Merge branch 'develop' of github.com:baidu/Paddle into feature/c_api

Author: reyoung

doc/api/v2/config/optimizer.rst

@@ -1,5 +1,3 @@
-..  _api_v2.optimizer:
-
 ==========
 Optimizer
 ==========

doc/api/v2/data.rst

@@ -1,6 +1,6 @@
-========
-Datasets
-========
+==================================
+Data Reader Interface and DataSets
+==================================
 
 
 DataTypes
@@ -49,7 +49,6 @@ mnist
     :members:
     :noindex:
 
-
 cifar
 +++++
 
@@ -61,7 +60,7 @@ conll05
 +++++++
 
 ..  automodule:: paddle.v2.dataset.conll05
-    :members:
+    :members: get_dict,get_embedding,test
     :noindex:
 
 imdb
@@ -85,6 +84,12 @@ movielens
     :members:
     :noindex:
 
+..  autoclass:: paddle.v2.dataset.movielens.MovieInfo
+    :noindex:
+    
+..  autoclass:: paddle.v2.dataset.movielens.UserInfo
+    :noindex:
+
 sentiment
 +++++++++
 
@@ -102,7 +107,7 @@ uci_housing
 wmt14
 +++++
 
-..  automodule:: paddle.v2.dataset.uci_housing
+..  automodule:: paddle.v2.dataset.wmt14
     :members:
     :noindex:
 

doc/api/v2/run_logic.rst

@@ -6,22 +6,26 @@ Parameters
 ==========
 
 ..  automodule:: paddle.v2.parameters
+    :members: Parameters
     :noindex:
 
 Trainer
 =======
 
 ..  automodule:: paddle.v2.trainer
+    :members: SGD
     :noindex:
 
 Event
 =====
 
 ..  automodule:: paddle.v2.event
+    :members:
     :noindex:
 
 Inference
 =========
 
 ..  autofunction:: paddle.v2.infer
     :noindex:
+    

doc/tutorials/embedding_model/index_cn.md

@@ -6,9 +6,10 @@
 
 ## 介绍 ###
 ### 中文字典 ###
-我们的字典使用内部的分词工具对百度知道和百度百科的语料进行分词后产生。分词风格如下： "《红楼梦》"将被分为 "《"，"红楼梦"，"》"，和 "《红楼梦》"。字典采用UTF8编码，输出有2列：词本身和词频。字典共包含 3206325个词和3个特殊标记：
+我们的字典使用内部的分词工具对百度知道和百度百科的语料进行分词后产生。分词风格如下： "《红楼梦》"将被分为 "《"，"红楼梦"，"》"，和 "《红楼梦》"。字典采用UTF8编码，输出有2列：词本身和词频。字典共包含 3206326个词和4个特殊标记：
   - `<s>`: 分词序列的开始
   - `<e>`: 分词序列的结束
+  - `PALCEHOLDER_JUST_IGNORE_THE_EMBEDDING`: 占位符，没有实际意义
   - `<unk>`: 未知词
 
 ### 中文词向量的预训练模型 ###

doc/tutorials/embedding_model/index_en.md

@@ -6,9 +6,10 @@ We thank @lipeng for the pull request that defined the model schemas and pretrai
 
 ## Introduction ###
 ### Chinese Word Dictionary ###
-Our Chinese-word dictionary is created on Baidu ZhiDao and Baidu Baike by using in-house word segmentor. For example, the participle of "《红楼梦》" is "《"，"红楼梦"，"》"，and "《红楼梦》". Our dictionary (using UTF-8 format) has has two columns: word and its frequency. The total word count is 3206325, including 3 special token:
+Our Chinese-word dictionary is created on Baidu ZhiDao and Baidu Baike by using in-house word segmentor. For example, the participle of "《红楼梦》" is "《"，"红楼梦"，"》"，and "《红楼梦》". Our dictionary (using UTF-8 format) has has two columns: word and its frequency. The total word count is 3206326, including 4 special token:
   - `<s>`: the start of a sequence
   - `<e>`: the end of a sequence
+  - `PALCEHOLDER_JUST_IGNORE_THE_EMBEDDING`: a placeholder, just ignore it and its embedding
   - `<unk>`: a word not included in dictionary
 
 ### Pretrained Chinese Word Embedding Model ###

paddle/gserver/gradientmachines/GradientMachine.h

@@ -134,9 +134,7 @@ class GradientMachine {
     backward(callback);
   }
 
-  virtual Argument getLayerOutput(const std::string& layerName) {
-    return *((Argument*)nullptr);
-  }
+  virtual Argument getLayerOutput(const std::string& layerName) = 0;
 
   // see comment in Layer.h for the function with the same name
   virtual void resetState() {}

paddle/scripts/docker/build.sh

@@ -9,6 +9,9 @@ if [ ${WITH_GPU} == "ON" ]; then
   GPU_DOCKER_PKG="python-pip python-dev"
 else
   BASE_IMAGE="python:2.7.13-slim"
+  # FIXME: python base image uses different python version than WITH_GPU
+  # need to change PYTHONHOME to /usr/local when using python base image
+  CPU_DOCKER_PYTHON_HOME_ENV="ENV PYTHONHOME /usr/local"
 fi
 
 DOCKERFILE_GPU_ENV=""
@@ -97,7 +100,8 @@ ADD build/*.deb /usr/local/opt/paddle/deb/
 RUN dpkg -i /usr/local/opt/paddle/deb/*.deb && \
     rm -f /usr/local/opt/paddle/deb/*.deb && \
     paddle version
-${DOCKERFILE_CUDNN_DSO} 
+${CPU_DOCKER_PYTHON_HOME_ENV}
+${DOCKERFILE_CUDNN_DSO}
 ${DOCKERFILE_GPU_ENV}
 # default command shows the paddle version and exit
 CMD ["paddle", "version"]

paddle/utils/.gitignore

@@ -1,2 +1 @@
 enable_virtualenv.c
-PythonUtil.cpp

paddle/utils/CMakeLists.txt

@@ -1,7 +1,4 @@
 # The utilities for paddle
-
-configure_file(PythonUtil.cpp.in ${PROJ_ROOT}/paddle/utils/PythonUtil.cpp)
-
 file(GLOB UTIL_HEADERS . *.h)
 file(GLOB UTIL_SOURCES . *.cpp)
 create_resources(enable_virtualenv.py enable_virtualenv.c)

paddle/utils/PythonUtil.cpp.in renamed to paddle/utils/PythonUtil.cpp

@@ -195,15 +195,6 @@ extern const char enable_virtualenv_py[];
 }
 void initPython(int argc, char** argv) {
 #ifndef PADDLE_NO_PYTHON
-  std::string pyHome;
-#if defined(__APPLE__) || defined(__OSX__)
-  pyHome = "/usr/local/Frameworks/Python.framework/Versions/2.7";
-  Py_SetPythonHome(const_cast<char*>(pyHome.c_str()));
-#endif
-  pyHome = "@PYTHON_INSTALL_DIR@"; // NOLINT
-  if (!pyHome.empty()) {
-    Py_SetPythonHome(const_cast<char*>(pyHome.c_str()));
-  }
   Py_SetProgramName(argv[0]);
   Py_Initialize();
   PySys_SetArgv(argc, argv);