diff --git a/README.md b/README.md index ceeb6d9e5193763293d3fce76e464340fbce533f..577528e7aaf45ce002467590ec66b19afb145920 100644 --- a/README.md +++ b/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! diff --git a/doc/api/v2/fluid/layers.rst b/doc/api/v2/fluid/layers.rst index d2bb7a8187ba7f1ae78a5337794a8882c96d2346..92ca1cf0f836a376387f3e6f2b5a24c78109323d 100644 --- a/doc/api/v2/fluid/layers.rst +++ b/doc/api/v2/fluid/layers.rst @@ -305,3 +305,10 @@ lstm_unit --------- .. autofunction:: paddle.v2.fluid.layers.lstm_unit :noindex: + + +sequence_softmax +--------- +.. autofunction:: paddle.v2.fluid.layers.sequence_softmax + :noindex: + diff --git a/doc/howto/usage/cluster/cluster_train_cn.md b/doc/howto/usage/cluster/cluster_train_cn.md index c9f90538a669d4705d18c3cd9b6dbf4a535c35b8..659bae9c0ceaf2fb2df8446b9d406a822a9df0ea 100644 --- a/doc/howto/usage/cluster/cluster_train_cn.md +++ b/doc/howto/usage/cluster/cluster_train_cn.md @@ -1,4 +1,4 @@ -# PaddlePaddle分布式训练 +# 分布式训练 ## 概述 @@ -181,8 +181,8 @@ PaddlePaddle可以使用多种分布式计算平台构建分布式计算任务 ## 在不同集群中运行 - - [fabric](fabric_cn.md) - - [openmpi](openmpi_cn.md) - - [kubernetes](k8s_cn.md) - - [kubernetes distributed](k8s_distributed_cn.md) - - [kubernetes on AWS](k8s_aws_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) diff --git a/doc/howto/usage/cluster/cluster_train_en.md b/doc/howto/usage/cluster/cluster_train_en.md index f9819470c0c622b4bc0ea064303d742385603230..915405ca5b446981515e301ca4b7ee065a82a9ff 100644 --- a/doc/howto/usage/cluster/cluster_train_en.md +++ b/doc/howto/usage/cluster/cluster_train_en.md @@ -1,4 +1,4 @@ -# PaddlePaddle Distributed Training +# Distributed Training ## Introduction @@ -188,5 +188,4 @@ These cluster platforms provide API or environment variables for training proces - [fabric](fabric_en.md) - [openmpi](openmpi_en.md) - [kubernetes](k8s_en.md) - - kubernetes distributed - [kubernetes on AWS](k8s_aws_en.md) diff --git a/doc/howto/usage/cluster/k8s_aws_en.md b/doc/howto/usage/cluster/k8s_aws_en.md index ce72b0803818d5bf0c18753c421848cf2fc1b668..0dfa8237a3fa2c9c3ee11e873c9fbbed3cd6018f 100644 --- a/doc/howto/usage/cluster/k8s_aws_en.md +++ b/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 diff --git a/doc/howto/usage/cluster/k8s_cn.md b/doc/howto/usage/cluster/k8s_cn.md index ab07cb9cd5b135ddea82b3360720537f1dc5a801..c1a11f7165a2f9da9dd044641274447e7943a597 100644 --- a/doc/howto/usage/cluster/k8s_cn.md +++ b/doc/howto/usage/cluster/k8s_cn.md @@ -1,21 +1,22 @@ # 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： diff --git a/doc/howto/usage/cluster/k8s_distributed_cn.md b/doc/howto/usage/cluster/k8s_distributed_cn.md index 0fc9e37a990104e942636fc807f67a99f0df9da8..701a9a75d78b53d7dab94529dbd1be382ff0d04e 100644 --- a/doc/howto/usage/cluster/k8s_distributed_cn.md +++ b/doc/howto/usage/cluster/k8s_distributed_cn.md @@ -28,7 +28,7 @@ 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/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 +62,7 @@ spec: hostNetwork: true containers: - name: paddle-data - image: paddledev/paddle-tutorial:k8s_data + image: paddlepaddle/paddle-tutorial:k8s_data