Merge branch 'develop' into memory.set_input

.pre-commit-config.yaml

@@ -2,12 +2,12 @@
     sha: c25201a00e6b0514370501050cf2a8538ac12270
     hooks:
     -   id: remove-crlf
-        files: (?!.*third_party)^.*$
+        files: (?!.*third_party)^.*$ | (?!.*book)^.*$
 -   repo: https://github.com/reyoung/mirrors-yapf.git
     sha: v0.13.2
     hooks:
     - id: yapf
-      files: (.*\.(py|bzl)|BUILD|.*\.BUILD|WORKSPACE)$  # Bazel BUILD files follow Python syntax.
+      files: (.*\.(py|bzl)|BUILD|.*\.BUILD|WORKSPACE)$
 -   repo: https://github.com/pre-commit/pre-commit-hooks
     sha: 7539d8bd1a00a3c1bfd34cdb606d3a6372e83469
     hooks:
@@ -15,7 +15,7 @@
     -   id: check-merge-conflict
     -   id: check-symlinks
     -   id: detect-private-key
-        files: (?!.*third_party)^.*$
+        files: (?!.*third_party)^.*$ | (?!.*book)^.*$
     -   id: end-of-file-fixer
 -   repo: https://github.com/PaddlePaddle/clang-format-pre-commit-hook.git
     sha: 28c0ea8a67a3e2dbbf4822ef44e85b63a0080a29

.travis.yml

@@ -8,17 +8,10 @@ sudo: required
 dist: trusty
 os:
   - linux
-  - osx
 env:
   - JOB=DOCS
   - JOB=BUILD_AND_TEST
   - JOB=PRE_COMMIT
-matrix:
-  exclude:
-    - os: osx
-      env: JOB=DOCS  # Only generate documentation in linux.
-    - os: osx
-      env: JOB=PRE_COMMIT # Only check pre-commit hook in linux
 
 addons:
   apt:
@@ -52,11 +45,10 @@ before_install:
         fi
       fi
     fi
-  - if [[ "$TRAVIS_OS_NAME" == "osx" ]]; then paddle/scripts/travis/before_install.osx.sh; fi
   - if [[ "$JOB" == "PRE_COMMIT" ]]; then sudo ln -s /usr/bin/clang-format-3.8 /usr/bin/clang-format; fi
   # Paddle is using protobuf 3.1 currently. Protobuf 3.2 breaks the compatibility. So we specify the python 
   # protobuf version.
-  - pip install numpy wheel 'protobuf==3.1' sphinx recommonmark sphinx_rtd_theme virtualenv pre-commit requests==2.9.2 LinkChecker
+  - pip install numpy wheel 'protobuf==3.1' sphinx recommonmark sphinx-rtd-theme==0.1.9 virtualenv pre-commit requests==2.9.2 LinkChecker
 script:
   - paddle/scripts/travis/main.sh
 notifications:

paddle/scripts/docker/Dockerfile.gpu → Dockerfile

+# A image for building paddle binaries
+# Use cuda devel base image for both cpu and gpu environment
 FROM nvidia/cuda:7.5-cudnn5-devel-ubuntu14.04
 MAINTAINER PaddlePaddle Authors <paddle-dev@baidu.com>
 
 ARG DEBIAN_FRONTEND=noninteractive
 ARG UBUNTU_MIRROR
-RUN /bin/bash -c 'if [[ -n ${UBUNTU_MIRROR} ]]; then sed -i 's#http://archive.ubuntu.com#${UBUNTU_MIRROR}#g' /etc/apt/sources.list; fi'
+RUN /bin/bash -c 'if [[ -n ${UBUNTU_MIRROR} ]]; then sed -i 's#http://archive.ubuntu.com/ubuntu#${UBUNTU_MIRROR}#g' /etc/apt/sources.list; fi'
+
+# ENV variables
+ARG BUILD_WOBOQ
+ARG BUILD_AND_INSTALL
+ARG WITH_GPU
+ARG WITH_AVX
+ARG WITH_DOC
+ARG WITH_STYLE_CHECK
+
+ENV BUILD_WOBOQ=${BUILD_WOBOQ:-OFF}
+ENV BUILD_AND_INSTALL=${BUILD_AND_INSTALL:-OFF}
+ENV WITH_GPU=${WITH_AVX:-OFF}
+ENV WITH_AVX=${WITH_AVX:-ON}
+ENV WITH_DOC=${WITH_DOC:-OFF}
+ENV WITH_STYLE_CHECK=${WITH_STYLE_CHECK:-OFF}
+
+ENV HOME /root
+# Add bash enhancements
+COPY ./paddle/scripts/docker/root/ /root/
 
 RUN apt-get update && \
     apt-get install -y git python-pip python-dev openssh-server bison && \
     apt-get install -y wget unzip tar xz-utils bzip2 gzip coreutils && \
     apt-get install -y curl sed grep graphviz libjpeg-dev zlib1g-dev && \
     apt-get install -y python-numpy python-matplotlib gcc g++ gfortran && \
-    apt-get install -y automake clang-3.8 llvm-3.8 libclang-3.8-dev && \
+    apt-get install -y automake locales clang-format-3.8 && \
     apt-get clean -y
 
