cudnn5_avx_mkl", "`paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
-    "cuda8.0_cudnn5_avx_mkl", "`paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
-    "cuda8.0_cudnn7_avx_mkl", "`paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
+    "cpu_avx_mkl", "`paddlepaddle-0.11.0-cp27-cp27mu-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddlepaddle-0.11.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.11.0-cp27-cp27m-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddlepaddle-0.11.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddle.tgz>`_"
+    "cpu_avx_openblas", "`paddlepaddle-0.11.0-cp27-cp27mu-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxOpenblas/.lastSuccessful/paddlepaddle-0.11.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.11.0-cp27-cp27m-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxOpenblas/.lastSuccessful/paddlepaddle-0.11.0-cp27-cp27m-linux_x86_64.whl>`_", "暂无"
+    "cuda7.5_cudnn5_avx_mkl", "`paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
+    "cuda8.0_cudnn5_avx_mkl", "`paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
+    "cuda8.0_cudnn7_avx_mkl", "`paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
 
 .. _pip_dependency:
 

doc/getstarted/build_and_install/pip_install_en.rst

@@ -40,11 +40,11 @@ If the links below shows up the login form, just click "Log in as guest" to star
     :header: "version", "cp27-cp27mu", "cp27-cp27m", "C-API"
     :widths: 1, 3, 3, 3
 
-    "cpu_avx_mkl", "`paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddle.tgz>`_"
-    "cpu_avx_openblas", "`paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxOpenblas/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxOpenblas/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl>`_", "Not Available"
-    "cuda7.5_cudnn5_avx_mkl", "`paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
-    "cuda8.0_cudnn5_avx_mkl", "`paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
-    "cuda8.0_cudnn7_avx_mkl", "`paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddlepaddle-0.10.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <http://guest@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
+    "cpu_avx_mkl", "`paddlepaddle-0.11.0-cp27-cp27mu-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddlepaddle-0.11.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.11.0-cp27-cp27m-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddlepaddle-0.11.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxCp27cp27mu/.lastSuccessful/paddle.tgz>`_"
+    "cpu_avx_openblas", "`paddlepaddle-0.11.0-cp27-cp27mu-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxOpenblas/.lastSuccessful/paddlepaddle-0.11.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle-0.11.0-cp27-cp27m-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_CpuAvxOpenblas/.lastSuccessful/paddlepaddle-0.11.0-cp27-cp27m-linux_x86_64.whl>`_", "Not Available"
+    "cuda7.5_cudnn5_avx_mkl", "`paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda75cudnn5cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
+    "cuda8.0_cudnn5_avx_mkl", "`paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda80cudnn5cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
+    "cuda8.0_cudnn7_avx_mkl", "`paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27mu-linux_x86_64.whl>`_", "`paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddlepaddle_gpu-0.11.0-cp27-cp27m-linux_x86_64.whl>`_", "`paddle.tgz <https://guest:@paddleci.ngrok.io/repository/download/Manylinux1_Cuda8cudnn7cp27cp27mu/.lastSuccessful/paddle.tgz>`_"
 
 .. _pip_dependency:
 

doc/getstarted/concepts/src/infer.py

+import paddle.v2 as paddle
+import numpy as np
+
+paddle.init(use_gpu=False)
+x = paddle.layer.data(name='x', type=paddle.data_type.dense_vector(2))
+y_predict = paddle.layer.fc(input=x, size=1, act=paddle.activation.Linear())
+
+# loading the model which generated by training
+with open('params_pass_90.tar', 'r') as f:
+    parameters = paddle.parameters.Parameters.from_tar(f)
+
+# Input multiple sets of data，Output the infer result in a array.
+i = [[[1, 2]], [[3, 4]], [[5, 6]]]
+print paddle.infer(output_layer=y_predict, parameters=parameters, input=i)
+# Will print:
+# [[ -3.24491572]
+#  [ -6.94668722]
+#  [-10.64845848]]

doc/getstarted/concepts/src/train.py

@@ -26,6 +26,11 @@ def event_handler(event):
         if event.batch_id % 1 == 0:
             print "Pass %d, Batch %d, Cost %f" % (event.pass_id, event.batch_id,
                                                   event.cost)
+    # product model every 10 pass
+    if isinstance(event, paddle.event.EndPass):
+        if event.pass_id % 10 == 0:
+            with open('params_pass_%d.tar' % event.pass_id, 'w') as f:
+                trainer.save_parameter_to_tar(f)
 
 
 # define training dataset reader

doc/getstarted/concepts/use_concepts_cn.rst

@@ -147,4 +147,9 @@ PaddlePaddle支持不同类型的输入数据，主要包括四种类型，和
 ..  literalinclude:: src/train.py
     :linenos:
 
+使用以上训练好的模型进行预测，取其中一个模型params_pass_90.tar，输入需要预测的向量组，然后打印输出：
+
+..  literalinclude:: src/infer.py
+    :linenos:
+
 有关线性回归的实际应用，可以参考PaddlePaddle book的 `第一章节 <http://book.paddlepaddle.org/index.html>`_。

doc/howto/dev/new_op_cn.md

@@ -53,7 +53,7 @@ Kernel实现       | CPU、CUDA共享Kernel实现在`.h`文件中，否则，CPU
 ```cpp
 class MulOpMaker : public framework::OpProtoAndCheckerMaker {
  public:
-  MulOpMaker(framework::OpProto *proto, framework::OpAttrChecker *op_checker)
+  MulOpMaker(OpProto *proto, OpAttrChecker *op_checker)
       : OpProtoAndCheckerMaker(proto, op_checker) {
     AddInput("X", "(Tensor), 2D tensor of size (M x K)");
     AddInput("Y", "(Tensor), 2D tensor of size (K x N)");
@@ -82,7 +82,7 @@ The equation is: Out = X * Y
 template <typename AttrType>
 class ScaleOpMaker : public framework::OpProtoAndCheckerMaker {
  public:
-  ScaleOpMaker(framework::OpProto *proto, framework::OpAttrChecker *op_checker)
+  ScaleOpMaker(OpProto *proto, OpAttrChecker *op_checker)
       : OpProtoAndCheckerMaker(proto, op_checker) {
     AddInput("X", "The input tensor of scale operator.").NotInGradient();
     AddOutput("Out", "The output tensor of scale operator.").NotInGradient();

doc/howto/dev/new_op_en.md

@@ -50,7 +50,7 @@ First, define `ProtoMaker` to describe the Operator's input, output, and additio
 ```cpp
 class MulOpMaker : public framework::OpProtoAndCheckerMaker {
  public:
-  MulOpMaker(framework::OpProto *proto, framework::OpAttrChecker *op_checker)
+  MulOpMaker(OpProto *proto, OpAttrChecker *op_checker)
       : OpProtoAndCheckerMaker(proto, op_checker) {
     AddInput("X", "(Tensor), 2D tensor of size (M x K)");
     AddInput("Y", "(Tensor), 2D tensor of size (K x N)");
@@ -79,7 +79,7 @@ An additional example [`ScaleOp`](https://github.com/PaddlePaddle/Paddle/blob/de
 template <typename AttrType>
 class ScaleOpMaker : public framework::OpProtoAndCheckerMaker {
  public:
-  ScaleOpMaker(framework::OpProto *proto, framework::OpAttrChecker *op_checker)
+  ScaleOpMaker(OpProto *proto, OpAttrChecker *op_checker)
       : OpProtoAndCheckerMaker(proto, op_checker) {
     AddInput("X", "The input tensor of scale operator.").NotInGradient();
     AddOutput("Out", "The output tensor of scale operator.").NotInGradient();

doc/howto/index_cn.rst

@@ -9,9 +9,6 @@
 
   usage/cmd_parameter/index_cn.rst
   usage/cluster/cluster_train_cn.md
-  usage/k8s/k8s_basis_cn.md
-  usage/k8s/k8s_cn.md
-  usage/k8s/k8s_distributed_cn.md
 
 开发标准
 --------

doc/howto/index_en.rst

@@ -9,8 +9,6 @@ Usage
 
   usage/cmd_parameter/index_en.rst
   usage/cluster/cluster_train_en.md
-  usage/k8s/k8s_en.md
-  usage/k8s/k8s_aws_en.md
 
 Development
 ------------

doc/howto/read_source.md

@@ -6,10 +6,10 @@ Core: https://github.com/PaddlePaddle/Paddle/tree/develop/paddle/framework
 
 Operator: https://github.com/PaddlePaddle/Paddle/tree/develop/paddle/operators
 
-Optimizer: https://github.com/PaddlePaddle/Paddle/tree/develop/paddle/optimizer
-
 Memory: https://github.com/PaddlePaddle/Paddle/tree/develop/paddle/memory
 
+Platform: https://github.com/PaddlePaddle/Paddle/tree/develop/paddle/platform
+
 # Compile Time
 
 The following **defines** the NN. The definition goes into this [protocol buffer](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/framework/framework.proto).

doc/howto/usage/cluster/cluster_train_cn.md

-# PaddlePaddle分布式训练
-
-* [概述](#概述)
-* [环境准备](#环境准备)
-* [启动参数说明](#启动参数说明)
-  * [启动参数服务器](#启动参数服务器)
-  * [启动计算节点](#启动计算节点)
-  * [准备数据集](#准备数据集)
-  * [准备训练程序](#准备训练程序)
-* [使用分布式计算平台或工具](#使用分布式计算平台或工具)
-  * [使用Fabric启动集群作业](#使用fabric启动集群作业)
-     * [准备一个Linux集群](#准备一个linux集群)
-     * [启动集群作业](#启动集群作业)
-     * [终止集群作业](#终止集群作业)
-     * [检查集群训练结果](#检查集群训练结果)
-     * [检查模型输出](#检查模型输出)
-  * [在OpenMPI集群中提交训练作业](#在openmpi集群中提交训练作业)
-     * [准备OpenMPI集群](#准备OpenMPI集群)
-     * [启动集群作业](#启动集群作业-1)
-  * [在Kubernetes集群中提交训练作业](#在kubernetes集群中提交训练作业)
+# 分布式训练
+
 
 ## 概述
+
 本文将介绍如何使用PaddlePaddle在不同的集群框架下完成分布式训练。分布式训练架构如下图所示：
 
 <img src="https://user-images.githubusercontent.com/13348433/31772175-5f419eca-b511-11e7-9db7-5231fe3d9ccb.png" width="500">
@@ -32,10 +15,11 @@
 
 在使用同步SGD训练神经网络时，PaddlePaddle使用同步屏障（barrier），使梯度的提交和参数的更新按照顺序方式执行。在异步SGD中，则并不会等待所有trainer提交梯度才更新参数，这样极大地提高了计算的并行性：参数服务器之间不相互依赖，并行地接收梯度和更新参数，参数服务器也不会等待计算节点全部都提交梯度之后才开始下一步，计算节点之间也不会相互依赖，并行地执行模型的训练。可以看出，虽然异步SGD方式会提高参数更新并行度, 但是并不能保证参数同步更新，在任意时间某一台参数服务器上保存的参数可能比另一台要更新，与同步SGD相比，梯度会有噪声。
 
+
 ## 环境准备
 
 1. 准备您的计算集群。计算集群通常由一组（几台到几千台规模）的Linux服务器组成。服务器之间可以通过局域网（LAN）联通，每台服务器具有集群中唯一的IP地址（或者可被DNS解析的主机名）。集群中的每台计算机通常被成为一个“节点”。
-1. 我们需要在集群的所有节点上安装 PaddlePaddle。 如果要启用GPU，还需要在节点上安装对应的GPU驱动以及CUDA。PaddlePaddle的安装可以参考[build_and_install](https://github.com/PaddlePaddle/Paddle/tree/develop/doc/getstarted/build_and_install)的多种安装方式。我们推荐使用[Docker](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/getstarted/build_and_install/docker_install_cn.rst)安装方式来快速安装PaddlePaddle。
+1. 我们需要在集群的所有节点上安装 PaddlePaddle。 如果要启用GPU，还需要在节点上安装对应的GPU驱动以及CUDA。PaddlePaddle的安装可以参考[build_and_install](http://www.paddlepaddle.org/docs/develop/documentation/zh/getstarted/build_and_install/index_cn.html)的多种安装方式。我们推荐使用[Docker](http://www.paddlepaddle.org/docs/develop/documentation/zh/getstarted/build_and_install/docker_install_cn.html)安装方式来快速安装PaddlePaddle。
 
 安装完成之后，执行下面的命令可以查看已经安装的版本（docker安装方式可以进入docker容器执行：`docker run -it paddlepaddle/paddle:[tag] /bin/bash`）：
 ```bash
@@ -63,12 +47,12 @@ $ paddle pserver --port=7164 --ports_num=1 --ports_num_for_sparse=1 --num_gradie
 $ stdbuf -oL /usr/bin/nohup paddle pserver --port=7164 --ports_num=1 --ports_num_for_sparse=1 --num_gradient_servers=1 &> pserver.log
 ```
 
-| 参数  | 是否必选 | 默认值 | 说明 |
-| ------------- | ------------- | ------------- | ------------- |
-| port  | 必选 | 7164 | pserver监听的起始端口，根据ports_num决定<br>总端口个数，从起始端口监听多个端口用于通信  |
-| ports_num  | 必选 | 1 | 监听的端口个数  |
-| ports_num_for_sparse  | 必选 | 1 | 用于稀疏类型参数通信的端口个数  |
-| num_gradient_servers  | 必选 | 1 | 当前训练任务pserver总数 |
+参数说明
+
+- port：**必选，默认7164**，pserver监听的起始端口，根据ports_num决定总端口个数，从起始端口监听多个端口用于通信
+- ports_num：**必选，默认1**，监听的端口个数
+- ports_num_for_sparse：**必选，默认1**，用于稀疏类型参数通信的端口个数
+- num_gradient_servers：**必选，默认1**，当前训练任务pserver总数
 
 ### 启动计算节点
 执行以下命令启动使用python编写的trainer程序（文件名为任意文件名，如train.py）
@@ -105,16 +89,16 @@ paddle.init(
         pservers="127.0.0.1")
 ```
 
-| 参数  | 是否必选 | 默认 | 说明 |
-| ------------- | ------------- | ------------- | ------------- |
-| use_gpu  | 可选 | False | 是否启用GPU训练 |
-| trainer_count  | 必选 | 1 | 当前训练任务trainer总个数 |
-| port  | 必选 | 7164 | 连接到pserver的端口  |
-| ports_num  | 必选 | 1 | 连接到pserver的端口个数  |
-| ports_num_for_sparse  | 必选 | 1 | 和pserver之间用于稀疏类型参数通信的端口个数  |
-| num_gradient_servers  | 必选 | 1 | 当前训练任务pserver总数 |
-| trainer_id  | 必选 | 0 | 每个trainer的唯一ID，从0开始的整数 |
-| pservers  | 必选 | 127.0.0.1 | 当前训练任务启动的pserver的IP列表，多个IP使用“,”隔开 |
+参数说明
+
+- use_gpu： **可选，默认False**，是否启用GPU训练
+- trainer_count：**必选，默认1**，当前训练任务trainer总个数
+- port：**必选，默认7164**，连接到pserver的端口
+- ports_num：**必选，默认1**，连接到pserver的端口个数
+- ports_num_for_sparse：**必选，默认1**，和pserver之间用于稀疏类型参数通信的端口个数
+- num_gradient_servers：**必选，默认1**，当前训练任务pserver总数
+- trainer_id：**必选，默认0**，每个trainer的唯一ID，从0开始的整数
+- pservers：**必选，默认127.0.0.1**，当前训练任务启动的pserver的IP列表，多个IP使用“,”隔开
 
 
 ### 准备数据集
@@ -171,7 +155,7 @@ test.txt-00002
 
 - `my_lib.py`：会被`train.py`调用的一些用户定义的库函数，比如PIL库等。
 - `word_dict.pickle`：在`train.py`中会使用到的字典数据文件。
-- `train.py`：训练程序，代码参考[api_train_v2_cluster.py](https://github.com/PaddlePaddle/Paddle/tree/develop/doc/howto/usage/cluster/src/word2vec/prepare.py)。***注意：*** 对于本样例代码，在使用不同的分布式计算平台时，您可能需要修改`train.py`开头的部分（如下），以便获得训练数据的位置和获取环境变量配置：
+- `train.py`：训练程序，代码参考[api_train_v2_cluster.py](https://github.com/PaddlePaddle/Paddle/tree/develop/doc/howto/usage/cluster/src/word2vec/api_train_v2_cluster.py)。***注意：*** 对于本样例代码，在使用不同的分布式计算平台时，您可能需要修改`train.py`开头的部分（如下），以便获得训练数据的位置和获取环境变量配置：
 
   ```python
   cluster_train_file = "./train_data_dir/train/train.txt"
@@ -195,91 +179,10 @@ PaddlePaddle可以使用多种分布式计算平台构建分布式计算任务
 
