Fix conflicts with develop branch

.gitmodules

@@ -0,0 +1,3 @@
+[submodule "warp-ctc"]
+	path = warp-ctc
+	url = https://github.com/baidu-research/warp-ctc.git

.pre-commit-config.yaml

@@ -2,6 +2,7 @@
     sha: c25201a00e6b0514370501050cf2a8538ac12270
     hooks:
     -   id: remove-crlf
+        files: (?!.*warp-ctc)^.*$
 -   repo: https://github.com/reyoung/mirrors-yapf.git
     sha: v0.13.2
     hooks:
@@ -13,6 +14,7 @@
     -   id: check-merge-conflict
     -   id: check-symlinks
     -   id: detect-private-key
+        files: (?!.*warp-ctc)^.*$
     -   id: end-of-file-fixer
 -   repo: https://github.com/PaddlePaddle/clang-format-pre-commit-hook.git
     sha: 28c0ea8a67a3e2dbbf4822ef44e85b63a0080a29

.travis.yml

@@ -50,7 +50,7 @@ before_install:
     fi
   - if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then sudo paddle/scripts/travis/before_install.linux.sh; fi
   - if [[ "$TRAVIS_OS_NAME" == "osx" ]]; then paddle/scripts/travis/before_install.osx.sh; fi
-  - pip install wheel protobuf sphinx breathe recommonmark virtualenv numpy
+  - pip install wheel protobuf sphinx breathe recommonmark virtualenv numpy sphinx_rtd_theme
 script:
   - paddle/scripts/travis/main.sh
 notifications:

CMakeLists.txt

@@ -71,10 +71,10 @@ find_package(Git REQUIRED)
 include(version)
 add_definitions(-DPADDLE_VERSION=\"${PADDLE_VERSION}\")
 
-
 if(NOT WITH_GPU)
     add_definitions(-DPADDLE_ONLY_CPU)
     add_definitions(-DHPPL_STUB_FUNC)
+
     list(APPEND CMAKE_CXX_SOURCE_FILE_EXTENSIONS cu)
 else()
     if(${CUDA_VERSION_MAJOR} VERSION_LESS 7)
@@ -91,15 +91,15 @@ else()
         set(CUDA_NVCC_FLAGS ${CUDA_NVCC_FLAGS} "-Xcompiler ${SSE3_FLAG}")
     endif(WITH_AVX)
 
-    if(WITH_DSO)
-        add_definitions(-DPADDLE_USE_DSO)
-    endif(WITH_DSO)
-
     # Include cuda and cudnn
     include_directories(${CUDNN_INCLUDE_DIR})
     include_directories(${CUDA_TOOLKIT_INCLUDE})
 endif(NOT WITH_GPU)
 
+if(WITH_DSO)
+    add_definitions(-DPADDLE_USE_DSO)
+endif(WITH_DSO)
+
 if(WITH_DOUBLE)
     add_definitions(-DPADDLE_TYPE_DOUBLE)
     set(ACCURACY double)

cmake/util.cmake

@@ -148,6 +148,11 @@ function(link_paddle_exe TARGET_NAME)
             target_link_libraries(${TARGET_NAME} rt)
         endif()
     endif()
+
+    if(NOT WITH_DSO)
+        target_link_libraries(${TARGET_NAME}
+            ${WARPCTC_LIBRARY})
+    endif()
 endfunction()
 
 # link_paddle_test

demo/seqToseq/dataprovider.py

@@ -19,27 +19,43 @@ START = "<s>"
 END = "<e>"
 
 
-def hook(settings, src_dict, trg_dict, file_list, **kwargs):
+def hook(settings, src_dict_path, trg_dict_path, is_generating, file_list,
+         **kwargs):
     # job_mode = 1: training mode
     # job_mode = 0: generating mode
-    settings.job_mode = trg_dict is not None
-    settings.src_dict = src_dict
+    settings.job_mode = not is_generating
+
+    def fun(dict_path):
+        out_dict = dict()
+        with open(dict_path, "r") as fin:
+            out_dict = {
+                line.strip(): line_count
+                for line_count, line in enumerate(fin)
+            }
+        return out_dict
+
+    settings.src_dict = fun(src_dict_path)
+    settings.trg_dict = fun(trg_dict_path)
+
     settings.logger.info("src dict len : %d" % (len(settings.src_dict)))
-    settings.sample_count = 0
 