imagePullPolicy: Always volumeMounts: - mountPath: "/mnt" diff --git a/doc/howto/usage/cluster/k8s_en.md b/doc/howto/usage/cluster/k8s_en.md index 0c3ab05b708e7a924577c26496b8c55126e76c62..c374f00a495d705ceddf8d3d930768ceeb93282b 100644 --- a/doc/howto/usage/cluster/k8s_en.md +++ b/doc/howto/usage/cluster/k8s_en.md @@ -1,18 +1,27 @@ -# 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. diff --git a/doc/howto/usage/cluster/src/Dockerfile b/doc/howto/usage/cluster/src/Dockerfile index 3a73606c61432329b4cc2d2f8daadc5af8735c96..e178bf4da0f32fca9586b5b69a2c7419de5d9cb1 100644 --- a/doc/howto/usage/cluster/src/Dockerfile +++ b/doc/howto/usage/cluster/src/Dockerfile @@ -1,4 +1,4 @@ -FROM paddledev/paddle:cpu-latest +FROM paddlepaddle/paddle:latest MAINTAINER zjsxzong89@gmail.com diff --git a/doc/howto/usage/cluster/src/k8s_train/Dockerfile b/doc/howto/usage/cluster/src/k8s_train/Dockerfile index c0fca1f9a945921e6e8899fee2db8845e66136a1..77f021a89a70d934bf70424eaa3c6dc3f7c93a28 100644 --- a/doc/howto/usage/cluster/src/k8s_train/Dockerfile +++ b/doc/howto/usage/cluster/src/k8s_train/Dockerfile @@ -1,4 +1,4 @@ -FROM paddledev/paddle:cpu-latest +FROM paddlepaddle/paddle:latest COPY start.sh /root/ COPY start_paddle.py /root/ diff --git a/paddle/pybind/CMakeLists.txt b/paddle/pybind/CMakeLists.txt index 1fb69de90d2fb5386dffdd95825c496a8fa559d3..6afed7eec7001b646d55cef0bc3f59782b80b15f 100644 --- a/paddle/pybind/CMakeLists.txt +++ b/paddle/pybind/CMakeLists.txt @@ -1,6 +1,6 @@ if(WITH_PYTHON) cc_library(paddle_pybind SHARED - SRCS pybind.cc exception.cc protobuf.cc + SRCS pybind.cc exception.cc protobuf.cc const_value.cc DEPS pybind python backward proto_desc paddle_memory executor prune init ${GLOB_OP_LIB}) endif(WITH_PYTHON) diff --git a/paddle/pybind/const_value.cc b/paddle/pybind/const_value.cc new file mode 100644 index 0000000000000000000000000000000000000000..b13ad42ea29453354798d88bff8ef47339d1a614 --- /dev/null +++ b/paddle/pybind/const_value.cc @@ -0,0 +1,29 @@ +/* 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 "const_value.h" +#include "paddle/framework/operator.h" + +namespace paddle { +namespace pybind { + +void BindConstValue(pybind11::module& m) { + m.def("kEmptyVarName", [] { return framework::kEmptyVarName; }); + m.def("kTempVarName", [] { return framework::kTempVarName; }); + m.def("kGradVarSuffix", [] { return framework::kGradVarSuffix; }); + m.def("kZeroVarSuffix", [] { return framework::kZeroVarSuffix; }); +} + +} // namespace pybind +} // namespace paddle diff --git a/paddle/pybind/const_value.h b/paddle/pybind/const_value.h new file mode 100644 index 0000000000000000000000000000000000000000..3d57c972a9d5339c0e155fa1d6395af9face8744 --- /dev/null +++ b/paddle/pybind/const_value.h @@ -0,0 +1,26 @@ +/* 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 +#include "paddle/platform/enforce.h" +#include "pybind11/pybind11.h" + +namespace py = pybind11; + +namespace paddle { +namespace pybind { +extern void BindConstValue(pybind11::module& m); +} // namespace pybind +} // namespace paddle diff --git a/paddle/pybind/pybind.cc b/paddle/pybind/pybind.cc