 在使用分布式计算平台进行训练时，任务被调度在集群中时，分布式计算平台通常会通过API或者环境变量提供任务运行需要的参数，比如节点的ID、IP和任务节点个数等。
 
-### 使用Fabric启动集群作业
-
-#### 准备一个Linux集群
-可以在`paddle/scripts/cluster_train_v2/fabric/docker_cluster`目录下，执行`kubectl -f ssh_servers.yaml`启动一个测试集群，并使用`kubectl get po -o wide`获得这些节点的IP地址。
-
-#### 启动集群作业
-
-`paddle.py` 提供了自动化脚本来启动不同节点中的所有 PaddlePaddle 集群进程。默认情况下，所有命令行选项可以设置为 `paddle.py` 命令选项并且 `paddle.py` 将透明、自动地将这些选项应用到 PaddlePaddle 底层进程。
-
-`paddle.py` 为方便作业启动提供了两个独特的命令选项。
-
--  `job_dispatch_package`  设为本地 `workspace` 目录，它将被分发到 `conf.py` 中设置的所有节点。它有助于帮助频繁修改和访问工作区文件的用户减少负担，否则频繁的多节点工作空间部署可能会很麻烦。
--  `job_workspace`  设为已部署的工作空间目录，`paddle.py` 将跳过分发阶段直接启动所有节点的集群作业。它可以帮助减少分发延迟。
-
-`cluster_train/run.sh` 提供了命令样例来运行 `doc/howto/usage/cluster/src/word2vec` 集群任务，只需用您定义的目录修改 `job_dispatch_package` 和 `job_workspace`，然后：
-```
-sh run.sh
-```
-
-集群作业将会在几秒后启动。
-
-#### 终止集群作业
-`paddle.py`能获取`Ctrl + C` SIGINT 信号来自动终止它启动的所有进程。只需中断 `paddle.py` 任务来终止集群作业。如果程序崩溃你也可以手动终止。
-
-#### 检查集群训练结果
-详细信息请检查 $workspace/log 里的日志，每一个节点都有相同的日志结构。
-
-`paddle_trainer.INFO`
-提供几乎所有训练的内部输出日志，与本地训练相同。这里检验运行时间模型的收敛。
-
-`paddle_pserver2.INFO`
-提供 pserver 运行日志，有助于诊断分布式错误。
-
-`server.log`
-提供 parameter server 进程的 stderr 和 stdout。训练失败时可以检查错误日志。
-
-`train.log`
-提供训练过程的 stderr 和 stdout。训练失败时可以检查错误日志。
-
-#### 检查模型输出
-运行完成后，模型文件将被写入节点 0 的 `output` 目录中。
-工作空间中的 `nodefile` 表示当前集群作业的节点 ID。
-
-### 在OpenMPI集群中提交训练作业
-
-#### 准备OpenMPI集群
-
-执行下面的命令以启动3个节点的OpenMPI集群和一个"head"节点：
-
-```bash
-paddle/scripts/cluster_train_v2/openmpi/docker_cluster
-kubectl create -f head.yaml
-kubectl create -f mpi-nodes.yaml
-```
-
-然后可以从head节点ssh无密码登录到OpenMPI的每个节点上。
-
-#### 启动集群作业
-
-您可以按照下面的步骤在OpenMPI集群中提交paddle训练任务：
-
-```bash
-# 获得head和node节点的IP地址
-kubectl get po -o wide
-# 将node节点的IP地址保存到machines文件中
-kubectl get po -o wide | grep nodes | awk '{print $6}' > machines
-# 拷贝必要的文件到head节点
-scp -i ssh/id_rsa.mpi.pub machines prepare.py train.py start_mpi_train.sh tutorial@[headIP]:~
-# ssh 登录到head节点
-ssh -i ssh/id_rsa.mpi.pub tutorial@[headIP]
-# --------------- 以下操作均在head节点中执行 ---------------
-# 准备训练数据
-python prepare.py
-# 拷贝训练程序和字典文件到每台MPI节点
-cat machines | xargs -i scp word_dict.pickle train.py start_mpi_train.sh machines {}:/home/tutorial
-# 创建日志目录
-mpirun -hostfile machines -n 3 mkdir /home/tutorial/logs
-# 拷贝训练数据到各自的节点
-scp train.txt-00000 test.txt-00000 [node1IP]:/home/tutorial
-scp train.txt-00001 test.txt-00001 [node2IP]:/home/tutorial
-scp train.txt-00002 test.txt-00002 [node3IP]:/home/tutorial
-# 启动训练任务
-mpirun -hostfile machines -n 3  /home/tutorial/start_mpi_train.sh
-```
-
-### 在Kubernetes集群中提交训练作业
+## 在不同集群中运行
 
-此部分的使用方法可以参考[here](../k8s/k8s_distributed_cn.md)。
+  - [fabric集群](fabric_cn.md)
+  - [openmpi集群](openmpi_cn.md)
+  - [kubernetes单机](k8s_cn.md)
+  - [kubernetes distributed分布式](k8s_distributed_cn.md)
+  - [AWS上运行kubernetes集群训练](k8s_aws_cn.md)

doc/howto/usage/cluster/cluster_train_en.md

-# PaddlePaddle Distributed Training
-
-* [Introduction](#introduction)
-* [Preparations](#preparations)
-* [Command-line arguments](#command-line-arguments)
-   * [Starting parameter server](#starting-parameter-server)
-   * [Starting trainer](#starting-trainer)
-   * [Prepare Training Dataset](#prepare-training-dataset)
-   * [Prepare Training program](#prepare-training-program)
-* [Use cluster platforms or cluster management tools](#use-cluster-platforms-or-cluster-management-tools)
-   * [Cluster Training Using Fabric](#cluster-training-using-fabric)
-      * [Prepare a Linux cluster](#prepare-a-linux-cluster)
-      * [Launching Cluster Job](#launching-cluster-job)
-      * [Kill Cluster Job](#kill-cluster-job)
-      * [Check Cluster Training Result](#check-cluster-training-result)
-      * [Check Model Output](#check-model-output)
-   * [Cluster Training Using OpenMPI](#cluster-training-using-openmpi)
-      * [Prepare an OpenMPI cluster](#prepare-an-openmpi-cluster)
-      * [Launching Cluster Job](#launching-cluster-job-1)
-   * [Cluster Training Using Kubernetes](#cluster-training-using-kubernetes)
+# Distributed Training
 
 ## Introduction
 
@@ -35,7 +16,7 @@ When training with synchronize SGD, PaddlePaddle uses an internal "synchronize b
 
 ## Preparations
 1. Prepare your computer cluster. It's normally a bunch of Linux servers connected by LAN. Each server will be assigned a unique IP address. The computers in the cluster can be called "nodes".
-2. Install PaddlePaddle on every node. If you are going to take advantage of GPU cards, you'll also need to install proper driver and CUDA libraries. To install PaddlePaddle please read [this build and install](https://github.com/PaddlePaddle/Paddle/tree/develop/doc/getstarted/build_and_install) document. We strongly recommend using [Docker installation](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/getstarted/build_and_install/docker_install_en.rst).
+2. Install PaddlePaddle on every node. If you are going to take advantage of GPU cards, you'll also need to install proper driver and CUDA libraries. To install PaddlePaddle please read [this build and install](http://www.paddlepaddle.org/docs/develop/documentation/en/getstarted/build_and_install/index_en.html) document. We strongly recommend using [Docker installation](http://www.paddlepaddle.org/docs/develop/documentation/en/getstarted/build_and_install/docker_install_en.html).
 
 After installation, you can check the version by typing the below command (run a docker container  if using docker: `docker run -it paddlepaddle/paddle:[tag] /bin/bash`):
 
@@ -67,12 +48,12 @@ If you wish to run parameter servers in background, and save a log file, you can
 $ stdbuf -oL /usr/bin/nohup paddle pserver --port=7164 --ports_num=1 --ports_num_for_sparse=1 --num_gradient_servers=1 &> pserver.log
 ```
 
-| param  | required | default | description |
-| ------------- | ------------- | ------------- | ------------- |
-| port  | required | 7164 | port which parameter server will listen on. If ports_num greater than 1, parameter server will listen on multiple ports for more network throughput |
-| ports_num  | required | 1 | total number of ports will listen on  |
-| ports_num_for_sparse  | required | 1 | number of ports which serves sparse parameter update  |
-| num_gradient_servers  | required | 1 | total number of gradient servers |
+Parameter Description
+
+- port: **required, default 7164**, port which parameter server will listen on. If ports_num greater than 1, parameter server will listen on multiple ports for more network throughput.
+- ports_num: **required, default 1**, total number of ports will listen on.
+- ports_num_for_sparse: **required, default 1**, number of ports which serves sparse parameter update.
+- num_gradient_servers: **required, default 1**, total number of gradient servers.
 
 ### Starting trainer
 Type the command below to start the trainer(name the file whatever you want, like "train.py")
@@ -111,16 +92,16 @@ paddle.init(
         pservers="127.0.0.1")
 ```
 
-| param  | required | default | description |
-| ------------- | ------------- | ------------- | ------------- |
-| use_gpu  | optional | False | set to "True" to enable GPU training |
-| trainer_count  | required | 1 | total count of trainers in the training job |
-| port  | required | 7164 | port to connect to parameter server  |
-| ports_num  | required | 1 | number of ports for communication |
-| ports_num_for_sparse  | required | 1 | number of ports for sparse type caculation |
-| num_gradient_servers  | required | 1 | total number of gradient server |
-| trainer_id  | required | 0 | ID for every trainer, start from 0 |
-| pservers  | required | 127.0.0.1 | list of IPs of parameter servers, separated by "," |
+Parameter Description
+
+- use_gpu: **optional, default False**, set to "True" to enable GPU training.
+- trainer_count: **required, default 1**, total count of trainers in the training job.
+- port: **required, default 7164**, port to connect to parameter server.
+- ports_num: **required, default 1**, number of ports for communication.
+- ports_num_for_sparse: **required, default 1**, number of ports for sparse type caculation.
+- num_gradient_servers: **required, default 1**, total number of gradient server.
+- trainer_id: **required, default 0**, ID for every trainer, start from 0.
+- pservers: **required, default 127.0.0.1**, list of IPs of parameter servers, separated by ",".
 
 ### Prepare Training Dataset
 
@@ -178,7 +159,7 @@ Your workspace may looks like:
 
 - `my_lib.py`: user defined libraries, like PIL libs. This is optional.
 - `word_dict.pickle`: dict file for training word embeding.
-- `train.py`: training program. Sample code: [api_train_v2_cluster.py](https://github.com/PaddlePaddle/Paddle/tree/develop/doc/howto/usage/cluster/src/word2vec/prepare.py). ***NOTE:*** You may need to modify the head part of `train.py` when using different cluster platform to retrive configuration environment variables:
+- `train.py`: training program. Sample code: [api_train_v2_cluster.py](https://github.com/PaddlePaddle/Paddle/tree/develop/doc/howto/usage/cluster/src/word2vec/api_train_v2_cluster.py). ***NOTE:*** You may need to modify the head part of `train.py` when using different cluster platform to retrive configuration environment variables:
 
   ```python
   cluster_train_file = "./train_data_dir/train/train.txt"
@@ -202,92 +183,9 @@ We'll introduce cluster job management on these platforms. The examples can be f
 
 These cluster platforms provide API or environment variables for training processes, when the job is dispatched to different nodes. Like node ID, IP or total number of nodes etc.
 
-### Cluster Training Using Fabric
-
-#### Prepare a Linux cluster
-
-Run `kubectl -f ssh_servers.yaml` under the directory:  `paddle/scripts/cluster_train_v2/fabric/docker_cluster` will launch a demo cluster. Run `kubectl get po -o wide` to get IP addresses of these nodes.
-
-#### Launching Cluster Job
-`paddle.py` provides automatical scripts to start all PaddlePaddle cluster processes in different nodes. By default, all command line options can be set as `paddle.py` command options and `paddle.py` will transparently and automatically set these options to PaddlePaddle lower level processes.
-
-`paddle.py`provides two distinguished command option for easy job launching.
-
-- `job_dispatch_package` set it with local `workspace` directory, it will be dispatched to all nodes which is set in `conf.py`. It could be helpful for frequently manipulating workspace files. otherwise, frequent multi-nodes workspace deployment is very annoying.
-- `job_workspace`  set it with already deployed workspace directory, `paddle.py` will skip dispatch stage to directly launch cluster job with all nodes. It could help to reduce heavy
-dispatch latency.
-
-`cluster_train/run.sh` provides command line sample to run `demo/recommendation` cluster job, just modify `job_dispatch_package` and `job_workspace` with your defined directory, then:
-```
-sh run.sh
-```
-
-The cluster Job will start in several seconds.
-
-#### Kill Cluster Job
-`paddle.py` can capture `Ctrl + C` SIGINT signal to automatically kill all processes launched by it. So just stop `paddle.py` to kill cluster job. You should manually kill the job if the program crashed.
-
-#### Check Cluster Training Result
-Check log in $workspace/log for details, each node owns same log structure.
-
-`paddle_trainer.INFO`
-It provides almost all internal output log for training,  same as local training. Check runtime model convergence here.
-
-`paddle_pserver2.INFO`
-It provides parameter server running log, which could help to diagnose distributed error.
-
-`server.log`
-It provides stderr and stdout of parameter server process. Check error log if training crashes.
-
-`train.log`
-It provides stderr and stdout of trainer process. Check error log if training crashes.
-
-#### Check Model Output
-After one pass finished, model files will be written in `output` directory in node 0.
-`nodefile` in workspace indicates the node id of current cluster job.
-
-### Cluster Training Using OpenMPI
-
-#### Prepare an OpenMPI cluster
-
-Run the following command to start a 3-node MPI cluster and one "head" node.
-
-```bash
-cd paddle/scripts/cluster_train_v2/openmpi/docker_cluster
-kubectl create -f head.yaml
-kubectl create -f mpi-nodes.yaml
-```
-
-Then you can log in to every OpenMPI node using ssh without input any passwords.
-
-#### Launching Cluster Job
-
-Follow the steps to launch a PaddlePaddle training job in OpenMPI cluster:\
-
-```bash
-# find out node IP addresses
-kubectl get po -o wide
-# generate a "machines" file containing node IP addresses
-kubectl get po -o wide | grep nodes | awk '{print $6}' > machines
-# copy necessary files onto "head" node
-scp -i ssh/id_rsa.mpi.pub machines prepare.py train.py start_mpi_train.sh tutorial@[headIP]:~
-# login to head node using ssh
-ssh -i ssh/id_rsa.mpi.pub tutorial@[headIP]
-# --------------- in head node ---------------
-# prepare training data
-python prepare.py
-# copy training data and dict file to MPI nodes
-cat machines | xargs -i scp word_dict.pickle train.py start_mpi_train.sh machines {}:/home/tutorial
-# creat a directory for storing log files
-mpirun -hostfile machines -n 3 mkdir /home/tutorial/logs
-# copy training data to every node
-scp train.txt-00000 test.txt-00000 [node1IP]:/home/tutorial
-scp train.txt-00001 test.txt-00001 [node2IP]:/home/tutorial
-scp train.txt-00002 test.txt-00002 [node3IP]:/home/tutorial
-# start the job
-mpirun -hostfile machines -n 3  /home/tutorial/start_mpi_train.sh
-```
-
-### Cluster Training Using Kubernetes
+## Use different clusters
 
-The details can be found [here](../k8s/k8s_cn.md)
+  - [fabric](fabric_en.md)
+  - [openmpi](openmpi_en.md)
+  - [kubernetes](k8s_en.md)
+  - [kubernetes on AWS](k8s_aws_en.md)

doc/howto/usage/cluster/fabric_cn.md

+# 使用fabric启动集群训练
+
+## 准备一个Linux集群
+可以在`paddle/scripts/cluster_train_v2/fabric/docker_cluster`目录下，执行`kubectl -f ssh_servers.yaml`启动一个测试集群，并使用`kubectl get po -o wide`获得这些节点的IP地址。
+
+## 启动集群作业
+
+`paddle.py` 提供了自动化脚本来启动不同节点中的所有 PaddlePaddle 集群进程。默认情况下，所有命令行选项可以设置为 `paddle.py` 命令选项并且 `paddle.py` 将透明、自动地将这些选项应用到 PaddlePaddle 底层进程。
+
+`paddle.py` 为方便作业启动提供了两个独特的命令选项。
+
+-  `job_dispatch_package`  设为本地 `workspace` 目录，它将被分发到 `conf.py` 中设置的所有节点。它有助于帮助频繁修改和访问工作区文件的用户减少负担，否则频繁的多节点工作空间部署可能会很麻烦。
+-  `job_workspace`  设为已部署的工作空间目录，`paddle.py` 将跳过分发阶段直接启动所有节点的集群作业。它可以帮助减少分发延迟。
+
+`cluster_train/run.sh` 提供了命令样例来运行 `doc/howto/usage/cluster/src/word2vec` 集群任务，只需用您定义的目录修改 `job_dispatch_package` 和 `job_workspace`，然后：
+```
+sh run.sh
+```
+
+集群作业将会在几秒后启动。
+
+## 终止集群作业
+`paddle.py`能获取`Ctrl + C` SIGINT 信号来自动终止它启动的所有进程。只需中断 `paddle.py` 任务来终止集群作业。如果程序崩溃你也可以手动终止。
+
+## 检查集群训练结果
+详细信息请检查 $workspace/log 里的日志，每一个节点都有相同的日志结构。
+
+`paddle_trainer.INFO`
+提供几乎所有训练的内部输出日志，与本地训练相同。这里检验运行时间模型的收敛。
+
+`paddle_pserver2.INFO`
+提供 pserver 运行日志，有助于诊断分布式错误。
+
+`server.log`
+提供 parameter server 进程的 stderr 和 stdout。训练失败时可以检查错误日志。
+
+`train.log`
+提供训练过程的 stderr 和 stdout。训练失败时可以检查错误日志。
+
+## 检查模型输出
+运行完成后，模型文件将被写入节点 0 的 `output` 目录中。
+工作空间中的 `nodefile` 表示当前集群作业的节点 ID。

doc/howto/usage/cluster/fabric_en.md

+# Cluster Training Using Fabric
+
+## Prepare a Linux cluster
+
+Run `kubectl -f ssh_servers.yaml` under the directory:  `paddle/scripts/cluster_train_v2/fabric/docker_cluster` will launch a demo cluster. Run `kubectl get po -o wide` to get IP addresses of these nodes.
+
+## Launching Cluster Job
+`paddle.py` provides automatical scripts to start all PaddlePaddle cluster processes in different nodes. By default, all command line options can be set as `paddle.py` command options and `paddle.py` will transparently and automatically set these options to PaddlePaddle lower level processes.
+
+`paddle.py`provides two distinguished command option for easy job launching.
+
+- `job_dispatch_package` set it with local `workspace` directory, it will be dispatched to all nodes which is set in `conf.py`. It could be helpful for frequently manipulating workspace files. otherwise, frequent multi-nodes workspace deployment is very annoying.
+- `job_workspace`  set it with already deployed workspace directory, `paddle.py` will skip dispatch stage to directly launch cluster job with all nodes. It could help to reduce heavy
+dispatch latency.
+
+`cluster_train/run.sh` provides command line sample to run `demo/recommendation` cluster job, just modify `job_dispatch_package` and `job_workspace` with your defined directory, then:
+```
+sh run.sh
+```
+
+The cluster Job will start in several seconds.
+
+## Kill Cluster Job
+`paddle.py` can capture `Ctrl + C` SIGINT signal to automatically kill all processes launched by it. So just stop `paddle.py` to kill cluster job. You should manually kill the job if the program crashed.
+
+## Check Cluster Training Result
+Check log in $workspace/log for details, each node owns same log structure.
+
+`paddle_trainer.INFO`
+It provides almost all internal output log for training,  same as local training. Check runtime model convergence here.
+
+`paddle_pserver2.INFO`
+It provides parameter server running log, which could help to diagnose distributed error.
+
+`server.log`
+It provides stderr and stdout of parameter server process. Check error log if training crashes.
+
+`train.log`
+It provides stderr and stdout of trainer process. Check error log if training crashes.
+
+## Check Model Output
+After one pass finished, model files will be written in `output` directory in node 0.
+`nodefile` in workspace indicates the node id of current cluster job.

doc/howto/usage/cluster/k8s_aws_cn.md

+k8s_aws_en.md
\ No newline at end of file

doc/howto/usage/k8s/k8s_aws_en.md → doc/howto/usage/cluster/k8s_aws_en.md

@@ -493,7 +493,7 @@ spec:
     spec:
       containers:
       - name: paddle-data
-        image: paddledev/paddle-tutorial:k8s_data
+        image: paddlepaddle/paddle-tutorial:k8s_data
         imagePullPolicy: Always
         volumeMounts:
         - mountPath: "/efs"
@@ -522,7 +522,7 @@ NAME          DESIRED   SUCCESSFUL   AGE
 paddle-data   1         1            6m
 ```
 