     if settings.job_mode:
-        settings.trg_dict = trg_dict
-        settings.slots = [
+        settings.slots = {
+            'source_language_word':
             integer_value_sequence(len(settings.src_dict)),
+            'target_language_word':
             integer_value_sequence(len(settings.trg_dict)),
+            'target_language_next_word':
             integer_value_sequence(len(settings.trg_dict))
-        ]
+        }
         settings.logger.info("trg dict len : %d" % (len(settings.trg_dict)))
     else:
-        settings.slots = [
+        settings.slots = {
+            'source_language_word':
             integer_value_sequence(len(settings.src_dict)),
+            'sent_id':
             integer_value_sequence(len(open(file_list[0], "r").readlines()))
-        ]
+        }
 
 
 def _get_ids(s, dictionary):
@@ -69,6 +85,10 @@ def process(settings, file_name):
                     continue
                 trg_ids_next = trg_ids + [settings.trg_dict[END]]
                 trg_ids = [settings.trg_dict[START]] + trg_ids
-                yield src_ids, trg_ids, trg_ids_next
+                yield {
+                    'source_language_word': src_ids,
+                    'target_language_word': trg_ids,
+                    'target_language_next_word': trg_ids_next
+                }
             else:
-                yield src_ids, [line_count]
+                yield {'source_language_word': src_ids, 'sent_id': [line_count]}

demo/seqToseq/seqToseq_net.py

@@ -37,17 +37,10 @@ def seq_to_seq_data(data_dir,
     """
     src_lang_dict = os.path.join(data_dir, 'src.dict')
     trg_lang_dict = os.path.join(data_dir, 'trg.dict')
-    src_dict = dict()
-    for line_count, line in enumerate(open(src_lang_dict, "r")):
-        src_dict[line.strip()] = line_count
-    trg_dict = dict()
-    for line_count, line in enumerate(open(trg_lang_dict, "r")):
-        trg_dict[line.strip()] = line_count
 
     if is_generating:
         train_list = None
         test_list = os.path.join(data_dir, gen_list)
-        trg_dict = None
     else:
         train_list = os.path.join(data_dir, train_list)
         test_list = os.path.join(data_dir, test_list)
@@ -57,8 +50,11 @@ def seq_to_seq_data(data_dir,
         test_list,
         module="dataprovider",
         obj="process",
-        args={"src_dict": src_dict,
-              "trg_dict": trg_dict})
+        args={
+            "src_dict_path": src_lang_dict,
+            "trg_dict_path": trg_lang_dict,
+            "is_generating": is_generating
+        })
 
     return {
         "src_dict_path": src_lang_dict,

doc/conf.py.in

@@ -23,7 +23,7 @@ AutoStructify = transform.AutoStructify
 # documentation root, use os.path.abspath to make it absolute, like shown here.
 sys.path.insert(0, '@PROJ_ROOT@/python')
 
-templates_path = ["@PROJ_ROOT@/doc/templates"]
+templates_path = ["@PROJ_ROOT@/doc_theme/templates"]
 
 # -- General configuration ------------------------------------------------
 
@@ -113,13 +113,12 @@ todo_include_todos = False
 
 # The theme to use for HTML and HTML Help pages.  See the documentation for
 # a list of builtin themes.
-#html_theme = 'sphinx_rtd_theme'
-html_theme = 'classic'
+html_theme = 'sphinx_rtd_theme'
 
 # Add any paths that contain custom static files (such as style sheets) here,
 # relative to this directory. They are copied after the builtin static files,
 # so a file named "default.css" will overwrite the builtin "default.css".
-html_static_path = ['_static']
+html_static_path = ['@PROJ_ROOT@/doc_theme/static']
 
 # Output file base name for HTML help builder.
 htmlhelp_basename = project + 'doc'

doc/getstarted/build_and_install/build_from_source.md

@@ -11,6 +11,7 @@ You can download PaddlePaddle from the [github source](https://github.com/Paddle
 ```bash
 git clone https://github.com/PaddlePaddle/Paddle paddle
 cd paddle
+git submodule update --init --recursive
 ```
 
 ## <span id="requirements">Requirements</span>

doc/getstarted/build_and_install/docker_install.rst

@@ -19,8 +19,8 @@ automatically runs the following commands:
 
 .. code-block:: base
 
-   docker build -t paddle:cpu-noavx -f paddle/scripts/docker/Dockerfile .
-   docker build -t paddle:gpu-noavx -f paddle/scripts/docker/Dockerfile.gpu .
+   docker build -t paddle:cpu -f paddle/scripts/docker/Dockerfile .
+   docker build -t paddle:gpu -f paddle/scripts/docker/Dockerfile.gpu .
 