index 4248db34c6345bd62e63628c7794b40d8a1adab6..4a82f1596eb0b7b3cfe9b9bbce32549f58efdbc9 100644 --- a/paddle/pybind/pybind.cc +++ b/paddle/pybind/pybind.cc @@ -30,6 +30,7 @@ limitations under the License. */ #include "paddle/operators/net_op.h" #include "paddle/platform/enforce.h" #include "paddle/platform/place.h" +#include "paddle/pybind/const_value.h" #include "paddle/pybind/exception.h" #include "paddle/pybind/pybind.h" #include "paddle/pybind/tensor_py.h" @@ -431,6 +432,7 @@ All parameter, weight, gradient are variables in Paddle. BindBlockDesc(m); BindVarDsec(m); BindOpDesc(m); + BindConstValue(m); py::class_(m, "LodRankTable") .def("items", [](framework::LoDRankTable &table) { diff --git a/python/paddle/v2/fluid/__init__.py b/python/paddle/v2/fluid/__init__.py index 59986c9f0ca8e4b793463db0e8c5da0489654ee9..9b3792ee9e3e4c6f319b3e2b13c4aa3a05cce8be 100644 --- a/python/paddle/v2/fluid/__init__.py +++ b/python/paddle/v2/fluid/__init__.py @@ -16,12 +16,13 @@ import regularizer from param_attr import ParamAttr from data_feeder import DataFeeder from core import LoDTensor, CPUPlace, GPUPlace +import clip Tensor = LoDTensor __all__ = framework.__all__ + executor.__all__ + [ 'io', 'initializer', 'layers', 'nets', 'optimizer', 'backward', 'regularizer', 'LoDTensor', 'CPUPlace', 'GPUPlace', 'Tensor', 'ParamAttr' - 'DataFeeder' + 'DataFeeder', 'clip' ] diff --git a/python/paddle/v2/fluid/clip.py b/python/paddle/v2/fluid/clip.py new file mode 100644 index 0000000000000000000000000000000000000000..d7ec2fbe13fe6d9158345099b8668afc5c7d4571 --- /dev/null +++ b/python/paddle/v2/fluid/clip.py @@ -0,0 +1,61 @@ +import functools +import layers + +__all__ = ['GradientClipByValue', 'append_gradient_clip_ops'] + + +class BaseGradientClipAttr(object): + def process_context(self, context, p_g): + raise NotImplementedError() + + def create_operators(self, param, grad): + raise NotImplementedError() + + +class NullGradientClipAttr(BaseGradientClipAttr): + def process_context(self, context, p_g): + pass + + def create_operators(self, param, grad): + return param, grad + + +class GradientClipByValue(BaseGradientClipAttr): + def __init__(self, max, min=None): + max = float(max) + if min is None: + min = -max + else: + min = float(min) + self.max = max + self.min = min + + def process_context(self, context, p_g): + pass + + def create_operators(self, param, grad): + new_grad = layers.clip(x=grad, min=self.min, max=self.max) + return param, new_grad + + +def append_gradient_clip_ops(param_grad): + context = dict() + create_op_callbacks = [] + for p, g in param_grad: + clip_attr = getattr(p, 'clip_attr', NullGradientClipAttr()) + if clip_attr is None: + clip_attr = NullGradientClipAttr() + if not isinstance(clip_attr, BaseGradientClipAttr): + raise TypeError( + "clip attribute should be an instance of BaseGradientClippingAttr" + ) + + clip_attr.process_context(context=context, p_g=param_grad) + create_op_callbacks.append( + functools.partial( + clip_attr.create_operators, param=p, grad=g)) + + return [each_callback() for each_callback in create_op_callbacks] + + +ClipByValue = GradientClipByValue