-Data preparation is done by docker image `paddledev/paddle-tutorial:k8s_data`, see [here](src/k8s_data/README.md) for how to build this docker image and source code.
+Data preparation is done by docker image `paddlepaddle/paddle-tutorial:k8s_data`, see [here](src/k8s_data/README.md) for how to build this docker image and source code.
 
 #### Start Training
 
@@ -545,7 +545,7 @@ spec:
           claimName: efsvol
       containers:
       - name: trainer
-        image: paddledev/paddle-tutorial:k8s_train
+        image: paddlepaddle/paddle-tutorial:k8s_train
         command: ["bin/bash",  "-c", "/root/start.sh"]
         env:
         - name: JOB_NAME
@@ -617,7 +617,7 @@ kubectl --kubeconfig=kubeconfig log -f POD_NAME
 
 Run `kubectl --kubeconfig=kubeconfig describe job paddle-cluster-job` to check training job status. It will complete in around 20 minutes.
 
-The details for start `pserver` and `trainer` are hidden inside docker image `paddledev/paddle-tutorial:k8s_train`, see [here](src/k8s_train/README.md) for how to build the docker image and source code.
+The details for start `pserver` and `trainer` are hidden inside docker image `paddlepaddle/paddle-tutorial:k8s_train`, see [here](src/k8s_train/README.md) for how to build the docker image and source code.
 
 #### Inspect Training Output
 

doc/howto/usage/k8s/k8s_cn.md → doc/howto/usage/cluster/k8s_cn.md

 # Kubernetes单机训练
 
-在这篇文档里，我们介绍如何在 Kubernetes 集群上启动一个单机使用CPU的Paddle训练作业。在下一篇中，我们将介绍如何启动分布式训练作业。
+在这篇文档里，我们介绍如何在 Kubernetes 集群上启动一个单机使用CPU的PaddlePaddle训练作业。在下一篇中，我们将介绍如何启动分布式训练作业。
 
 ## 制作Docker镜像
 
-在一个功能齐全的Kubernetes机群里，通常我们会安装Ceph等分布式文件系统来存储训练数据。这样的话，一个分布式Paddle训练任务中的每个进程都可以从Ceph读取数据。在这个例子里，我们只演示一个单机作业，所以可以简化对环境的要求，把训练数据直接放在
-Paddle的Docker image里。为此，我们需要制作一个包含训练数据的Paddle镜像。
+在一个功能齐全的Kubernetes机群里，通常我们会安装Ceph等分布式文件系统来存储训练数据。这样的话，一个分布式PaddlePaddle训练任务中
+的每个进程都可以从Ceph读取数据。在这个例子里，我们只演示一个单机作业，所以可以简化对环境的要求，把训练数据直接放在
+PaddlePaddle的Docker Image里。为此，我们需要制作一个包含训练数据的PaddlePaddle镜像。
+
+PaddlePaddle的 `paddlepaddle/paddle:cpu-demo-latest` 镜像里有PaddlePaddle的源码与demo，
+（请注意，默认的PaddlePaddle生产环境镜像 `paddlepaddle/paddle:latest` 是不包括源码的，PaddlePaddle的各版本镜像可以参考
+[Docker Installation Guide](http://paddlepaddle.org/docs/develop/documentation/zh/getstarted/build_and_install/docker_install_cn.html)），
+下面我们使用这个镜像来下载数据到Docker Container中，并把这个包含了训练数据的Container保存为一个新的镜像。
 
-Paddle 的 [Quick Start Tutorial](http://www.paddlepaddle.org/doc/demo/quick_start/index_en.html) 
-里介绍了用Paddle源码中的脚本下载训练数据的过程。
-而 `paddledev/paddle:cpu-demo-latest` 镜像里有 Paddle 源码与demo，（ 请注意，默认的
-Paddle镜像 `paddledev/paddle:cpu-latest` 是不包括源码的, Paddle的各版本镜像可以参考 [Docker installation guide](http://www.paddlepaddle.org/doc/build/docker_install.html) ），所以我们使用这个镜像来下载训练数据到Docker container中，然后把这个包含了训练数据的container保存为一个新的镜像。
-  
 ### 运行容器
 
 ```
-$ docker run --name quick_start_data -it paddledev/paddle:cpu-demo-latest
+$ docker run --name quick_start_data -it paddlepaddle/paddle:cpu-demo-latest
 ```
 
 ### 下载数据
@@ -103,7 +104,7 @@ spec:
       restartPolicy: Never
 ```
 
-### 创建Paddle Job
+### 创建PaddlePaddle Job
 
 使用上文创建的yaml文件创建Kubernetes Job，命令为：
 

doc/howto/usage/k8s/k8s_distributed_cn.md → doc/howto/usage/cluster/k8s_distributed_cn.md

 # Kubernetes分布式训练
 
-前一篇文章介绍了如何在Kubernetes集群上启动一个单机PaddlePaddle训练作业 (Job)。在这篇文章里，我们介绍如何在Kubernetes集群上进行分布式PaddlePaddle训练作业。关于PaddlePaddle的分布式训练，文章 [Cluster Training](https://github.com/baidu/Paddle/blob/develop/doc/cluster/opensource/cluster_train.md)介绍了一种通过SSH远程分发任务，进行分布式训练的方法，与此不同的是，本文将介绍在Kubernetes容器管理平台上快速构建PaddlePaddle容器集群，进行分布式训练的方案。
-
-有关Kubernetes相关概念以及如何搭建和配置Kubernetes集群，可以参考[k8s_basis](./k8s_basis_cn.md)。
+前一篇文章介绍了如何在Kubernetes集群上启动一个单机PaddlePaddle训练作业 (Job)。在这篇文章里，我们介绍如何在Kubernetes集群上进行分布式PaddlePaddle训练作业。关于PaddlePaddle的分布式训练，文章 [Cluster Training](http://www.paddlepaddle.org/docs/develop/documentation/zh/howto/usage/cluster/cluster_train_cn.html)介绍了一种通过SSH远程分发任务，进行分布式训练的方法，与此不同的是，本文将介绍在Kubernetes容器管理平台上快速构建PaddlePaddle容器集群，进行分布式训练的方案。
 
 ## 整体方案
 
@@ -28,7 +26,7 @@ PaddlePaddle镜像需要提供`paddle pserver`与`paddle train`进程的运行
 - 拷贝训练文件到容器内
 - 生成`paddle pserver`与`paddle train`进程的启动参数，并且启动训练
 
-因为官方镜像 `paddledev/paddle:cpu-latest` 内已经包含PaddlePaddle的执行程序但是还没上述功能，所以我们可以在这个基础上，添加启动脚本，制作新镜像来完成以上的工作。参考镜像的[*Dockerfile*](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/howto/usage/cluster/k8s/src/k8s_train/Dockerfile)。
+因为官方镜像 `paddlepaddle/paddle:latest` 内已经包含PaddlePaddle的执行程序但是还没上述功能，所以我们可以在这个基础上，添加启动脚本，制作新镜像来完成以上的工作。参考镜像的[*Dockerfile*](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/howto/usage/cluster/src/k8s_train/Dockerfile)。
 
 ```bash
 $ cd doc/howto/usage/k8s/src/k8s_train
@@ -62,7 +60,7 @@ spec:
       hostNetwork: true
       containers:
       - name: paddle-data
-        image: paddledev/paddle-tutorial:k8s_data
+        image: paddlepaddle/paddle-tutorial:k8s_data
         imagePullPolicy: Always
         volumeMounts:
         - mountPath: "/mnt"
@@ -149,20 +147,19 @@ spec:
 
 文件中，`metadata`下的`name`表示这个job的名字。`parallelism，completions`字段表示这个job会同时开启3个PaddlePaddle节点，成功训练且退出的pod数目为3时，这个job才算成功结束。然后申明一个存储卷`jobpath`，代表宿主机目录`/home/work/mfs`，在对容器的描述`containers`字段中，将此目录挂载为容器的`/home/jobpath`目录，这样容器的`/home/jobpath`目录就成为了共享存储，放在这个目录里的文件其实是保存到了MFS上。
 
-`env`字段表示容器的环境变量，我们将`paddle`运行的一些参数通过这种方式传递到容器内。
+`env`字段表示容器的环境变量，我们将`paddle`运行的一些参数通过这种方式传递到容器内：
+
 
-环境变量 | 说明
---- | ---
-JOB_PATH | 共享存储挂在的路径
-JOB_NAME | Job的名字
-TRAIN_CONFIG_DIR | 本次训练文件所在目录，与JOB_PATH,JOB_NAME组合可以找到本次训练需要的文件路径
-CONF_PADDLE_NIC | `paddle pserver`进程需要的`--nics`参数，即网卡名
-CONF_PADDLE_PORT | `paddle paserver`的`--port`参数
-CONF_PADDLE_PORTS_NUM | 稠密更新的端口数量，即`--ports_num`参数
-CONF_PADDLE_PORTS_NUM_SPARSE | 稀疏更新的端口数量，即`--ports_num_for_sparse`参数
-CONF_PADDLE_GRADIENT_NUM | 训练节点数量，即`--num_gradient_servers参数`
+- JOB_PATH：共享存储挂在的路径
+- JOB_NAME：Job的名字
+- TRAIN_CONFIG_DIR：本次训练文件所在目录，与JOB_PATH,JOB_NAME组合可以找到本次训练需要的文件路径
+- CONF_PADDLE_NIC：`paddle pserver`进程需要的`--nics`参数，即网卡名
+- CONF_PADDLE_PORT：`paddle paserver`的`--port`参数
+- CONF_PADDLE_PORTS_NUM：稠密更新的端口数量，即`--ports_num`参数
+- CONF_PADDLE_PORTS_NUM_SPARSE：稀疏更新的端口数量，即`--ports_num_for_sparse`参数
+- CONF_PADDLE_GRADIENT_NUM：训练节点数量，即`--num_gradient_servers参数`
 
-这些参数的具体描述，读者可以查看[这里](http://www.paddlepaddle.org/doc/ui/cmd_argument/detail_introduction.html#parameter-server-and-distributed-communication)。
+这些参数的具体描述，读者可以查看[这里](http://www.paddlepaddle.org/docs/develop/documentation/zh/howto/usage/cmd_parameter/detail_introduction_cn.html)。
 
 编写完YAML文件后，可以使用Kubernetes的命令行工具创建job。
 

doc/howto/usage/k8s/k8s_en.md → doc/howto/usage/cluster/k8s_en.md

-# Paddle On Kubernetes
+# PaddlePaddle On Kubernetes
 
->In this article, we will introduce how to run Paddle training job on single CPU machine using Kubernetes. In next article, we will introduce how to run Paddle training job on distributed cluster.
+In this article, we will introduce how to run PaddlePaddle training job on single CPU machine using Kubernetes. In next article, we will introduce how to run PaddlePaddle training job on distributed cluster.
 
 ## Build Docker Image
 
-In distributed Kubernetes cluster, we will use Ceph or other shared storage system for storing training related data so that all processes in Paddle training can retrieve data from Ceph. In this example, we will only demo training job on single machine. In order to simplify the requirement of the environment, we will directly put training data into Paddle's Docker Image, so we need to create a Paddle Docker image that already includes the training data.
+In distributed Kubernetes cluster, we will use Ceph or other distributed
+storage system for storing training related data so that all processes in
+PaddlePaddle training can retrieve data from Ceph. In this example, we will
+only demo training job on single machine. In order to simplify the requirement
+of the environment, we will directly put training data into the PaddlePaddle Docker Image,
+so we need to create a PaddlePaddle Docker image that includes the training data.
+
+The production Docker Image `paddlepaddle/paddle:cpu-demo-latest` has the PaddlePaddle
+source code and demo. (Caution: Default PaddlePaddle Docker Image `paddlepaddle/paddle:latest` doesn't include
+the source code, PaddlePaddle's different versions of Docker Image can be referred here:
+[Docker Installation Guide](http://paddlepaddle.org/docs/develop/documentation/zh/getstarted/build_and_install/docker_install_en.html)),
+so we run this Docker Image and download the training data, and then commit the whole
+Container to be a new Docker Image.
 