 
 To run the CPU-only image as an interactive container:
@@ -79,5 +79,28 @@ source code:
    cd ~
    git clone github.com/PaddlePaddle/Paddle
    cd Paddle
+   git submodule update --init --recursive
    docker build --build-arg WITH_AVX=OFF -t paddle:cpu-noavx -f paddle/scripts/docker/Dockerfile .
    docker build --build-arg WITH_AVX=OFF -t paddle:gpu-noavx -f paddle/scripts/docker/Dockerfile.gpu .
+
+
+Documentation
+-------------
+
+Paddle Docker images include an HTML version of C++ source code
+generated using `woboq code browser
+<https://github.com/woboq/woboq_codebrowser>`_.  This makes it easy
+for users to browse and understand the C++ source code.
+
+As long as we give the Paddle Docker container a name, we can run an
+additional nginx Docker container to serve the volume from the Paddle
+container:
+
+.. code-block:: bash
+
+   docker run -d --name paddle-cpu-doc paddle:cpu
+   docker run -d --volumes-from paddle-cpu-doc -p 8088:80 nginx
+
+
+Then we can direct our Web browser to the HTML version of source code
+at http://localhost:8088/paddle/

doc/howto/cmd_parameter/arguments.md

@@ -143,7 +143,7 @@ It looks like there are a lot of arguments. However, most of them are for develo
 </tr>
 
 <tr>
-<td class="left" rowspan = "2">testing during training</td><td class="left">test_all_data_in_one_period</td>
+<td class="left" rowspan = "2">testing during training</td><td class="left">test_period</td>
 <td class="left">√</td><td class="left">√</td><td class="left"></td><td class="left"></td>
 </tr>
 

doc/howto/cmd_parameter/detail_introduction.md

@@ -31,7 +31,7 @@
   - type: string (default: null).
 
 * `--version`
-  - Whether to print version infomatrion.
+  - Whether to print version information.
   - type: bool (default: 0).
 
 * `--show_layer_stat`
@@ -110,8 +110,8 @@
   - type: int32 (default: -1).
 
 * `--test_period`
-  - Run testing every test_period train batches. If not set, run testing each pass.
-  - type: int32 (default: 1000).
+   - if equal 0, do test on all test data at the end of each pass. While if equal non-zero, do test on all test data every test_period batches.
+  - type: int32 (default: 0).
 
 * `--test_wait`
   - Whether to wait for parameter per pass if not exist. If set test_data_path in submitting environment of cluster, it will launch one process to perfom testing, so we need to set test_wait=1. Note that in the cluster submitting environment, this argument has been set True by default.
@@ -121,10 +121,6 @@
   - File that saves the model list when testing. It was set automatically when using cluster submitting environment after setting model_path.
   - type: string (default: "", null).
 
-* `--test_all_data_in_one_period`
-  - This argument is usually used in testing period during traning. If true, all data will be tested in one test period. Otherwise (batch_size * log_peroid) data will be tested.
-  - type: bool (default: 0).
-
 * `--predict_output_dir`
   - Directory that saves the layer output. It is configured in Outputs() in network config. Default, this argument is null, meaning save nothing. Specify this directory if you want to save feature map of some layers in testing mode. Note that, layer outputs are values after activation function.
   - type: string (default: "", null).

doc/howto/cmd_parameter/use_case.md

@@ -10,9 +10,8 @@ paddle train \
   --config=network_config \
   --save_dir=output \
   --trainer_count=COUNT \                #(default:1)
-  --test_period=M \                      #(default:1000）
-  --test_all_data_in_one_period=true \   #(default:false) 
-  --num_passes=N \                       #(defalut:100）
+  --test_period=M \                      #(default:0) 
+  --num_passes=N \                       #(defalut:100)
   --log_period=K \                       #(default:100)
   --dot_period=1000 \                    #(default:1)
   #[--show_parameter_stats_period=100] \ #(default:0)

doc/howto/contribute_to_paddle.md

@@ -36,8 +36,9 @@ If your repository doesn't contain **develop** branch, just create it by your ow
 git clone https://github.com/USERNAME/Paddle.git Paddle
 cd Paddle
 git checkout -b develop  # create develop branch.
-git remote add upstream https://github.com/baidu/Paddle.git  # add upstream to baidu/Paddle
+git remote add upstream https://github.com/PaddlePaddle/Paddle.git  # add upstream to baidu/Paddle
 git pull upstream develop  # update to upstream
+git submodule update --init --recursive
 ```
 
 Then you can start to develop by making a local developement branch
@@ -69,7 +70,7 @@ To do this, you'll need to add a remote at first:
 # see the current configured remote repository
 git remote -v
 # add upstream repository
-git remote add upstream https://github.com/baidu/Paddle.git
+git remote add upstream https://github.com/PaddlePaddle/Paddle.git
 # verify the new upstream
 git remote -v
 ```

doc_cn/conf.py.in

@@ -22,7 +22,7 @@ AutoStructify = transform.AutoStructify
 # add these directories to sys.path here. If the directory is relative to the
 # documentation root, use os.path.abspath to make it absolute, like shown here.
 sys.path.insert(0, '@PROJ_ROOT@/python')
-templates_path = ["@PROJ_ROOT@/doc/templates"]
+templates_path = ["@PROJ_ROOT@/doc_theme/templates"]
 