diff --git a/python/paddle/v2/fluid/framework.py b/python/paddle/v2/fluid/framework.py index e9319cbe2a4f390f591de4d3b7426453a9ed5d1b..713d8dd1655aabeeab0807cce4257e5d56196aa5 100644 --- a/python/paddle/v2/fluid/framework.py +++ b/python/paddle/v2/fluid/framework.py @@ -1,10 +1,10 @@ import collections +import contextlib import numpy as np -from . import core + import proto.framework_pb2 as framework_pb2 -import google.protobuf.message -import contextlib +from . import core __all__ = [ 'Block', 'Variable', 'Program', 'Operator', 'default_startup_program', @@ -12,6 +12,18 @@ __all__ = [ 'switch_main_program' ] +EMPTY_VAR_NAME = core.kEmptyVarName() +TEMP_VAR_NAME = core.kTempVarName() +GRAD_VAR_SUFFIX = core.kGradVarSuffix() +ZERO_VAR_SUFFIX = core.kZeroVarSuffix() + + +def grad_var_name(var_name): + """ + return gradient name for a certain var name + """ + return var_name + GRAD_VAR_SUFFIX + def unique_name(prefix): """ @@ -704,6 +716,7 @@ class Block(object): trainable=p.trainable, optimize_attr=p.optimize_attr, regularizer=p.regularizer, + clip_attr=p.clip_attr, name=v.name) self.vars[new_p.name] = new_p @@ -866,6 +879,8 @@ class Parameter(Variable): self.regularizer = kwargs.get('regularizer', None) + self.clip_attr = kwargs.get('clip_attr', None) + # program is a global instance. _main_program_ = Program() diff --git a/python/paddle/v2/fluid/layers/ops.py b/python/paddle/v2/fluid/layers/ops.py index fa312ace60390e5fdd9637dccc71ccf8b247ca47..d2ff6841a317aaf6903edadc9213f69ef6c41216 100644 --- a/python/paddle/v2/fluid/layers/ops.py +++ b/python/paddle/v2/fluid/layers/ops.py @@ -2,7 +2,7 @@ from ..registry import register_layer __all__ = [ 'mean', 'mul', 'dropout', 'reshape', 'sigmoid', 'scale', 'transpose', 'sigmoid_cross_entropy_with_logits', 'elementwise_add', 'elementwise_div', - 'elementwise_sub', 'elementwise_mul', 'clip', 'abs' + 'elementwise_sub', 'elementwise_mul', 'clip', 'abs', 'sequence_softmax' ] for _OP in set(__all__): diff --git a/python/paddle/v2/fluid/optimizer.py b/python/paddle/v2/fluid/optimizer.py index 9f03eeea83e6d212da5fbe3d090d82028fa378ac..84fcbcdc2f2868a84bad5e145a934e33485b1fef 100644 --- a/python/paddle/v2/fluid/optimizer.py +++ b/python/paddle/v2/fluid/optimizer.py @@ -6,6 +6,7 @@ from framework import unique_name, program_guard from initializer import Constant from layer_helper import LayerHelper from regularizer import append_regularization_ops +from clip import append_gradient_clip_ops __all__ = ['SGD', 'Momentum', 'Adagrad', 'Adam', 'Adamax', 'DecayedAdagrad'] @@ -197,9 +198,13 @@ class Optimizer(object): `create_optimization_pass()` into one. """ params_grads = append_backward_ops(loss, parameter_list, no_grad_set) + + params_grads = append_gradient_clip_ops(params_grads) + # Add regularization if any params_grads = append_regularization_ops(params_grads, self.regularization) + optimize_ops = self.create_optimization_pass(params_grads, loss, startup_program) return optimize_ops diff --git a/python/paddle/v2/fluid/param_attr.py b/python/paddle/v2/fluid/param_attr.py index 7952a5ea51c00f72664443fb26faa455e89da7be..f6f320c788e7e08d44df8aff5ad3792b237e103a 