-Paddle's [Quick Start Tutorial](http://www.paddlepaddle.org/doc/demo/quick_start/index_en.html) introduces how to download and train data by using script from Paddle's source code.
-And `paddledev/paddle:cpu-demo-latest` image has the Paddle source code and demo. (Caution: Default Paddle image `paddledev/paddle:cpu-latest` doesn't include the source code, Paddle's different versions of image can be referred here: [Docker installation guide](http://www.paddlepaddle.org/doc/build/docker_install.html)), so we run this container and download the training data, and then commit the whole container to be a new Docker image.
-  
 ### Run Docker Container
 
 ```
-$ docker run --name quick_start_data -it paddledev/paddle:cpu-demo-latest
+$ docker run --name quick_start_data -it paddlepaddle/paddle:cpu-demo-latest
 ```
 
 ### Download Training Data
@@ -67,7 +76,7 @@ $ docker commit quick_start_data mypaddle/paddle:quickstart
 
 ## Use Kubernetes For Training
 
->We will use Kubernetes job for training process, following steps shows how to do the training with Kubernetes.
+We will use Kubernetes job for training process, following steps shows how to do the training with Kubernetes.
 
 ### Create Yaml Files
 
@@ -99,7 +108,7 @@ spec:
       restartPolicy: Never
 ```
 
-### Start Paddle Job
+### Start PaddlePaddle Job
 
 Using the above yaml file to start the Kubernetes job.
 

doc/howto/usage/cluster/openmpi_cn.md

+# 在OpenMPI集群中提交训练作业
+
+## 准备OpenMPI集群
+
+执行下面的命令以启动3个节点的OpenMPI集群和一个"head"节点：
+
+```bash
+paddle/scripts/cluster_train_v2/openmpi/docker_cluster
+kubectl create -f head.yaml
+kubectl create -f mpi-nodes.yaml
+```
+
+然后可以从head节点ssh无密码登录到OpenMPI的每个节点上。
+
+## 启动集群作业
+
+您可以按照下面的步骤在OpenMPI集群中提交paddle训练任务：
+
+```bash
+# 获得head和node节点的IP地址
+kubectl get po -o wide
+# 将node节点的IP地址保存到machines文件中
+kubectl get po -o wide | grep nodes | awk '{print $6}' > machines
+# 拷贝必要的文件到head节点
+scp -i ssh/id_rsa.mpi.pub machines prepare.py train.py start_mpi_train.sh tutorial@[headIP]:~
+# ssh 登录到head节点
+ssh -i ssh/id_rsa.mpi.pub tutorial@[headIP]
+# --------------- 以下操作均在head节点中执行 ---------------
+# 准备训练数据
+python prepare.py
+# 拷贝训练程序和字典文件到每台MPI节点
+cat machines | xargs -i scp word_dict.pickle train.py start_mpi_train.sh machines {}:/home/tutorial
+# 创建日志目录
+mpirun -hostfile machines -n 3 mkdir /home/tutorial/logs
+# 拷贝训练数据到各自的节点
+scp train.txt-00000 test.txt-00000 [node1IP]:/home/tutorial
+scp train.txt-00001 test.txt-00001 [node2IP]:/home/tutorial
+scp train.txt-00002 test.txt-00002 [node3IP]:/home/tutorial
+# 启动训练任务
+mpirun -hostfile machines -n 3  /home/tutorial/start_mpi_train.sh
+```

doc/howto/usage/cluster/openmpi_en.md

+# Cluster Training Using OpenMPI
+
+## Prepare an OpenMPI cluster
+
+Run the following command to start a 3-node MPI cluster and one "head" node.
+
+```bash
+cd paddle/scripts/cluster_train_v2/openmpi/docker_cluster
+kubectl create -f head.yaml
+kubectl create -f mpi-nodes.yaml
+```
+
+Then you can log in to every OpenMPI node using ssh without input any passwords.
+
+## Launching Cluster Job
+
+Follow the steps to launch a PaddlePaddle training job in OpenMPI cluster:\
+
+```bash
+# find out node IP addresses
+kubectl get po -o wide
+# generate a "machines" file containing node IP addresses
+kubectl get po -o wide | grep nodes | awk '{print $6}' > machines
+# copy necessary files onto "head" node
+scp -i ssh/id_rsa.mpi.pub machines prepare.py train.py start_mpi_train.sh tutorial@[headIP]:~
+# login to head node using ssh
+ssh -i ssh/id_rsa.mpi.pub tutorial@[headIP]
+# --------------- in head node ---------------
+# prepare training data
+python prepare.py
+# copy training data and dict file to MPI nodes
+cat machines | xargs -i scp word_dict.pickle train.py start_mpi_train.sh machines {}:/home/tutorial
+# creat a directory for storing log files
+mpirun -hostfile machines -n 3 mkdir /home/tutorial/logs
+# copy training data to every node
+scp train.txt-00000 test.txt-00000 [node1IP]:/home/tutorial
+scp train.txt-00001 test.txt-00001 [node2IP]:/home/tutorial
+scp train.txt-00002 test.txt-00002 [node3IP]:/home/tutorial
+# start the job
+mpirun -hostfile machines -n 3  /home/tutorial/start_mpi_train.sh
+```

doc/howto/usage/k8s/src/Dockerfile → doc/howto/usage/cluster/src/Dockerfile

-FROM paddledev/paddle:cpu-latest
+FROM paddlepaddle/paddle:latest
 
 MAINTAINER zjsxzong89@gmail.com
 

doc/howto/usage/k8s/src/k8s_train/Dockerfile → doc/howto/usage/cluster/src/k8s_train/Dockerfile

-FROM paddledev/paddle:cpu-latest
+FROM paddlepaddle/paddle:latest
 
 COPY start.sh /root/
 COPY start_paddle.py /root/

doc/howto/usage/k8s/k8s_basis_cn.md

-# Kubernetes 简介
-
-[*Kubernetes*](http://kubernetes.io/)是Google开源的容器集群管理系统，其提供应用部署、维护、扩展机制等功能，利用Kubernetes能方便地管理跨机器运行容器化的应用。Kubernetes可以在物理机或虚拟机上运行，且支持部署到[AWS](http://kubernetes.io/docs/getting-started-guides/aws)，[Azure](http://kubernetes.io/docs/getting-started-guides/azure/)，[GCE](http://kubernetes.io/docs/getting-started-guides/gce)等多种公有云环境。介绍分布式训练之前，需要对[Kubernetes](http://kubernetes.io/)有一个基本的认识，下面先简要介绍一下本文用到的几个Kubernetes概念。
-
-- [*Node*](http://kubernetes.io/docs/admin/node/) 表示一个Kubernetes集群中的一个工作节点，这个节点可以是物理机或者虚拟机，Kubernetes集群就是由node节点与master节点组成的。
-
-- [*Pod*](http://kubernetes.io/docs/user-guide/pods/) 是一组(一个或多个)容器，pod是Kubernetes的最小调度单元，一个pod中的所有容器会被调度到同一个node上。Pod中的容器共享NET，PID，IPC，UTS等Linux namespace。由于容器之间共享NET namespace，所以它们使用同一个IP地址，可以通过*localhost*互相通信。不同pod之间可以通过IP地址访问。
-
-- [*Job*](http://kubernetes.io/docs/user-guide/jobs/) 描述Kubernetes上运行的作业，一次作业称为一个job，通常每个job包括一个或者多个pods，job启动后会创建这些pod并开始执行一个程序，等待这个程序执行成功并返回0则成功退出，如果执行失败，也可以配置不同的重试机制。
-
-- [*Volume*](http://kubernetes.io/docs/user-guide/volumes/) 存储卷，是pod内的容器都可以访问的共享目录，也是容器与node之间共享文件的方式，因为容器内的文件都是暂时存在的，当容器因为各种原因被销毁时，其内部的文件也会随之消失。通过volume，就可以将这些文件持久化存储。Kubernetes支持多种volume，例如hostPath(宿主机目录)，gcePersistentDisk，awsElasticBlockStore等。
-
-- [*Namespaces*](https://kubernetes.io/docs/user-guide/namespaces/) 命名空间，在kubernetes中创建的所有资源对象(例如上文的pod，job)等都属于一个命名空间，在同一个命名空间中，资源对象的名字是唯一的，不同空间的资源名可以重复，命名空间主要为了对象进行逻辑上的分组便于管理。本文只使用了默认命名空间。
-
-- [*PersistentVolume*](https://kubernetes.io/docs/user-guide/persistent-volumes/): 和[*PersistentVolumeClaim*](https://kubernetes.io/docs/user-guide/persistent-volumes/#persistentvolumeclaims)结合，将外部的存储服务在Kubernetes中描述成为统一的资源形式，便于存储资源管理和Pod引用。
-
-## 部署Kubernetes集群
-
-Kubernetes提供了多种集群部署的方案，本文档内不重复介绍。这里给出集中常见的部署方法：
-
-- [*minikube*](https://kubernetes.io/docs/getting-started-guides/minikube/): 快速在本地启动一个单机的kubernetes服务器，便于本地验证和测试。
-- [*kubeadm*](http://kubernetes.io/docs/getting-started-guides/kubeadm/): 在不同操作系统，不同主机(Bare-Metal, AWS, GCE)条件下，快速部署集群。
-- [*AWS EC2*](https://kubernetes.io/docs/getting-started-guides/aws/): 在aws上快速部署集群。
-- [*Bare-Metal*](https://kubernetes.io/docs/getting-started-guides/centos/centos_manual_config/): 在物理机上手动部署。
-
-可以参考[这个表格](https://kubernetes.io/docs/getting-started-guides/#table-of-solutions)选择适合您的场景的合适方案。
-
-## 选择存储方案
-
-容器不会保留在运行时生成的数据，job或者应用程序在容器中运行时生成的数据会在容器销毁时消失。为了完成分布式机器学习训练任务，需要有一个外部的存储服务来保存训练所需数据和训练输出。
-常见的可选存储服务包括：
-
-- [*NFS*](https://github.com/kubernetes/kubernetes/tree/master/examples/volumes/nfs): 可以将磁盘上某个目录共享给网络中其他机器访问。部署和配置比较简单，可以用于小量数据的验证。不提供分布式存储，高可用，冗余等功能。NFS的部署方法可以参考[这里](http://www.tecmint.com/how-to-setup-nfs-server-in-linux/)。
-- [*GlusterFS*](http://gluster.readthedocs.io/en/latest/Quick-Start-Guide/Quickstart/): 网络分布式文件系统，可以在Kubernetes中按照[这个](https://github.com/kubernetes/kubernetes/tree/master/examples/volumes/glusterfs)例子使用。
-- [*Ceph*](http://docs.ceph.com/docs/master/): 分布式文件系统，支持rbd，POSIX API接口(ceph fs)和对象存储API，参考[这里](https://kubernetes.io/docs/user-guide/volumes/#rbd)。
-- [*MooseFS*](https://moosefs.com/documentation.html): 一个分布式的存储系统。需要先挂载到服务器Node上再通过kubernetes hostPath Volume挂载到容器中。
-
-## 配置kubectl
-
-### 安装kubectl
-```
-# OS X
-curl -LO https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/darwin/amd64/kubectl
-
-# Linux
-curl -LO https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/linux/amd64/kubectl
-
-# Windows
-curl -LO https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/windows/amd64/kubectl.exe
-```
-
-### 配置kubectl访问你的kubernetes集群
-
-编辑`~/.kube/config`这个配置文件，修改`Master-IP`的地址。如果使用SSL认证，则需要配置`certificate-authority`和`users`中的用户证书。如果是使用非SSL方式访问（比如通过8080端口），也可以去掉这些证书的配置。
-```
-apiVersion: v1
-clusters:
-- cluster:
-    certificate-authority: /path/to/ca.crt
-    server: https://[Master-IP]:443
-  name: minikube
-contexts:
-- context:
-    cluster: minikube
-    user: minikube
-  name: minikube
-current-context: minikube
-kind: Config
-preferences: {}
-users:
-- name: minikube
-  user:
-    client-certificate: /path/to/apiserver.crt
-    client-key: /Users/wuyi/.minikube/apiserver.key
-```

doc/mobile/cross_compiling_for_ios_cn.md

@@ -18,11 +18,11 @@ PaddlePaddle为交叉编译提供了工具链配置文档[cmake/cross_compiling/
 
 - `CMAKE_SYSTEM_NAME`，CMake编译的目标平台，必须设置为`iOS`。在设置`CMAKE_SYSTEM_NAME=iOS`后，PaddlePaddle的CMake系统会自动编译所有的第三方依赖库，并且强制设置一些PaddlePaddle参数的值（`WITH_C_API=ON`、`WITH_GPU=OFF`、`WITH_AVX=OFF`、`WITH_PYTHON=OFF`、`WITH_RDMA=OFF`）。
 - `WITH_C_API`，是否编译C-API预测库，必须设置为ON。在iOS平台上只支持使用C-API来预测。
-- `WITH_SWIG_PY`，必须设置为ON。在iOS平台上不支持通过swig调用来训练或者预测。
+- `WITH_SWIG_PY`，必须设置为`OFF`。在iOS平台上不支持通过swig调用来训练或者预测。
 
 iOS平台可选配置参数：
 
-- `IOS_PLATFORM`，可设置为`OS/SIMULATOR`，默认值为`OS`。
+- `IOS_PLATFORM`，可设置为`OS`（默认值）或`SIMULATOR`。
   - `OS`，构建目标为`arm`架构的iPhone或者iPad等物理设备。
   - `SIMULATOR`，构建目标为`x86`架构的模拟器平台。
 - `IOS_ARCH`，目标架构。针对不同的`IOS_PLATFORM`，可设置的目标架构如下表所示，默认编译所有架构：

doc/mobile/cross_compiling_for_ios_en.md

+# PaddlePaddle Compiling Guide for iOS
+
+This tutorial will walk you through cross compiling the PaddlePaddle library for iOS from the source in MacOS.
+
+## Preparation
+
+Apple provides Xcode for cross-compiling and IDE for iOS development. Download from App store or [here](https://developer.apple.com/cn/xcode/). To verify your installation, run command as follows
+
+```bash
+$ xcodebuild -version
+Xcode 9.0
+Build version 9A235
+```
+
+## Cross-compiling configurations
+
+PaddlePaddle provides cross-compiling toolchain configuration documentation [cmake/cross_compiling/ios.cmake](https://github.com/PaddlePaddle/Paddle/blob/develop/cmake/cross_compiling/ios.cmake), which has some default settings for frequently used compilers.
+
+There are some mandatory environment variables need to be set before cross compiling PaddlePaddle for iOS:
+
+- `CMAKE_SYSTEM_NAME`, CMake compiling target platform name, has to be `iOS`. PaddlePaddle CMake will compile all the third party dependencies and enforce some parameters (`WITH_C_API=ON`, `WITH_GPU=OFF`, `WITH_AVX=OFF`, `WITH_PYTHON=OFF`,`WITH_RDMA=OFF`) when this variable is set with value `iOS`.
+
+- `WITH_C_API`, Whether to compile inference C-API library, has to be `ON`, since C-API is the only supported interface for inferencing in iOS.
+- `WITH_SWIG_PY`, has to be `OFF`. It's not supported to inference or train via swig in iOS.
+
+Optional environment variables for iOS are:
+
+- `IOS_PLATFORM`, either `OS` (default) or `SIMULATOR`.
+  - `OS`, build targets ARM-based physical devices like iPhone or iPad.
+  - `SIMULATOR`, build targets x86 architecture simulators.
+- `IOS_ARCH`, target architecture. By default, all architecture types will be compiled. If you need to specify the architecture to compile for, please find valid values for different `IOS_PLATFORM` settings from the table below:
+
+    <table class="docutils">
+    <colgroup>
+      <col width="35%" />
+      <col width="65%" />
+    </colgroup>
+    <thead valign="bottom">
+      <tr class="row-odd">
+      <th class="head">IOS_PLATFORM</th>
+      <th class="head">IOS_ARCH</th>
+    </tr>
+    </thead>
+    <tbody valign="top">
+      <tr class="row-even">
+      <td>OS</td>
+      <td>armv7, armv7s, arm64 </td>
+    </tr>
+    <tr class="row-odd">
+      <td>SIMULATOR</td>
+      <td>i386, x86_64 </td>
+    </tr>
+    </tbody>
+    </table>
+
+- `IOS_DEPLOYMENT_TARGET`, minimum iOS version to deployment, `7.0` by default.
+- `IOS_ENABLE_BITCODE`, whether to enable [Bitcode](https://developer.apple.com/library/content/documentation/IDEs/Conceptual/AppDistributionGuide/AppThinning/AppThinning.html#//apple_ref/doc/uid/TP40012582-CH35-SW3), values can be `ON/OFF`, `ON` by default.
+- `IOS_USE_VECLIB_FOR_BLAS`, whether to use [vecLib](https://developer.apple.com/documentation/accelerate/veclib) framework for BLAS computing. values can be `ON/OFF`, `OFF` by default.
+- `IOS_DEVELOPMENT_ROOT`, the path to `Developer` directory, can be explicitly set with your `/path/to/platform/Developer`. If left blank, PaddlePaddle will automatically pick the Xcode corresponding `platform`'s `Developer` directory based on your `IOS_PLATFORM` value.
+- `IOS_SDK_ROOT`, the path to `SDK` root, can be explicitly set with your  `/path/to/platform/Developer/SDKs/SDK`. if left black, PaddlePaddle will pick the latest SDK in the directory of `IOS_DEVELOPMENT_ROOT`.
+
+other settings：
+
+- `USE_EIGEN_FOR_BLAS`, whether to use Eigen for matrix computing. effective when `IOS_USE_VECLIB_FOR_BLAS=OFF`. Values can be `ON/OFF`, `OFF` by default.
+- `HOST_C/CXX_COMPILER`, host C/C++ compiler. Uses value from environment variable `CC/CXX` by default or `cc/c++` if `CC/CXX` doesn't exist.
+
+some typical cmake configurations:
+
+```bash
+cmake -DCMAKE_SYSTEM_NAME=iOS \
+      -DIOS_PLATFORM=OS \
+      -DIOS_ARCH="armv7;arm64" \
+      -DIOS_ENABLE_BITCODE=ON \
+      -DIOS_USE_VECLIB_FOR_BLAS=ON \
+      -DCMAKE_INSTALL_PREFIX=your/path/to/install \
+      -DWITH_C_API=ON \
+      -DWITH_TESTING=OFF \
+      -DWITH_SWIG_PY=OFF \
+      ..
+```
+
+```bash
+cmake -DCMAKE_SYSTEM_NAME=iOS \
+      -DIOS_PLATFORM=SIMULATOR \
+      -DIOS_ARCH="x86_64" \
+      -DIOS_USE_VECLIB_FOR_BLAS=ON \
+      -DCMAKE_INSTALL_PREFIX=your/path/to/install \
+      -DWITH_C_API=ON \
+      -DWITH_TESTING=OFF \