 # -- General configuration ------------------------------------------------
 
@@ -112,12 +112,12 @@ todo_include_todos = False
 
 # The theme to use for HTML and HTML Help pages.  See the documentation for
 # a list of builtin themes.
-#html_theme = 'sphinx_rtd_theme'  # sphinx_rtd_theme will cause table bad style
-html_theme = 'classic'
+html_theme = 'sphinx_rtd_theme'
+
 # Add any paths that contain custom static files (such as style sheets) here,
 # relative to this directory. They are copied after the builtin static files,
 # so a file named "default.css" will overwrite the builtin "default.css".
-html_static_path = ['_static']
+html_static_path = ['@PROJ_ROOT@/doc_theme/static']
 
 # Output file base name for HTML help builder.
 htmlhelp_basename = project + 'doc'

doc_cn/faq/index.rst

@@ -214,3 +214,41 @@ PaddlePaddle的参数使用名字 :code:`name` 作为参数的ID，相同名字
         cmake .. -DPYTHON_EXECUTABLE=<exc_path> -DPYTHON_LIBRARY=<lib_path>  -DPYTHON_INCLUDE_DIR=<inc_path>
 
 用户需要指定本机上Python的路径：``<exc_path>``, ``<lib_path>``, ``<inc_path>``
+
+10. A protocol message was rejected because it was too big
+----------------------------------------------------------
+
+如果在训练NLP相关模型时，出现以下错误：
+
+..  code-block:: bash
+
+    [libprotobuf ERROR google/protobuf/io/coded_stream.cc:171] A protocol message was rejected because it was too big (more than 67108864 bytes).  To increase the limit (or to disable these warnings), see CodedInputStream::SetTotalBytesLimit() in google/protobuf/io/coded_stream.h.
+    F1205 14:59:50.295174 14703 TrainerConfigHelper.cpp:59] Check failed: m->conf.ParseFromString(configProtoStr) 
+
+可能的原因是：传给dataprovider的某一个args过大，一般是由于直接传递大字典导致的。错误的define_py_data_sources2类似：
+
+..  code-block:: python
+
+     src_dict = dict()
+     for line_count, line in enumerate(open(src_dict_path, "r")):
+        src_dict[line.strip()] = line_count
+
+     define_py_data_sources2(
+        train_list,
+        test_list,
+        module="dataprovider",
+        obj="process",
+        args={"src_dict": src_dict})
+
+解决方案是：将字典的地址作为args传给dataprovider，然后在dataprovider里面根据该地址加载字典。即define_py_data_sources2应改为：
+
+..  code-block:: python
+
+     define_py_data_sources2(
+        train_list,
+        test_list,
+        module="dataprovider",
+        obj="process",
+        args={"src_dict_path": src_dict_path})
+
+完整源码可参考 `seqToseq <https://github.com/PaddlePaddle/Paddle/tree/develop/demo/seqToseq>`_ 示例。

doc_cn/ui/data_provider/dataprovider.rst

@@ -0,0 +1,13 @@
+DataProvider的介绍
+==================
+
+DataProvider是PaddlePaddle负责提供数据的模块。其作用是将数据传入内存或显存，让神经网络可以进行训练或预测。用户可以通过简单使用Python接口 `PyDataProvider2 <pydataprovider2.html>`_ ，来自定义传数据的过程。如果有更复杂的使用，或者需要更高的效率，用户也可以在C++端自定义一个 ``DataProvider`` 。
+
+PaddlePaddle需要用户在网络配置（trainer_config.py）中定义使用哪种DataProvider，并且在DataProvider中实现如何访问训练文件列表（train.list）或测试文件列表（test.list）。
+
+- train.list和test.list存放在本地（推荐直接存放到训练目录，以相对路径引用)。一般情况下，两者均为纯文本文件，其中每一行对应一个数据文件地址：
+  
+  - 如果数据文件存于本地磁盘，这个地址则为它的绝对路径或相对路径(相对于PaddlePaddle程序运行时的路径)。
+  - 地址也可以为hdfs文件路径，或者数据库连接路径等。
+  - 由于这个地址会被DataProvider使用，因此，如何解析该地址也是用户自定义DataProvider时需要考虑的地方。
+- 如果没有设置test.list，或设置为None，那么在训练过程中不会执行测试操作；否则，会根据命令行参数指定的测试方式，在训练过程中进行测试，从而防止过拟合。