100644 --- a/python/paddle/v2/fluid/param_attr.py +++ b/python/paddle/v2/fluid/param_attr.py @@ -1,6 +1,8 @@ from initializer import Initializer, Xavier, Constant from regularizer import WeightDecayRegularizer +__all__ = ['ParamAttr'] + class ParamAttr(object): def __init__(self, @@ -8,12 +10,14 @@ class ParamAttr(object): initializer=None, learning_rate=1.0, regularizer=None, - trainable=True): + trainable=True, + clip=None): self.name = name self.initializer = initializer self.learning_rate = learning_rate self.regularizer = regularizer self.trainable = trainable + self.clip = clip def set_default_initializer(self, initializer): if initializer is None: @@ -56,7 +60,8 @@ class ParamAttr(object): 'name': self.name, 'learning_rate': self.learning_rate, 'regularizer': self.regularizer, - 'trainable': self.trainable + 'trainable': self.trainable, + 'clip_attr': self.clip } if with_initializer: kwargs['initializer'] = self.initializer diff --git a/python/paddle/v2/fluid/tests/book/test_recognize_digits_mlp.py b/python/paddle/v2/fluid/tests/book/test_recognize_digits_mlp.py index d77f19660ebcd470837e8b4e63509683de4a7a82..fc073f6be8563a363c0f98b9235ae267fa68562d 100644 --- a/python/paddle/v2/fluid/tests/book/test_recognize_digits_mlp.py +++ b/python/paddle/v2/fluid/tests/book/test_recognize_digits_mlp.py @@ -11,7 +11,9 @@ regularizer = fluid.regularizer.L2Decay(0.0005 * BATCH_SIZE) hidden1 = fluid.layers.fc(input=image, size=128, act='relu', - param_attr=regularizer) + param_attr=fluid.ParamAttr( + regularizer=regularizer, + clip=fluid.clip.ClipByValue(10))) hidden2 = fluid.layers.fc(input=hidden1, size=64, act='relu', diff --git a/python/paddle/v2/fluid/tests/test_batch_norm_op.py b/python/paddle/v2/fluid/tests/test_batch_norm_op.py index e766a68c0e338b07e47260e40edc544c98555382..dee2febb83d171ed4a13921e3b7d37322ead2786 100644 --- a/python/paddle/v2/fluid/tests/test_batch_norm_op.py +++ b/python/paddle/v2/fluid/tests/test_batch_norm_op.py @@ -3,10 +3,7 @@ import numpy as np from op_test import OpTest import paddle.v2.fluid.core as core from paddle.v2.fluid.op import Operator - - -def grad_var_name(var_name): - return var_name + "@GRAD" +from paddle.v2.fluid.framework import grad_var_name def get_backward_op(scope, op, no_grad_set): diff --git a/python/paddle/v2/fluid/tests/test_const_value.py b/python/paddle/v2/fluid/tests/test_const_value.py new file mode 100644 index 0000000000000000000000000000000000000000..f8c17c2c98674fa67458efa090e166e37f5a6a8a --- /dev/null +++ b/python/paddle/v2/fluid/tests/test_const_value.py @@ -0,0 +1,14 @@ +import unittest +import paddle.v2.fluid.framework as framework + + +class ConditionalBlock(unittest.TestCase): + def test_const_value(self): + self.assertEqual(framework.GRAD_VAR_SUFFIX, "@GRAD") + self.assertEqual(framework.TEMP_VAR_NAME, "@TEMP@") + self.assertEqual(framework.GRAD_VAR_SUFFIX, "@GRAD") + self.assertEqual(framework.ZERO_VAR_SUFFIX, "@ZERO") + + +if __name__ == '__main__': + unittest.main() diff --git a/python/paddle/v2/fluid/tests/test_layers.py b/python/paddle/v2/fluid/tests/test_layers.py index 