+      -DWITH_SWIG_PY=OFF \
+      ..
+```
+
+You can set other compiling parameters for your own need. I.E. if you are trying to minimize the library size, set `CMAKE_BUILD_TYPE` with `MinSizeRel`; or if the performance is your concern, set `CMAKE_BUILD_TYPE` with `Release`. You can even manipulate the PaddlePaddle compiling procedure by manually set `CMAKE_C/CXX_FLAGS` values.
+
+**TIPS for a better performance**:
+
+- set `CMAKE_BUILD_TYPE` with `Release`
+- set `IOS_USE_VECLIB_FOR_BLAS` with `ON`
+
+## Compile and install
+
+After CMake, run following commands, PaddlePaddle will download the compile 3rd party dependencies, compile and install PaddlePaddle inference library.
+
+```
+$ make
+$ make install
+```
+
+Please Note: if you compiled PaddlePaddle in the source directory for other platforms, do remove `third_party` and `build` directory within the source with `rm -rf` to ensure that all the 3rd party libraries dependencies and PaddlePaddle is newly compiled with current CMake configuration.
+
+`your/path/to/install` directory will have following directories after `compile` and `install`:
+
+- `include`, contains all the C-API header files.
+- `lib`, contains PaddlePaddle C-API static library.
+- `third_party` contains all the 3rd party libraries.
+
+Please note: if PaddlePaddle library need to support both physical devices and simulators, you will need to compile correspondingly, then merge fat library with `lipo`.
+
+Now you will have PaddlePaddle library compiled and installed, the fat library can be used in deep learning related iOS APPs. Please refer to C-API documentation for usage guides.

doc/mobile/index_en.rst

@@ -5,4 +5,5 @@ MOBILE
   :maxdepth: 1
 
   cross_compiling_for_android_en.md
+  cross_compiling_for_ios_en.md
   cross_compiling_for_raspberry_en.md

go/pserver/client/c/test/test_cclient.c

 /* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
 
-   Licensed under the Apache License, Version 2.0 (the "License");
-   you may not use this file except in compliance with the License.
-   You may obtain a copy of the License at
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
 
-   http://www.apache.org/licenses/LICENSE-2.0
+    http://www.apache.org/licenses/LICENSE-2.0
 
-   Unless required by applicable law or agreed to in writing, software
-   distributed under the License is distributed on an "AS IS" BASIS,
-   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-   See the License for the specific language governing permissions and
-   limitations under the License. */
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
 
 #include <stdio.h>
 #include <stdlib.h>

paddle/framework/CMakeLists.txt

@@ -21,6 +21,8 @@ cc_test(variable_test SRCS variable_test.cc)
 cc_library(scope SRCS scope.cc DEPS glog)
 cc_test(scope_test SRCS scope_test.cc DEPS scope)
 
+cc_library(data_transform SRCS data_transform.cc DEPS tensor framework_proto)
+cc_test(data_transform_test SRCS data_transform_test.cc DEPS data_transform device_context)
 
 cc_library(attribute SRCS attribute.cc DEPS framework_proto)
 cc_test(program_desc_test SRCS program_desc_test.cc DEPS proto_desc
@@ -29,12 +31,13 @@ cc_library(op_proto_maker SRCS op_proto_maker.cc DEPS framework_proto attribute)
 cc_test(op_proto_maker_test SRCS op_proto_maker_test.cc DEPS op_proto_maker)
 cc_library(op_info SRCS op_info.cc DEPS attribute framework_proto)
 cc_library(shape_inference SRCS shape_inference.cc DEPS ddim attribute)
-cc_library(operator SRCS operator.cc DEPS op_info device_context tensor scope glog shape_inference)
-cc_test(operator_test SRCS operator_test.cc DEPS operator op_registry)
+cc_library(operator SRCS operator.cc DEPS op_info device_context tensor scope glog
+    shape_inference data_transform)
+cc_test(operator_test SRCS operator_test.cc DEPS operator op_registry init)
 cc_library(proto_desc SRCS var_desc.cc op_desc.cc block_desc.cc program_desc.cc DEPS shape_inference op_info operator glog)
 
 cc_library(op_registry SRCS op_registry.cc DEPS op_proto_maker op_info operator glog proto_desc)
-cc_test(op_registry_test SRCS op_registry_test.cc DEPS op_registry)
+nv_test(op_registry_test SRCS op_registry_test.cc DEPS op_registry)
 
 py_proto_compile(framework_py_proto SRCS framework.proto)
 # Generate an empty __init__.py to make framework_py_proto as a valid python module.
@@ -58,3 +61,10 @@ cc_test(var_type_inference_test SRCS var_type_inference_test.cc DEPS op_registry
         proto_desc)
 cc_library(selected_rows SRCS selected_rows.cc DEPS tensor)
 cc_test(selected_rows_test SRCS selected_rows_test.cc DEPS selected_rows)
+
+cc_library(threadpool SRCS threadpool.cc)
+cc_test(threadpool_test SRCS threadpool_test.cc DEPS threadpool)
+cc_library(init SRCS init.cc DEPS gflags device_context place stringpiece)
+cc_test(init_test SRCS init_test.cc DEPS init)
+
+cc_test(op_kernel_type_test SRCS op_kernel_type_test.cc DEPS place device_context framework_proto)

paddle/framework/attribute.cc

@@ -19,42 +19,42 @@ limitations under the License. */
 namespace paddle {
 namespace framework {
 
-Attribute GetAttrValue(const OpDesc::Attr& attr_desc) {
+Attribute GetAttrValue(const proto::OpDesc::Attr& attr_desc) {
   switch (attr_desc.type()) {
-    case framework::AttrType::BOOLEAN: {
+    case proto::AttrType::BOOLEAN: {
       return attr_desc.b();
     }
-    case framework::AttrType::INT: {
+    case proto::AttrType::INT: {
       return attr_desc.i();
     }
-    case framework::AttrType::FLOAT: {
+    case proto::AttrType::FLOAT: {
       return attr_desc.f();
     }
-    case framework::AttrType::STRING: {
+    case proto::AttrType::STRING: {
       return attr_desc.s();
     }
-    case framework::AttrType::BOOLEANS: {
+    case proto::AttrType::BOOLEANS: {
       std::vector<bool> val(attr_desc.bools_size());
       for (int i = 0; i < attr_desc.bools_size(); ++i) {
         val[i] = attr_desc.bools(i);
       }
       return val;
     }
-    case framework::AttrType::INTS: {
+    case proto::AttrType::INTS: {
       std::vector<int> val(attr_desc.ints_size());
       for (int i = 0; i < attr_desc.ints_size(); ++i) {
         val[i] = attr_desc.ints(i);
       }
       return val;
     }
-    case framework::AttrType::FLOATS: {
+    case proto::AttrType::FLOATS: {
       std::vector<float> val(attr_desc.floats_size());
       for (int i = 0; i < attr_desc.floats_size(); ++i) {
         val[i] = attr_desc.floats(i);
       }
       return val;
     }
-    case framework::AttrType::STRINGS: {
+    case proto::AttrType::STRINGS: {
       std::vector<std::string> val(attr_desc.strings_size());
       for (int i = 0; i < attr_desc.strings_size(); ++i) {
         val[i] = attr_desc.strings(i);

paddle/framework/attribute.h

@@ -27,12 +27,12 @@ limitations under the License. */
 namespace paddle {
 namespace framework {
 template <typename T>
-inline AttrType AttrTypeID() {
+inline proto::AttrType AttrTypeID() {
   Attribute tmp = T();
-  return static_cast<AttrType>(tmp.which() - 1);
+  return static_cast<proto::AttrType>(tmp.which() - 1);
 }
 
-Attribute GetAttrValue(const OpDesc::Attr& attr_desc);
+Attribute GetAttrValue(const proto::OpDesc::Attr& attr_desc);
 
 class AttrReader {
  public:

paddle/framework/backward.cc

 /* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
 
-   Licensed under the Apache License, Version 2.0 (the "License");
-   you may not use this file except in compliance with the License.
-   You may obtain a copy of the License at
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
 
-   http://www.apache.org/licenses/LICENSE-2.0
+    http://www.apache.org/licenses/LICENSE-2.0
 
-   Unless required by applicable law or agreed to in writing, software
-   distributed under the License is distributed on an "AS IS" BASIS,
-   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-   See the License for the specific language governing permissions and
-   limitations under the License. */
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
 
 #include "paddle/framework/backward.h"
 #include "paddle/operators/net_op.h"
@@ -42,7 +42,7 @@ static std::unordered_set<std::string>& CtrlFlowOps() {
 static inline std::unique_ptr<OperatorBase> CreateGradOp(
     const OperatorBase& op, const std::unordered_set<std::string>& no_grad_set,
     std::unordered_map<std::string, std::string>* grad_to_var) {
-  OpDescBind op_desc;
+  OpDesc op_desc;
   op_desc.SetInputMap(op.Inputs());
   op_desc.SetOutputMap(op.Outputs());
   op_desc.SetType(op.Type());
@@ -53,7 +53,7 @@ static inline std::unique_ptr<OperatorBase> CreateGradOp(
   grad_ops.reserve(grad_descs.size());
   std::transform(grad_descs.begin(), grad_descs.end(),
                  std::back_inserter(grad_ops),
-                 [](const std::unique_ptr<OpDescBind>& grad_desc) {
+                 [](const std::unique_ptr<OpDesc>& grad_desc) {
                    return OpRegistry::CreateOp(*grad_desc);
                  });
   PADDLE_ENFORCE(!grad_ops.empty());
@@ -217,7 +217,7 @@ static std::unique_ptr<OperatorBase> BackwardRecursive(
         // If part of input gradient of that operator is not calculated, fill
         // zero variables to that input gradient.
         net->AppendOp(OpRegistry::CreateOp("fill_zeros_like", {{"X", {prefix}}},
-                                           {{"Y", {grad_input}}},
+                                           {{"Out", {grad_input}}},
                                            AttributeMap{}));
       }
       return false;
@@ -296,7 +296,7 @@ static std::string FwdName(const std::string& grad_name) {
 static void CreateGradVarInBlock(
     size_t grad_op_start_index,
     const std::unordered_map<std::string, std::string>& param_name_map,
-    BlockDescBind* block_desc,
+    BlockDesc* block_desc,
     std::unordered_map<std::string, GradVarInfo>* grad_var_record) {
   auto ops = block_desc->AllOps();
   for (size_t op_index = grad_op_start_index; op_index < ops.size();
@@ -341,7 +341,7 @@ static void CreateGradVarInBlock(
       auto* param = block_desc->FindVarRecursive(pname);
       auto* grad = block_desc->FindVar(arg);
       if (param == nullptr) {
-        grad->SetDataType(DataType::FP32);
+        grad->SetDataType(proto::DataType::FP32);
       } else {
         grad->SetDataType(param->GetDataType());
       }
@@ -350,12 +350,11 @@ static void CreateGradVarInBlock(
   }
 }
 
-std::vector<std::unique_ptr<OpDescBind>> MakeOpGrad(
-    const OpDescBind* op_desc, std::unordered_set<std::string>* no_grad_vars,
+std::vector<std::unique_ptr<OpDesc>> MakeOpGrad(
+    const OpDesc* op_desc, std::unordered_set<std::string>* no_grad_vars,
     std::unordered_map<std::string, std::string>* grad_to_var,
-    const std::vector<BlockDescBind*>& grad_block =
-        std::vector<BlockDescBind*>()) {
-  std::vector<std::unique_ptr<OpDescBind>> grad_op_descs;
+    const std::vector<BlockDesc*>& grad_block = std::vector<BlockDesc*>()) {
+  std::vector<std::unique_ptr<OpDesc>> grad_op_descs;
   // All input gradients of forwarding operator do not need to calculate.
   const std::vector<std::string>& inputs = op_desc->InputArgumentNames();
   if (AllGradInSet(inputs, *no_grad_vars)) {
@@ -386,7 +385,7 @@ std::vector<std::unique_ptr<OpDescBind>> MakeOpGrad(
           .Get(op_desc->Type())
           .GradOpMaker()(*op_desc, *no_grad_vars, grad_to_var, grad_block);
 
-  std::list<std::unique_ptr<OpDescBind>> pending_fill_zeros_ops;
+  std::list<std::unique_ptr<OpDesc>> pending_fill_zeros_ops;
   for (auto& desc : grad_op_descs) {
     for (const std::string& in_name : desc->InputArgumentNames()) {
       if (no_grad_vars->count(in_name)) {
@@ -394,9 +393,9 @@ std::vector<std::unique_ptr<OpDescBind>> MakeOpGrad(
             0, in_name.size() - sizeof(kGradVarSuffix) / sizeof(char) + 1);
         std::string new_name = prefix + kZeroVarSuffix;
         desc->Rename(in_name, new_name);
-        std::unique_ptr<OpDescBind> fill_zeros_op(
-            new OpDescBind("fill_zeros_like", {{"X", {prefix}}},
-                           {{"Y", {new_name}}}, AttributeMap{}));
+        std::unique_ptr<OpDesc> fill_zeros_op(
+            new OpDesc("fill_zeros_like", {{"X", {prefix}}},
+                       {{"Out", {new_name}}}, AttributeMap{}));
         pending_fill_zeros_ops.push_back(std::move(fill_zeros_op));
       }
     }
@@ -408,34 +407,33 @@ std::vector<std::unique_ptr<OpDescBind>> MakeOpGrad(
   return grad_op_descs;
 }
 
-static BlockDescBind* CreateStepBlock(
-    ProgramDescBind& program_desc,
-    std::unordered_set<std::string>* no_grad_vars,
+static BlockDesc* CreateStepBlock(
+    ProgramDesc& program_desc, std::unordered_set<std::string>* no_grad_vars,
     std::unordered_map<std::string, std::string>* grad_to_var,
     int step_block_idx);
 
-std::vector<std::unique_ptr<OpDescBind>> MakeBlockBackward(
-    ProgramDescBind& program_desc, int block_idx,
+std::vector<std::unique_ptr<OpDesc>> MakeBlockBackward(
+    ProgramDesc& program_desc, int block_idx,
     std::unordered_set<std::string>* no_grad_vars,
     std::unordered_map<std::string, std::string>* grad_to_var) {
   VLOG(5) << "MakeBlockBackward";
-  BlockDescBind* cur_block = program_desc.MutableBlock(block_idx);
-  std::vector<OpDescBind*> op_descs = cur_block->AllOps();
+  BlockDesc* cur_block = program_desc.MutableBlock(block_idx);
+  std::vector<OpDesc*> op_descs = cur_block->AllOps();
   std::unordered_map<std::string, std::vector<size_t>> dup_out_ops;
   size_t grad_desc_idx = 0;
-  std::vector<std::unique_ptr<OpDescBind>> backward_descs;
+  std::vector<std::unique_ptr<OpDesc>> backward_descs;
 
   for (auto it = op_descs.rbegin(); it != op_descs.rend(); ++it) {
     VLOG(5) << "Making backward " << (*it)->Type() << " op";
-    std::vector<std::unique_ptr<OpDescBind>> op_grads;