doc_cn/ui/data_provider/index.rst

@@ -1,17 +0,0 @@
-PaddlePaddle的数据提供(DataProvider)介绍
-========================================
-
-数据提供(DataProvider)是PaddlePaddle负责提供数据的模块。其作用是将训练数据传入内存或者显存，让神经网络可以进行训练。简单的使用，用户可以使用Python的 :code:`PyDataProvider` 来自定义传数据的过程。如果有更复杂的使用，或者需要更高的效率，用户也可以在C++端自定义一个 :code:`DataProvider` 。
-
-PaddlePaddle需要用户在网络配置(trainer_config.py)中定义使用哪种DataProvider及其参数，训练文件列表(train.list)和测试文件列表(test.list)。
-
-其中，train.list和test.list均为本地的两个文件(推荐直接放置到训练目录，以相对路径引用)。如果test.list不设置，或者设置为None，那么在训练过程中，不会执行测试操作。否则，会根据命令行参数指定的测试方式，在训练过程中进行测试，从而防止过拟合。
-
-一般情况下，train.list和test.list为纯文本文件，一行对应一个数据文件，数据文件存放在本地磁盘中。将文件的绝对路径或相对路径(相对于PaddlePaddle程序运行时的路径)写在train.list和test.list中。当然，train.list和test.list也可以放置hdfs文件路径，或者数据库连接地址等等。
-
-用户在DataProvider中需要实现如何访问其中每一个文件。DataProvider的具体用法和如何实现一个新的DataProvider，请参考下述文章:
-
-..	toctree::
-
-	pydataprovider2.rst
-	write_new_dataprovider.rst

doc_cn/ui/data_provider/mnist_config.py

@@ -5,5 +5,6 @@ define_py_data_sources2(
     test_list=None,
     module='mnist_provider',
     obj='process')
+
 img = data_layer(name='pixel', size=784)
 label = data_layer(name='label', size=10)

doc_cn/ui/data_provider/mnist_provider.py

@@ -1,22 +0,0 @@
-from paddle.trainer.PyDataProvider2 import *
-
-
-# Define a py data provider
-@provider(input_types=[dense_vector(28 * 28), integer_value(10)])
-def process(settings, filename):  # settings is not used currently.
-    f = open(filename, 'r')  # open one of training file
-
-    for line in f:  # read each line
-        label, pixel = line.split(';')
-
-        # get features and label
-        pixels_str = pixel.split(' ')
-
-        pixels_float = []
-        for each_pixel_str in pixels_str:
-            pixels_float.append(float(each_pixel_str))
-
-        # give data to paddle.
-        yield pixels_float, int(label)
-
-    f.close()  # close file

doc_cn/ui/data_provider/sentimental_provider.py

@@ -8,19 +8,16 @@ def on_init(settings, dictionary, **kwargs):
 
     # set input types in runtime. It will do the same thing as
     # @provider(input_types) will do, but it is set dynamically during runtime.
-    settings.input_types = [
+    settings.input_types = {
         # The text is a sequence of integer values, and each value is a word id.
         # The whole sequence is the sentences that we want to predict its
         # sentimental.
-        integer_value(
-            len(dictionary), seq_type=SequenceType),  # text input
+        'data': integer_value_sequence(len(dictionary)),  # text input
+        'label': integer_value(2)  # label positive/negative
+    }
 
-        # label positive/negative
-        integer_value(2)
-    ]
-
-    # save dictionary as settings.dictionary. It will be used in process
-    # method.
+    # save dictionary as settings.dictionary. 
+    # It will be used in process method.
     settings.dictionary = dictionary
 
 

doc_cn/ui/data_provider/write_new_dataprovider.rst

@@ -1,4 +0,0 @@
-自定义一个DataProvider
-====================
-
-TBD

doc_cn/ui/index.rst

@@ -8,8 +8,8 @@
 ..  toctree::
     :maxdepth: 1
 
-    data_provider/index.rst
-
+    data_provider/dataprovider.rst
+    data_provider/pydataprovider2.rst
 
 命令及命令行参数
 ================
@@ -23,9 +23,8 @@
 * `参数分类 <../../doc/ui/cmd_argument/argument_outline.html>`_
 * `参数描述 <../../doc/ui/cmd_argument/detail_introduction.html>`_
 
-
 预测
-====
+=======
 
 ..  toctree::
     :maxdepth: 1