4fceb0eca5bb16cb39f9703b3a992d1936b1b2dc..c6f109bc959c6dee364dd06ae8bcd7f2e4fb6a8f 100644 --- a/python/paddle/v2/fluid/tests/test_layers.py +++ b/python/paddle/v2/fluid/tests/test_layers.py @@ -187,6 +187,15 @@ class TestBook(unittest.TestCase): x_t=x_t, hidden_t_prev=prev_hidden, cell_t_prev=prev_cell)) print(str(program)) + def test_sequence_softmax(self): + program = Program() + with program_guard(program): + seq_data = layers.data( + name='seq_data', shape=[10, 10], dtype='float32', lod_level=1) + seq = layers.fc(input=seq_data, size=20) + self.assertIsNotNone(layers.sequence_softmax(x=seq)) + print(str(program)) + def test_get_places(self): program = Program() with program_guard(program): diff --git a/python/paddle/v2/fluid/tests/test_operator.py b/python/paddle/v2/fluid/tests/test_operator.py index 4aa022ef90159cd96eed4e4dbe30cf5d1e8a41a7..c059a2b88b1324935f871b6e9c11efd5652ddd65 100644 --- a/python/paddle/v2/fluid/tests/test_operator.py +++ b/python/paddle/v2/fluid/tests/test_operator.py @@ -1,6 +1,6 @@ import unittest + import paddle.v2.fluid.op as op -import paddle.v2.fluid.core as core import paddle.v2.fluid.proto.framework_pb2 as framework_pb2 diff --git a/python/paddle/v2/fluid/tests/test_program.py b/python/paddle/v2/fluid/tests/test_program.py index e6da0b2be77533811c98751de1067dbbeac11309..447c746aacc1c9455d7a023bca625d548ab2638b 100644 --- a/python/paddle/v2/fluid/tests/test_program.py +++ b/python/paddle/v2/fluid/tests/test_program.py @@ -1,7 +1,7 @@ from __future__ import print_function import unittest -from paddle.v2.fluid.framework import Program, default_main_program, program_guard +from paddle.v2.fluid.framework import Program, default_main_program, program_guard, grad_var_name import paddle.v2.fluid.layers as layers main_program = default_main_program() @@ -109,12 +109,10 @@ class TestProgram(unittest.TestCase): self.assertEqual(add_op.idx, 1) param_to_grad = prog.append_backward(mean_out, set()) - def grad_name(name): - return name + "@GRAD" - for var_name in ("mul.x", "mul.y", "mul.out", "add.y", "add.out", "mean.out"): - self.assertEqual(param_to_grad[var_name][0], grad_name(var_name)) + self.assertEqual(param_to_grad[var_name][0], + grad_var_name(var_name)) self.assertEqual(param_to_grad[var_name][1], 0) expect_ops = [ diff --git a/python/paddle/v2/fluid/tests/test_recurrent_op.py b/python/paddle/v2/fluid/tests/test_recurrent_op.py index 694ff0d8dd794111aff51bb7d503a56b87514342..e38c763ddbcc5c8410f41d062c05499333a3ee55 100644 --- a/python/paddle/v2/fluid/tests/test_recurrent_op.py +++ b/python/paddle/v2/fluid/tests/test_recurrent_op.py @@ -1,7 +1,7 @@ import unittest import paddle.v2.fluid.layers as layers -from paddle.v2.fluid.framework import Program +from paddle.v2.fluid.framework import Program, grad_var_name from paddle.v2.fluid.executor import Executor from paddle.v2.fluid.backward import append_backward_ops import numpy as np @@ -164,7 +164,7 @@ class RecurrentOpTest1(unittest.TestCase): for x in self.data_field } fetch_list = [ - self.main_program.global_block().var(x + "@GRAD") + self.main_program.global_block().var(grad_var_name(x)) for x in self.data_field ]