+    std::vector<std::unique_ptr<OpDesc>> op_grads;
 
     if ((*it)->Type() == "recurrent" || (*it)->Type() == "while") {
       int step_block_idx = (*it)->GetBlockAttr("sub_block");
-      BlockDescBind* backward_block = CreateStepBlock(
-          program_desc, no_grad_vars, grad_to_var, step_block_idx);
+      BlockDesc* backward_block = CreateStepBlock(program_desc, no_grad_vars,
+                                                  grad_to_var, step_block_idx);
       op_grads = MakeOpGrad(*it, no_grad_vars, grad_to_var, {backward_block});
     } else if ((*it)->Type() == "conditional_block") {
-      BlockDescBind* backward_block =
+      BlockDesc* backward_block =
           CreateStepBlock(program_desc, no_grad_vars, grad_to_var,
                           (*it)->GetBlockAttr("sub_block"));
       op_grads = MakeOpGrad(*it, no_grad_vars, grad_to_var, {backward_block});
@@ -463,14 +461,14 @@ std::vector<std::unique_ptr<OpDescBind>> MakeBlockBackward(
       }
       ++grad_desc_idx;
     }
-    std::transform(
-        op_grads.begin(), op_grads.end(), std::back_inserter(backward_descs),
-        [](std::unique_ptr<OpDescBind>& ptr) { return std::move(ptr); });
+    std::transform(op_grads.begin(), op_grads.end(),
+                   std::back_inserter(backward_descs),
+                   [](std::unique_ptr<OpDesc>& ptr) { return std::move(ptr); });
   }
 
   VLOG(5) << "Appending Sums";
   // Check whether some variables are written more than once
-  std::list<std::pair<size_t, std::unique_ptr<OpDescBind>>> pending_sum_ops;
+  std::list<std::pair<size_t, std::unique_ptr<OpDesc>>> pending_sum_ops;
   for (const auto& dup : dup_out_ops) {
     const std::string& out_name = dup.first;
     const std::vector<size_t> dup_op = dup.second;
@@ -486,18 +484,17 @@ std::vector<std::unique_ptr<OpDescBind>> MakeBlockBackward(
         sum_op_inputs.emplace_back(new_name);
         next_g_name = sum_op_inputs.back();
       }
-      std::unique_ptr<OpDescBind> sum_op(
-          new OpDescBind("sum", {{"X", sum_op_inputs}}, {{"Out", {out_name}}},
-                         AttributeMap{}));
+      std::unique_ptr<OpDesc> sum_op(new OpDesc("sum", {{"X", sum_op_inputs}},
+                                                {{"Out", {out_name}}},
+                                                AttributeMap{}));
       pending_sum_ops.push_back({dup_op.back(), std::move(sum_op)});
     }
   }
 
-  pending_sum_ops.sort(
-      [](const std::pair<size_t, std::unique_ptr<OpDescBind>>& a,
-         const std::pair<size_t, std::unique_ptr<OpDescBind>>& b) {
-        return a.first > b.first;
-      });
+  pending_sum_ops.sort([](const std::pair<size_t, std::unique_ptr<OpDesc>>& a,
+                          const std::pair<size_t, std::unique_ptr<OpDesc>>& b) {
+    return a.first > b.first;
+  });
   for (auto& p : pending_sum_ops) {
     backward_descs.insert(backward_descs.begin() + p.first + 1,
                           std::move(p.second));
@@ -508,14 +505,13 @@ std::vector<std::unique_ptr<OpDescBind>> MakeBlockBackward(
   return backward_descs;
 }
 
-static BlockDescBind* CreateStepBlock(
-    ProgramDescBind& program_desc,
-    std::unordered_set<std::string>* no_grad_vars,
+static BlockDesc* CreateStepBlock(
+    ProgramDesc& program_desc, std::unordered_set<std::string>* no_grad_vars,
     std::unordered_map<std::string, std::string>* grad_to_var,
     int step_block_idx) {
   auto backward_block_op_descs = MakeBlockBackward(program_desc, step_block_idx,
                                                    no_grad_vars, grad_to_var);
-  BlockDescBind* backward_block =
+  BlockDesc* backward_block =
       program_desc.AppendBlock(*program_desc.MutableBlock(step_block_idx));
   for (auto& ptr : backward_block_op_descs) {
     backward_block->AppendAllocatedOp(move(ptr));
@@ -524,7 +520,7 @@ static BlockDescBind* CreateStepBlock(
 }
 
 ParamGradInfoMap AppendBackward(
-    ProgramDescBind& program_desc, const VarDescBind& target,
+    ProgramDesc& program_desc, const VarDesc& target,
     const std::unordered_set<std::string>& no_grad_vars) {
   std::unordered_set<std::string> no_grad_var_names;
   no_grad_var_names.reserve(no_grad_vars.size() + 1);
@@ -541,11 +537,11 @@ ParamGradInfoMap AppendBackward(
   PADDLE_ENFORCE(is_scalar, "target should be scalar");
   VLOG(3) << "backward from loss=" << target.Name()
           << " data_type=" << target.GetDataType();
-  std::unique_ptr<OpDescBind> fill_one_op(
-      new OpDescBind("fill_constant", {}, {{"Out", {fill_one_op_out}}},
-                     {{"shape", std::vector<int>{1}},
-                      {"value", static_cast<float>(1.0)},
-                      {"dtype", target.GetDataType()}}));
+  std::unique_ptr<OpDesc> fill_one_op(
+      new OpDesc("fill_constant", {}, {{"Out", {fill_one_op_out}}},
+                 {{"shape", std::vector<int>{1}},
+                  {"value", static_cast<float>(1.0)},
+                  {"dtype", target.GetDataType()}}));
   // infer var type of fill_one_op
   fill_one_op->InferVarType(root_block);
 

paddle/framework/backward.h

 /* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
 
-   Licensed under the Apache License, Version 2.0 (the "License");
-   you may not use this file except in compliance with the License.
-   You may obtain a copy of the License at
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
 
-   http://www.apache.org/licenses/LICENSE-2.0
+    http://www.apache.org/licenses/LICENSE-2.0
 
-   Unless required by applicable law or agreed to in writing, software
-   distributed under the License is distributed on an "AS IS" BASIS,
-   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-   See the License for the specific language governing permissions and
-   limitations under the License. */
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
 
 #pragma once
 
@@ -49,7 +49,7 @@ using ParamGradInfoMap = std::unordered_map<std::string /*fwd_var_name*/,
                                             GradVarInfo /*grad_var_info*/>;
 
 ParamGradInfoMap AppendBackward(
-    ProgramDescBind& program_desc, const VarDescBind& target,
+    ProgramDesc& program_desc, const VarDesc& target,
     const std::unordered_set<std::string>& no_grad_vars);
 
 }  // namespace framework

paddle/framework/backward_test.cc

 /* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
 
-   Licensed under the Apache License, Version 2.0 (the "License");
-   you may not use this file except in compliance with the License.
-   You may obtain a copy of the License at
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
 
-   http://www.apache.org/licenses/LICENSE-2.0
+    http://www.apache.org/licenses/LICENSE-2.0
 
-   Unless required by applicable law or agreed to in writing, software
-   distributed under the License is distributed on an "AS IS" BASIS,
-   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-   See the License for the specific language governing permissions and
-   limitations under the License. */
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
 
 #include "paddle/framework/backward.h"
 
@@ -58,13 +58,13 @@ class RowWiseAddGradMaker : public SingleGradOpDescMaker {
   using SingleGradOpDescMaker::SingleGradOpDescMaker;
 
  protected:
-  std::unique_ptr<OpDescBind> Apply() const override {
-    auto grad_op = new OpDescBind();
+  std::unique_ptr<OpDesc> Apply() const override {
+    auto grad_op = new OpDesc();
     grad_op->SetInput(GradVarName("Out"), OutputGrad("Out"));
     grad_op->SetOutput(GradVarName("X"), InputGrad("X"));
     grad_op->SetOutput(GradVarName("b"), InputGrad("b"));
     grad_op->SetType("rowwise_add_grad");
-    return std::unique_ptr<OpDescBind>(grad_op);
+    return std::unique_ptr<OpDesc>(grad_op);
   }
 };
 
@@ -159,14 +159,14 @@ class FillZeroOpMaker : public OpProtoAndCheckerMaker {
   FillZeroOpMaker(OpProto *proto, OpAttrChecker *op_checker)
       : OpProtoAndCheckerMaker(proto, op_checker) {
     AddInput("X", "x");
-    AddOutput("Y", "out");
+    AddOutput("Out", "out");
     AddComment("");
   }
 };
 
 class SumOpMaker : public framework::OpProtoAndCheckerMaker {
  public:
-  SumOpMaker(framework::OpProto *proto, framework::OpAttrChecker *op_checker)
+  SumOpMaker(OpProto *proto, OpAttrChecker *op_checker)
       : OpProtoAndCheckerMaker(proto, op_checker) {
     AddInput("X", "the input tensors of sum operator.").AsDuplicable();
     AddOutput("Out", "the output tensor of sum operator.");
@@ -190,11 +190,11 @@ class MinusGradOpDescMaker : public GradOpDescMakerBase {
  public:
   using GradOpDescMakerBase::GradOpDescMakerBase;
 
-  std::vector<std::unique_ptr<OpDescBind>> operator()() const override {
-    std::vector<std::unique_ptr<OpDescBind>> retv;
+  std::vector<std::unique_ptr<OpDesc>> operator()() const override {
+    std::vector<std::unique_ptr<OpDesc>> retv;
     auto x_g = InputGrad("X");
     if (!x_g.empty()) {
-      auto *op_desc = new OpDescBind();
+      auto *op_desc = new OpDesc();
       op_desc->SetType("scale");
       op_desc->SetInput("X", OutputGrad("Out"));
       op_desc->SetOutput("Out", x_g);
@@ -204,7 +204,7 @@ class MinusGradOpDescMaker : public GradOpDescMakerBase {
 
     auto y_g = InputGrad("Y");
     if (!y_g.empty()) {
-      auto *op_desc = new OpDescBind();
+      auto *op_desc = new OpDesc();
       op_desc->SetType("scale");
       op_desc->SetInput("X", OutputGrad("Out"));
       op_desc->SetOutput("Out", y_g);
@@ -430,8 +430,8 @@ TEST(Backward, op_part_of_output_are_not_need) {
   ASSERT_EQ("fill_zeros_like", fill_zero.Type());
   ASSERT_EQ(1UL, fill_zero.Inputs("X").size());
   ASSERT_EQ("Z", fill_zero.Input("X"));
-  ASSERT_EQ(1UL, fill_zero.Outputs("Y").size());
-  ASSERT_EQ(std::string("Z") + f::kZeroVarSuffix, fill_zero.Output("Y"));
+  ASSERT_EQ(1UL, fill_zero.Outputs("Out").size());
+  ASSERT_EQ(std::string("Z") + f::kZeroVarSuffix, fill_zero.Output("Out"));
 
   auto &d_many_out = *net->ops_[1];
   ASSERT_EQ("many_output_op_grad", d_many_out.Type());
@@ -505,25 +505,25 @@ TEST(Backward, linear_net_intermediate_variable_has_no_grad) {
 }
 
 TEST(Backward, simple_single_op) {
-  f::ProgramDescBind program;
-  f::BlockDescBind *block = program.MutableBlock(0);
+  f::ProgramDesc program;
+  f::BlockDesc *block = program.MutableBlock(0);
 
-  f::OpDescBind *op = block->AppendOp();
+  f::OpDesc *op = block->AppendOp();
   op->SetType("rowwise_add");
   op->SetInput("X", {"x"});
   op->SetInput("b", {"b"});
   op->SetOutput("Out", {"out"});
 
-  auto target = f::VarDescBind("out");
+  auto target = f::VarDesc("out");
   target.SetShape({1});
   auto var_to_grad =
       AppendBackward(program, target, std::unordered_set<std::string>{});
 
   ASSERT_EQ(block->AllOps().size(), 3UL);
-  f::OpDescBind *fill_op = block->AllOps()[1];
+  f::OpDesc *fill_op = block->AllOps()[1];
   EXPECT_EQ(fill_op->Type(), "fill_constant");
 
-  f::OpDescBind *grad_op = block->AllOps()[2];
+  f::OpDesc *grad_op = block->AllOps()[2];
   EXPECT_EQ(grad_op->Type(), "rowwise_add_grad");
   ASSERT_EQ(grad_op->InputNames().size(), 1UL);
   ASSERT_EQ(grad_op->OutputNames().size(), 2UL);
@@ -543,16 +543,16 @@ TEST(Backward, simple_single_op) {
 }
 
 TEST(Backward, default_attribute) {
-  f::ProgramDescBind program;
-  f::BlockDescBind *block = program.MutableBlock(0);
-  f::OpDescBind *op = block->AppendOp();
+  f::ProgramDesc program;
+  f::BlockDesc *block = program.MutableBlock(0);
+  f::OpDesc *op = block->AppendOp();
   op->SetType("mul");
   op->SetInput("X", {"x"});
   op->SetInput("Y", {"y"});
   op->SetOutput("Out", {"out"});
   op->CheckAttrs();
 
-  auto target = f::VarDescBind("out");
+  auto target = f::VarDesc("out");
   target.SetShape({1});
   AppendBackward(program, target, std::unordered_set<std::string>{});
 
@@ -560,47 +560,47 @@ TEST(Backward, default_attribute) {
   EXPECT_EQ(boost::get<int>(op->GetAttr("x_num_col_dims")), 1);
   EXPECT_EQ(boost::get<int>(op->GetAttr("y_num_col_dims")), 1);
 
-  f::OpDescBind *fill_op = block->AllOps()[1];
+  f::OpDesc *fill_op = block->AllOps()[1];
   EXPECT_EQ(fill_op->Type(), "fill_constant");
 
-  f::OpDescBind *grad_op = block->AllOps()[2];
+  f::OpDesc *grad_op = block->AllOps()[2];
   ASSERT_EQ(grad_op->Type(), "mul_grad");
   EXPECT_EQ(boost::get<int>(grad_op->GetAttr("x_num_col_dims")), 1);
   EXPECT_EQ(boost::get<int>(grad_op->GetAttr("y_num_col_dims")), 1);
 }
 
 TEST(Backward, simple_mult_op) {
-  f::ProgramDescBind program;
-  f::BlockDescBind *block = program.MutableBlock(0);
-  f::OpDescBind *op1 = block->AppendOp();
+  f::ProgramDesc program;
+  f::BlockDesc *block = program.MutableBlock(0);
+  f::OpDesc *op1 = block->AppendOp();
   op1->SetType("rowwise_add");
   op1->SetInput("X", {"x1"});
   op1->SetInput("b", {"b1"});
   op1->SetOutput("Out", {"out1"});
 
-  f::OpDescBind *op2 = block->AppendOp();
+  f::OpDesc *op2 = block->AppendOp();
   op2->SetType("mul");
   op2->SetInput("X", {"out1"});
   op2->SetInput("Y", {"y2"});
   op2->SetOutput("Out", {"out2"});
 
-  f::OpDescBind *op3 = block->AppendOp();
+  f::OpDesc *op3 = block->AppendOp();
   op3->SetType("rowwise_add");
   op3->SetInput("X", {"out2"});
   op3->SetInput("b", {"b3"});
   op3->SetOutput("Out", {"out3"});
 
-  auto target = f::VarDescBind("out3");
+  auto target = f::VarDesc("out3");
   target.SetShape({1});
   size_t forward_len = block->AllOps().size();
   auto var_to_grad =
       AppendBackward(program, target, std::unordered_set<std::string>{});
 
   ASSERT_EQ(block->AllOps().size(), 6UL + 1);
-  f::OpDescBind *fill_op = block->AllOps()[forward_len];
+  f::OpDesc *fill_op = block->AllOps()[forward_len];
   EXPECT_EQ(fill_op->Type(), "fill_constant");
 
-  f::OpDescBind *grad_op1 = block->AllOps()[6];
+  f::OpDesc *grad_op1 = block->AllOps()[6];
   EXPECT_EQ(grad_op1->Type(), "rowwise_add_grad");
   ASSERT_EQ(grad_op1->InputNames().size(), 1UL);
   ASSERT_EQ(grad_op1->OutputNames().size(), 2UL);
@@ -611,7 +611,7 @@ TEST(Backward, simple_mult_op) {
   EXPECT_EQ(grad_op1->Output(f::GradVarName("b")),
             std::vector<std::string>({f::GradVarName("b1")}));
 
-  f::OpDescBind *grad_op2 = block->AllOps()[5];
+  f::OpDesc *grad_op2 = block->AllOps()[5];
   EXPECT_EQ(grad_op2->Type(), "mul_grad");
   ASSERT_EQ(grad_op2->InputNames().size(), 4UL);
   ASSERT_EQ(grad_op2->OutputNames().size(), 2UL);
@@ -625,7 +625,7 @@ TEST(Backward, simple_mult_op) {
   EXPECT_EQ(grad_op2->Output(f::GradVarName("Y")),
             std::vector<std::string>({f::GradVarName("y2")}));
 
-  f::OpDescBind *grad_op3 = block->AllOps()[4];
+  f::OpDesc *grad_op3 = block->AllOps()[4];
   EXPECT_EQ(grad_op3->Type(), "rowwise_add_grad");
   ASSERT_EQ(grad_op3->InputNames().size(), 1UL);
   ASSERT_EQ(grad_op3->OutputNames().size(), 2UL);
@@ -655,42 +655,42 @@ TEST(Backward, simple_mult_op) {
 }
 
 TEST(Backward, intermedia_var_no_grad) {
-  f::ProgramDescBind program;
-  f::BlockDescBind *block = program.MutableBlock(0);
-  f::OpDescBind *op1 = block->AppendOp();
+  f::ProgramDesc program;
+  f::BlockDesc *block = program.MutableBlock(0);
+  f::OpDesc *op1 = block->AppendOp();
   op1->SetType("rowwise_add");
   op1->SetInput("X", {"x1"});
   op1->SetInput("b", {"b1"});
   op1->SetOutput("Out", {"out1"});
 
-  f::OpDescBind *op2 = block->AppendOp();
+  f::OpDesc *op2 = block->AppendOp();
   op2->SetType("mul");
   op2->SetInput("X", {"x2"});
   op2->SetInput("Y", {"y2"});
   op2->SetOutput("Out", {"out2"});
 
-  f::OpDescBind *op3 = block->AppendOp();
+  f::OpDesc *op3 = block->AppendOp();
   op3->SetType("rowwise_add");
   op3->SetInput("X", {"out2"});
   op3->SetInput("b", {"b3"});
   op3->SetOutput("Out", {"out3"});
 
-  f::OpDescBind *op4 = block->AppendOp();
+  f::OpDesc *op4 = block->AppendOp();
   op4->SetType("mul");
   op4->SetInput("X", {"out1"});
   op4->SetInput("Y", {"out3"});
   op4->SetOutput("Out", {"out4"});
 
-  auto target = f::VarDescBind("out4");
+  auto target = f::VarDesc("out4");
   target.SetShape({1});
   size_t forward_len = block->AllOps().size();
   auto var_to_grad = AppendBackward(program, target, {"out3"});
 
   ASSERT_EQ(block->AllOps().size(), 7UL);
-  f::OpDescBind *fill_op = block->AllOps()[forward_len];
+  f::OpDesc *fill_op = block->AllOps()[forward_len];
   EXPECT_EQ(fill_op->Type(), "fill_constant");
 
-  f::OpDescBind *grad_op1 = block->AllOps()[6];
+  f::OpDesc *grad_op1 = block->AllOps()[6];
   EXPECT_EQ(grad_op1->Type(), "rowwise_add_grad");
   ASSERT_EQ(grad_op1->InputNames().size(), 1UL);
   ASSERT_EQ(grad_op1->OutputNames().size(), 2UL);
@@ -701,7 +701,7 @@ TEST(Backward, intermedia_var_no_grad) {
   EXPECT_EQ(grad_op1->Output(f::GradVarName("b")),
             std::vector<std::string>({f::GradVarName("b1")}));
 
-  f::OpDescBind *grad_op4 = block->AllOps()[5];
+  f::OpDesc *grad_op4 = block->AllOps()[5];
   EXPECT_EQ(grad_op4->Type(), "mul_grad");
   ASSERT_EQ(grad_op4->InputNames().size(), 4UL);
   ASSERT_EQ(grad_op4->OutputNames().size(), 2UL);
@@ -726,32 +726,32 @@ TEST(Backward, intermedia_var_no_grad) {
 }
 
 TEST(Backward, var_no_grad) {
-  f::ProgramDescBind program;
-  f::BlockDescBind *block = program.MutableBlock(0);
-  f::OpDescBind *op1 = block->AppendOp();
+  f::ProgramDesc program;
+  f::BlockDesc *block = program.MutableBlock(0);
+  f::OpDesc *op1 = block->AppendOp();
   op1->SetType("mult_in_out");
   op1->SetInput("X", {"x1"});
   op1->SetInput("H", {"h1"});
   op1->SetOutput("Y", {"y1"});
   op1->SetOutput("Z", {"z1"});
 
-  f::OpDescBind *op2 = block->AppendOp();
+  f::OpDesc *op2 = block->AppendOp();
   op2->SetType("mult_in_out");
   op2->SetInput("X", {"y1"});
   op2->SetInput("H", {"z1"});
   op2->SetOutput("Y", {"y2"});
   op2->SetOutput("Z", {"z2"});
 
-  auto target = f::VarDescBind("z2");
+  auto target = f::VarDesc("z2");
   target.SetShape({1});
   size_t forward_len = block->AllOps().size();
   auto var_to_grad = AppendBackward(program, target, {"z1"});
 
   ASSERT_EQ(block->AllOps().size(), 6UL);
-  f::OpDescBind *fill_op = block->AllOps()[forward_len];
+  f::OpDesc *fill_op = block->AllOps()[forward_len];
   EXPECT_EQ(fill_op->Type(), "fill_constant");
 
-  f::OpDescBind *grad_op2 = block->AllOps()[3];
+  f::OpDesc *grad_op2 = block->AllOps()[3];
   ASSERT_EQ(grad_op2->Type(), "mult_in_out_grad");
   ASSERT_EQ(grad_op2->InputNames().size(), 6UL);
   ASSERT_EQ(grad_op2->OutputNames().size(), 2UL);
@@ -767,15 +767,15 @@ TEST(Backward, var_no_grad) {
             std::vector<std::string>({f::GradVarName("y1")}));
   EXPECT_EQ(grad_op2->Output(f::GradVarName("H")), std::vector<std::string>());
 
-  f::OpDescBind *fill_zero_op = block->AllOps()[4];
+  f::OpDesc *fill_zero_op = block->AllOps()[4];
   ASSERT_EQ(fill_zero_op->Type(), "fill_zeros_like");
   ASSERT_EQ(fill_zero_op->InputNames().size(), 1UL);
   ASSERT_EQ(fill_zero_op->OutputNames().size(), 1UL);
   EXPECT_EQ(fill_zero_op->Input("X"), std::vector<std::string>({"z1"}));
-  EXPECT_EQ(fill_zero_op->Output("Y"),
+  EXPECT_EQ(fill_zero_op->Output("Out"),
             std::vector<std::string>({std::string("z1") + f::kZeroVarSuffix}));
 
-  f::OpDescBind *grad_op1 = block->AllOps()[5];
+  f::OpDesc *grad_op1 = block->AllOps()[5];
   ASSERT_EQ(grad_op1->Type(), "mult_in_out_grad");
   ASSERT_EQ(grad_op1->InputNames().size(), 6UL);
   ASSERT_EQ(grad_op1->OutputNames().size(), 2UL);
@@ -803,37 +803,37 @@ TEST(Backward, var_no_grad) {
 }
 
 TEST(Backward, shared_var) {
-  f::ProgramDescBind program;
-  f::BlockDescBind *block = program.MutableBlock(0);
-  f::OpDescBind *op1 = block->AppendOp();
+  f::ProgramDesc program;
+  f::BlockDesc *block = program.MutableBlock(0);
+  f::OpDesc *op1 = block->AppendOp();
   op1->SetType("rowwise_add");
   op1->SetInput("X", {"x1"});
   op1->SetInput("b", {"b1"});
   op1->SetOutput("Out", {"out1"});
 
-  f::OpDescBind *op2 = block->AppendOp();
+  f::OpDesc *op2 = block->AppendOp();
   op2->SetType("mul");
   op2->SetInput("X", {"out1"});
   op2->SetInput("Y", {"y2"});
   op2->SetOutput("Out", {"out2"});
 
-  f::OpDescBind *op3 = block->AppendOp();
+  f::OpDesc *op3 = block->AppendOp();
   op3->SetType("rowwise_add");
   op3->SetInput("X", {"out1"});
   op3->SetInput("b", {"b3"});
   op3->SetOutput("Out", {"out3"});
 
-  auto target = f::VarDescBind("out3");
+  auto target = f::VarDesc("out3");
   target.SetShape({1});
   size_t forward_len = block->AllOps().size();
   auto var_to_grad =
       AppendBackward(program, target, std::unordered_set<std::string>{});
 
   ASSERT_EQ(block->AllOps().size(), 8UL);
-  f::OpDescBind *fill_op = block->AllOps()[forward_len];
+  f::OpDesc *fill_op = block->AllOps()[forward_len];
   EXPECT_EQ(fill_op->Type(), "fill_constant");
 
-  f::OpDescBind *grad_op3 = block->AllOps()[4];
+  f::OpDesc *grad_op3 = block->AllOps()[4];
   ASSERT_EQ(grad_op3->Type(), "rowwise_add_grad");
   ASSERT_EQ(grad_op3->InputNames().size(), 1UL);
   ASSERT_EQ(grad_op3->OutputNames().size(), 2UL);
@@ -844,7 +844,7 @@ TEST(Backward, shared_var) {
   EXPECT_EQ(grad_op3->Output(f::GradVarName("b")),
             std::vector<std::string>({f::GradVarName("b3")}));
 
-  f::OpDescBind *grad_op4 = block->AllOps()[5];
+  f::OpDesc *grad_op4 = block->AllOps()[5];
   ASSERT_EQ(grad_op4->Type(), "mul_grad");
   ASSERT_EQ(grad_op4->InputNames().size(), 4UL);
   ASSERT_EQ(grad_op4->OutputNames().size(), 2UL);
@@ -858,7 +858,7 @@ TEST(Backward, shared_var) {
   EXPECT_EQ(grad_op4->Output(f::GradVarName("Y")),
             std::vector<std::string>({f::GradVarName("y2")}));
 
-  f::OpDescBind *sum_op = block->AllOps()[6];
+  f::OpDesc *sum_op = block->AllOps()[6];
   ASSERT_EQ(sum_op->Type(), "sum");
   ASSERT_EQ(sum_op->InputNames().size(), 1UL);
   ASSERT_EQ(sum_op->OutputNames().size(), 1UL);
@@ -868,7 +868,7 @@ TEST(Backward, shared_var) {
   EXPECT_EQ(sum_op->Output("Out"),
             std::vector<std::string>({f::GradVarName("out1")}));
 
-  f::OpDescBind *grad_op1 = block->AllOps()[7];
+  f::OpDesc *grad_op1 = block->AllOps()[7];
   ASSERT_EQ(grad_op1->Type(), "rowwise_add_grad");
   ASSERT_EQ(grad_op1->InputNames().size(), 1UL);
   ASSERT_EQ(grad_op1->OutputNames().size(), 2UL);
@@ -895,19 +895,19 @@ TEST(Backward, shared_var) {
 }
 
 TEST(Backward, half_backward) {
-  f::ProgramDescBind program;
-  f::BlockDescBind *block = program.MutableBlock(0);
+  f::ProgramDesc program;
+  f::BlockDesc *block = program.MutableBlock(0);
   auto *op1 = block->AppendOp();
   op1->SetType("minus");
   op1->SetInput("X", {"a"});
   op1->SetInput("Y", {"b"});
   op1->SetOutput("Out", {"out"});
 
-  auto target = f::VarDescBind("out");
+  auto target = f::VarDesc("out");
   target.SetShape({1});
   size_t forward_len = block->AllOps().size();
   auto var_to_grad = AppendBackward(program, target, {"b"});
-  f::OpDescBind *fill_op = block->AllOps()[forward_len];
+  f::OpDesc *fill_op = block->AllOps()[forward_len];
   EXPECT_EQ(fill_op->Type(), "fill_constant");
   auto ops = block->AllOps();
   ASSERT_EQ(3UL, ops.size());

paddle/framework/block_desc.cc

@@ -19,18 +19,18 @@ limitations under the License. */
 namespace paddle {
 namespace framework {
 
-VarDescBind *BlockDescBind::Var(const std::string &name) {
+VarDesc *BlockDesc::Var(const std::string &name) {
   auto it = vars_.find(name);
   if (it != vars_.end()) {
     return it->second.get();
   }
   need_update_ = true;
-  auto *var = new VarDescBind(name);
+  auto *var = new VarDesc(name);
   vars_[name].reset(var);
   return var;
 }
 
-VarDescBind *BlockDescBind::FindVar(const std::string &name) const {
+VarDesc *BlockDesc::FindVar(const std::string &name) const {
   auto it = vars_.find(name);
   if (it == vars_.end()) {
     return nullptr;
@@ -38,11 +38,11 @@ VarDescBind *BlockDescBind::FindVar(const std::string &name) const {
   return it->second.get();
 }
 
-bool BlockDescBind::HasVar(const std::string &name) const {
+bool BlockDesc::HasVar(const std::string &name) const {
   return vars_.find(name) != vars_.end();
 }
 
-VarDescBind *BlockDescBind::FindVarRecursive(const std::string &name) const {
+VarDesc *BlockDesc::FindVarRecursive(const std::string &name) const {
   if (name == kEmptyVarName) return nullptr;
 
   auto it = vars_.find(name);
@@ -53,53 +53,67 @@ VarDescBind *BlockDescBind::FindVarRecursive(const std::string &name) const {
   return it->second.get();
 }
 
-VarDescBind *BlockDescBind::FindRecursiveOrCreateVar(
-    const std::string &name_bytes) {
-  VarDescBind *res = FindVarRecursive(name_bytes);
+VarDesc *BlockDesc::FindRecursiveOrCreateVar(const std::string &name_bytes) {
+  VarDesc *res = FindVarRecursive(name_bytes);
   if (res == nullptr) {
     res = Var(name_bytes);
   }
   return res;
 }
 
-bool BlockDescBind::HasVarRecursive(const std::string &name) const {
+bool BlockDesc::HasVarRecursive(const std::string &name) const {
   return FindVarRecursive(name) != nullptr;
 }
 
-std::vector<VarDescBind *> BlockDescBind::AllVars() const {
-  std::vector<VarDescBind *> res;
+std::vector<VarDesc *> BlockDesc::AllVars() const {
+  std::vector<VarDesc *> res;
   for (const auto &p : vars_) {
     res.push_back(p.second.get());
   }
   return res;
 }
 
-OpDescBind *BlockDescBind::AppendOp() {
+OpDesc *BlockDesc::AppendOp() {
   need_update_ = true;
-  ops_.emplace_back(new OpDescBind());
+  ops_.emplace_back(new OpDesc());
   return ops_.back().get();
 }
 
-void BlockDescBind::AppendAllocatedOp(std::unique_ptr<OpDescBind> &&op_desc) {
+void BlockDesc::AppendAllocatedOp(std::unique_ptr<OpDesc> &&op_desc) {
   need_update_ = true;
   ops_.emplace_back(std::move(op_desc));
 }
 
-OpDescBind *BlockDescBind::PrependOp() {
+OpDesc *BlockDesc::PrependOp() {
   need_update_ = true;
-  ops_.emplace_front(new OpDescBind());
+  ops_.emplace_front(new OpDesc());
   return ops_.front().get();
 }
 
-std::vector<OpDescBind *> BlockDescBind::AllOps() const {
-  std::vector<OpDescBind *> res;
+void BlockDesc::RemoveOp(size_t s, size_t e) {
+  if (ops_.begin() + s == ops_.end() || ops_.begin() + e == ops_.end()) {
+    return;
+  }
+  need_update_ = true;
+  for (auto it = ops_.begin() + s; it != ops_.begin() + e; it++) {
+    auto names = (*it)->InputArgumentNames();
+    for (auto n : names) {
+      // TODO(typhoonzero): delete vars if no other op use it.
+      VLOG(3) << "deleting var " << n;
+    }
+  }
+  ops_.erase(ops_.begin() + s, ops_.begin() + e);
+}
+
+std::vector<OpDesc *> BlockDesc::AllOps() const {
+  std::vector<OpDesc *> res;
   for (const auto &op : ops_) {
     res.push_back(op.get());
   }
   return res;
 }
 
-void BlockDescBind::Flush() {
+void BlockDesc::Flush() {
   for (auto &op_desc : ops_) {
     op_desc->Flush();
   }
@@ -121,43 +135,43 @@ void BlockDescBind::Flush() {
   }
 }
 
-BlockDescBind *BlockDescBind::ParentBlock() const {
+BlockDesc *BlockDesc::ParentBlock() const {
   if (this->desc_->parent_idx() == kNoneBlockIndex) {
     return nullptr;
   }
   return prog_->MutableBlock(static_cast<size_t>(this->desc_->parent_idx()));
 }
 
-BlockDesc *BlockDescBind::Proto() {
+proto::BlockDesc *BlockDesc::Proto() {
   Flush();
   return desc_;
 }
 
-BlockDescBind::BlockDescBind(ProgramDescBind *prog, BlockDesc *desc)
+BlockDesc::BlockDesc(ProgramDesc *prog, proto::BlockDesc *desc)
     : prog_(prog), desc_(desc), need_update_(false) {
-  for (const VarDesc &var_desc : desc_->vars()) {
-    vars_[var_desc.name()].reset(new VarDescBind(var_desc));
+  for (const proto::VarDesc &var_desc : desc_->vars()) {
+    vars_[var_desc.name()].reset(new VarDesc(var_desc));
   }
-  for (const OpDesc &op_desc : desc_->ops()) {
-    ops_.emplace_back(new OpDescBind(op_desc, prog));
+  for (const proto::OpDesc &op_desc : desc_->ops()) {
+    ops_.emplace_back(new OpDesc(op_desc, prog));
   }
 }
 
-BlockDescBind::BlockDescBind(const BlockDescBind &other, BlockDesc *desc,
-                             ProgramDescBind *prog)
+BlockDesc::BlockDesc(const BlockDesc &other, proto::BlockDesc *desc,
+                     ProgramDesc *prog)
     : prog_(prog), desc_(desc) {
   need_update_ = true;
   for (auto &op : other.ops_) {
-    ops_.emplace_back(new OpDescBind(*op));
+    ops_.emplace_back(new OpDesc(*op));
   }
 
   for (auto &it : other.vars_) {
-    auto *var = new VarDescBind(*it.second);
+    auto *var = new VarDesc(*it.second);
     vars_[it.first].reset(var);
   }
 }
 
-void BlockDescBind::ClearPBOps() {
+void BlockDesc::ClearPBOps() {
   auto ops = this->desc_->mutable_ops();
   while (!ops->empty()) {
     // we do not own the OpDesc, so release the ownership.
@@ -165,7 +179,7 @@ void BlockDescBind::ClearPBOps() {
   }
 }
 
-void BlockDescBind::ClearPBVars() {
+void BlockDesc::ClearPBVars() {
   auto vars = this->desc_->mutable_vars();
   while (!vars->empty()) {
     // we do not own the VarDesc, so release the ownership.

paddle/framework/block_desc.h

@@ -28,20 +28,19 @@ limitations under the License. */
 namespace paddle {
 namespace framework {
 
-class ProgramDescBind;
+class ProgramDesc;
 
 // Each Protobuf Message, we provide a XXXBind class. In that class, we optimize
 // read/write speed. Only when we want the protobuf message, the local changes
 // will be synchronized (by `Sync` method).
 
-class BlockDescBind {
+class BlockDesc {
  public:
-  BlockDescBind(ProgramDescBind *prog, BlockDesc *desc);
+  BlockDesc(ProgramDesc *prog, proto::BlockDesc *desc);
 
-  BlockDescBind(const BlockDescBind &other, BlockDesc *desc,
-                ProgramDescBind *prog);
+  BlockDesc(const BlockDesc &other, proto::BlockDesc *desc, ProgramDesc *prog);
 
-  ~BlockDescBind() {
+  ~BlockDesc() {
     this->ClearPBVars();
     this->ClearPBOps();
   }
@@ -50,15 +49,15 @@ class BlockDescBind {
 
   int32_t Parent() const { return desc_->parent_idx(); }
 
-  VarDescBind *Var(const std::string &name_bytes);
+  VarDesc *Var(const std::string &name_bytes);
 
-  VarDescBind *FindVar(const std::string &name_bytes) const;
+  VarDesc *FindVar(const std::string &name_bytes) const;
 
   bool HasVar(const std::string &var_name) const;
 
-  VarDescBind *FindVarRecursive(const std::string &name_bytes) const;
+  VarDesc *FindVarRecursive(const std::string &name_bytes) const;
 
-  VarDescBind *FindRecursiveOrCreateVar(const std::string &name_bytes);
+  VarDesc *FindRecursiveOrCreateVar(const std::string &name_bytes);
 
   bool HasVarRecursive(const std::string &var_name) const;
 
@@ -70,41 +69,43 @@ class BlockDescBind {
     return var_names;
   }
 
-  std::vector<VarDescBind *> AllVars() const;
+  std::vector<VarDesc *> AllVars() const;
 
-  BlockDescBind *ParentBlock() const;
+  BlockDesc *ParentBlock() const;
 
-  OpDescBind *AppendOp();
+  OpDesc *AppendOp();
 
-  void AppendAllocatedOp(std::unique_ptr<OpDescBind> &&op_desc);
+  void AppendAllocatedOp(std::unique_ptr<OpDesc> &&op_desc);
 
-  OpDescBind *PrependOp();
+  OpDesc *PrependOp();
 
-  std::vector<OpDescBind *> AllOps() const;
+  void RemoveOp(size_t s, size_t e);
+
+  std::vector<OpDesc *> AllOps() const;
 
   size_t OpSize() const { return ops_.size(); }
 
-  OpDescBind *Op(int idx) { return ops_.at(idx).get(); }
+  OpDesc *Op(int idx) { return ops_.at(idx).get(); }
 
   void Flush();
 
-  BlockDesc *Proto();
+  proto::BlockDesc *Proto();
 
-  ProgramDescBind *Program() { return this->prog_; }
+  ProgramDesc *Program() { return this->prog_; }
 
  private:
   void ClearPBOps();
   void ClearPBVars();
 
  private:
-  ProgramDescBind *prog_;  // not_own
-  BlockDesc *desc_;        // not_own
+  ProgramDesc *prog_;       // not_own
+  proto::BlockDesc *desc_;  // not_own
   bool need_update_;
 
-  std::deque<std::unique_ptr<OpDescBind>> ops_;
-  std::unordered_map<std::string, std::unique_ptr<VarDescBind>> vars_;
+  std::deque<std::unique_ptr<OpDesc>> ops_;
+  std::unordered_map<std::string, std::unique_ptr<VarDesc>> vars_;
 
-  DISABLE_COPY_AND_ASSIGN(BlockDescBind);
+  DISABLE_COPY_AND_ASSIGN(BlockDesc);
 };
 }  // namespace framework
 }  // namespace paddle

paddle/framework/data_layout.h

+/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
+
+#pragma once
+#include "paddle/platform/enforce.h"
+
+#include <iostream>
+#include "paddle/platform/enforce.h"
+
+namespace paddle {
+namespace framework {
+
+enum class DataLayout {
+  kNHWC = 0,
+  kNCHW = 1,
+  kAnyLayout = 2,
+};
+
+inline DataLayout StringToDataLayout(const std::string& str) {
+  if (str == "NHWC" || str == "nhwc") {
+    return DataLayout::kNHWC;
+  } else if (str == "NCHW" || str == "nchw") {
+    return DataLayout::kNCHW;
+  } else {
+    PADDLE_THROW("Unknown storage order string: %s", str);
+  }
+}
+
+inline std::string DataLayoutToString(const DataLayout& data_layout) {
+  switch (data_layout) {
+    case DataLayout::kNHWC:
+      return "NHWC";
+    case DataLayout::kNCHW:
+      return "NCHW";
+    case DataLayout::kAnyLayout:
+      return "ANY_LAYOUT";
+    default:
+      PADDLE_THROW("unknown DataLayou %d", data_layout);
+  }
+}
+
+inline std::ostream& operator<<(std::ostream& out, DataLayout l) {
+  out << DataLayoutToString(l);
+  return out;
+}
+
+}  // namespace framework
+}  // namespace paddle

paddle/framework/data_transform.cc

+/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
+
+#include "paddle/framework/data_transform.h"
+#include "paddle/framework/lod_tensor.h"
+
+namespace paddle {
+namespace framework {
+
+DataTransformFnMap& DataTransformFnMap::Instance() {
+  static DataTransformFnMap data_transform_map;
+  return data_transform_map;
+}
+
+}  // namespace framework
+}  // namespace paddle

paddle/framework/data_transform.h

+/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
+
+#pragma once
+
+#include <functional>
+#include <utility>
+#include <vector>
+
+#include "paddle/framework/op_kernel_type.h"
+#include "paddle/framework/tensor.h"
+#include "paddle/framework/variable.h"
+#include "paddle/platform/device_context.h"
+#include "paddle/platform/macros.h"
+
+namespace paddle {
+namespace framework {
+
+using DataTransformFn =
+    std::function<void(const std::vector<platform::DeviceContext*> ctx,
+                       const Variable& in, Variable* out)>;
+using KernelTypePair = std::pair<OpKernelType, OpKernelType>;
+
+struct KernelTypePairHash {
+  static void HashCombine(const OpKernelType& t, std::size_t* seed) {
+    OpKernelType::Hash kernel_type_hasher;
+    (*seed) ^= kernel_type_hasher(t) + 0x9e3779b9 + (*seed << 6) + (*seed >> 2);
+  }
+
+  size_t operator()(const KernelTypePair& kernel_pair) const {
+    std::size_t seed = 0;
+    HashCombine(kernel_pair.first, &seed);
+    HashCombine(kernel_pair.second, &seed);
+    return seed;
+  }
+};
+
+using DataTransformMap =
+    std::unordered_map<KernelTypePair, DataTransformFn, KernelTypePairHash>;
+
+class DataTransformFnMap {
+ public:
+  static DataTransformFnMap& Instance();
+
+  bool Has(const KernelTypePair& key_pair) const {
+    return map_.find(key_pair) != map_.end();
+  }
+
+  void Insert(const OpKernelType& left, const OpKernelType& right,
+              const DataTransformFn& data_tranform_fn) {
+    Insert(std::make_pair(left, right), data_tranform_fn);
+  }
+
+  void Insert(const KernelTypePair& kernel_type_pair,
+              const DataTransformFn& data_tranform_fn) {
+    PADDLE_ENFORCE(!Has(kernel_type_pair),
+                   "KernelTypePair %s has been registered", "");
+    map_.insert({kernel_type_pair, data_tranform_fn});
+  }
+
+  const DataTransformFn& Get(const KernelTypePair& key_pair) const {
+    auto data_transformer = GetNullable(key_pair);
+    PADDLE_ENFORCE_NOT_NULL(data_transformer,
+                            "DataTransformFn should not be NULL");
+    return *data_transformer;
+  }
+
+  const DataTransformFn* GetNullable(const KernelTypePair& key_pair) const {
+    auto it = map_.find(key_pair);
+    if (it == map_.end()) {
+      return nullptr;
+    } else {
+      return &(it->second);
+    }
+  }
+
+  const DataTransformMap& Map() const { return map_; }
+
+ private:
+  DataTransformFnMap() = default;
+  DataTransformMap map_;
+  DISABLE_COPY_AND_ASSIGN(DataTransformFnMap);
+};
+
+// generate unique name with __LINE__
+// refs https://stackoverflow.com/questions/1597007
+#define TOKENPASTE(x, y) x##y
+#define TOKENPASTE2(x, y) TOKENPASTE(x, y)
+#define REGISTER_DATA_TRANSFORM_FN(from, to, fn)                              \
+  static int TOKENPASTE2(fn_, __LINE__)() {                                   \
+    ::paddle::framework::DataTransformFnMap::Instance().Insert(from, to, fn); \
+    return 0;                                                                 \
+  }                                                                           \
+  static int TOKENPASTE2(var_, __LINE__) __attribute__((unused)) =            \
+      TOKENPASTE2(fn_, __LINE__)()
+
+}  // namespace framework
+}  // namespace paddle

paddle/framework/data_transform_test.cc

+/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
+#include <array>
+#include <vector>
+
+#include <gtest/gtest.h>
+
+#include "paddle/framework/data_transform.h"
+
+namespace paddle {
+namespace framework {
+using namespace platform;
+
+/**
+ * @brief cross validation of different kernel type transform
+ *  We use four bit map represent different combination.
+ *  If the field has multiple possible value, only choose two of them.
+ *  For DataType, only test the FP32(float), FP64(double).
+ *  e.g. 0000 -> FP32, CPUPlace, kNHWC, kPlain
+ *       1111 -> FP64, GPUPlace, kNCHW, kMKLDNN
+ */
+
+std::array<proto::DataType, 2> kDataType = {
+    {proto::DataType::FP32, proto::DataType::FP64}};
+
+std::array<Place, 2> kPlace = {{CPUPlace(), CUDAPlace(0)}};
+
+std::array<DataLayout, 2> kDataLayout = {
+    {DataLayout::kNHWC, DataLayout::kNCHW}};
+
+std::array<LibraryType, 2> kLibraryType = {
+    {LibraryType::kPlain, LibraryType::kMKLDNN}};
+
+OpKernelType GenFromBit(const std::vector<bool> bits) {
+  return OpKernelType(kDataType[bits[0]], kPlace[bits[1]], kDataLayout[bits[2]],
+                      kLibraryType[bits[3]]);
+}
+
+int test_value = 0;
+
+auto kernel0 = GenFromBit({0, 0, 0, 0});
+auto kernel1 = GenFromBit({0, 0, 0, 1});
+auto kernel2 = GenFromBit({0, 0, 1, 0});
+auto kernel3 = GenFromBit({0, 0, 1, 1});
+
+void TransDataType_t(std::vector<platform::DeviceContext*> ctx,
+                     const Variable& in, Variable* out) {
+  test_value++;
+}
+
+void TransDataLayout_t(std::vector<platform::DeviceContext*> ctx,
+                       const Variable& in, Variable* out) {
+  test_value--;
+}
+
+void TransLibraryType_t(std::vector<platform::DeviceContext*> ctx,
+                        const Variable& in, Variable* out) {
+  test_value += 2;
+}
+
+}  // namespace framework
+}  // namespace paddle
+
+namespace frw = paddle::framework;
+
+REGISTER_DATA_TRANSFORM_FN(frw::kernel0, frw::kernel1, frw::TransDataType_t);
+REGISTER_DATA_TRANSFORM_FN(frw::kernel1, frw::kernel2, frw::TransDataLayout_t);
+REGISTER_DATA_TRANSFORM_FN(frw::kernel0, frw::kernel2, frw::TransLibraryType_t);
+
+TEST(DataTransform, Register) {
+  using namespace paddle::framework;
+  using namespace paddle::platform;
+
+  auto& instance = DataTransformFnMap::Instance();
+  std::vector<DeviceContext*> ctx;
+  paddle::framework::Variable in;
+  paddle::framework::Variable out;
+
+  instance.Get(std::make_pair(frw::kernel0, frw::kernel1))(ctx, in, &out);
+  ASSERT_EQ(test_value, 1);
+
+  instance.Get(std::make_pair(frw::kernel1, frw::kernel2))(ctx, in, &out);
+  ASSERT_EQ(test_value, 0);
+
+  instance.Get(std::make_pair(frw::kernel0, frw::kernel2))(ctx, in, &out);
+  ASSERT_EQ(test_value, 2);
+}

paddle/framework/data_type.h

 /* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
 
-   Licensed under the Apache License, Version 2.0 (the "License");
-   you may not use this file except in compliance with the License.
-   You may obtain a copy of the License at
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
 
-   http://www.apache.org/licenses/LICENSE-2.0
+    http://www.apache.org/licenses/LICENSE-2.0
 
-   Unless required by applicable law or agreed to in writing, software
-   distributed under the License is distributed on an "AS IS" BASIS,
-   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-   See the License for the specific language governing permissions and
-   limitations under the License. */
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
 
 #pragma once
 #include <typeindex>
@@ -20,7 +20,8 @@
 namespace paddle {
 namespace framework {
 
-inline DataType ToDataType(std::type_index type) {
+inline proto::DataType ToDataType(std::type_index type) {
+  using namespace paddle::framework::proto;
   if (typeid(float).hash_code() == type.hash_code()) {
     return DataType::FP32;
   } else if (typeid(double).hash_code() == type.hash_code()) {
@@ -36,7 +37,8 @@ inline DataType ToDataType(std::type_index type) {
   }
 }
 
-inline std::type_index ToTypeIndex(DataType type) {
+inline std::type_index ToTypeIndex(proto::DataType type) {
+  using namespace paddle::framework::proto;
   switch (type) {
     case DataType::FP32:
       return typeid(float);
@@ -54,7 +56,8 @@ inline std::type_index ToTypeIndex(DataType type) {
 }
 
 template <typename Visitor>
-inline void VisitDataType(DataType type, Visitor visitor) {
+inline void VisitDataType(proto::DataType type, Visitor visitor) {
+  using namespace paddle::framework::proto;
   switch (type) {
     case DataType::FP32:
       visitor.template operator()<float>();