+# git credential to skip password typing
+RUN git config --global credential.helper store
+
+# Fix locales to en_US.UTF-8
+RUN localedef -i en_US -f UTF-8 en_US.UTF-8
+
 RUN pip install --upgrade pip && \
-    pip install -U protobuf && \
+    pip install -U 'protobuf==3.1.0' && \
     pip install -U wheel pillow BeautifulSoup && \
     pip install -U docopt PyYAML sphinx && \
-    pip install -U sphinx_rtd_theme recommonmark jupyter
+    pip install -U sphinx-rtd-theme==0.1.9 recommonmark && \
+    pip install -U pre-commit 'requests==2.9.2' jupyter
 
 RUN curl -sSL https://cmake.org/files/v3.4/cmake-3.4.1.tar.gz | tar -xz && \
-    cd cmake-3.4.1 && ./bootstrap && make -j4 && make install && \
+    cd cmake-3.4.1 && ./bootstrap && make -j `nproc` && make install && \
     cd .. && rm -rf cmake-3.4.1
 
-ARG BUILD_AND_INSTALL
-ARG WITH_AVX
-ARG WITH_DOC
-ARG WITH_STYLE_CHECK
+RUN apt-get install -y swig
 
-ENV BUILD_AND_INSTALL=${BUILD_AND_INSTALL:-OFF}
-ENV WITH_GPU=ON
-ENV WITH_AVX=${WITH_AVX:-ON}
-ENV WITH_DOC=${WITH_DOC:-ON}
-ENV WITH_STYLE_CHECK=${WITH_STYLE_CHECK:-OFF}
-
-RUN mkdir /paddle
-COPY . /paddle/
-RUN /paddle/paddle/scripts/docker/build.sh
 VOLUME ["/usr/share/nginx/html/data", "/usr/share/nginx/html/paddle"]
 
 # Configure OpenSSH server. c.f. https://docs.docker.com/engine/examples/running_ssh_service
@@ -46,12 +62,5 @@ RUN sed -ri 's/^PermitRootLogin\s+.*/PermitRootLogin yes/' /etc/ssh/sshd_config
 RUN sed -ri 's/UsePAM yes/#UsePAM yes/g' /etc/ssh/sshd_config
 EXPOSE 22
 
-# Jupyter Notebook directory.
-RUN mkdir /notes/
-WORKDIR "/notes"
-EXPOSE 8888
-
-RUN mkdir -p /opt/bin
-COPY ./paddle/scripts/docker/entrypoint /opt/bin/
-
-CMD ["/opt/bin/entrypoint"]
+# development image default do build work
+CMD ["bash", "/paddle/paddle/scripts/docker/build.sh"]

authors

@@ -29,13 +29,16 @@ Luo, Tao
 Lyu, Qin
 Mao, Hongyue
 Qian, Xiaojun
+Qiao, Longfei
 Qi, Jun
 Qin, Duohao
 Shen, Guolong
 Shi, Guangchuan
 Song, Xiang
+Wang, Helin
 Wang, Jiang
 Wang, Yanfei
+Wang, Yi
 Wang, Yong
 Weng, Renliang
 Xu, Tianbing

cmake/FindSphinx.cmake

@@ -72,7 +72,7 @@ function( Sphinx_add_target target_name builder conf cache source destination )
     ${source}
     ${destination}
     COMMENT "Generating sphinx documentation: ${builder}"
-    COMMAND ln -sf ${destination}/index_*.html ${destination}/index.html
+    COMMAND cd ${destination} && ln -s ./index_*.html index.html
     )
 
   set_property(

cmake/cblas.cmake

@@ -16,7 +16,8 @@
 set(CBLAS_FOUND OFF)
 
 ## Find MKL First.
-set(MKL_ROOT $ENV{MKLROOT} CACHE PATH "Folder contains MKL")
+set(INTEL_ROOT "/opt/intel" CACHE PATH "Folder contains intel libs")
+set(MKL_ROOT ${INTEL_ROOT}/mkl CACHE PATH "Folder contains MKL")
 
 find_path(MKL_INCLUDE_DIR mkl.h PATHS
   ${MKL_ROOT}/include)

cmake/ccache.cmake

 # Use ccache if found ccache program
 
-find_program(CCACHE_FOUND ccache)
+find_program(CCACHE_PATH ccache)
 
-if(CCACHE_FOUND)
+if(CCACHE_PATH)
     message(STATUS "Ccache is founded, use ccache to speed up compile.")
-    set_property(GLOBAL PROPERTY RULE_LAUNCH_COMPILE ccache)
-    set_property(GLOBAL PROPERTY RULE_LAUNCH_LINK ccache)
-endif(CCACHE_FOUND)
\ No newline at end of file
+    set_property(GLOBAL PROPERTY RULE_LAUNCH_COMPILE ${CCACHE_PATH})
+    set_property(GLOBAL PROPERTY RULE_LAUNCH_LINK ${CCACHE_PATH})
+endif(CCACHE_PATH)

cmake/coverallsGcovJsons.cmake

@@ -110,14 +110,13 @@ endmacro()
 
 # Get the coverage data.
 file(GLOB_RECURSE GCDA_FILES "${COV_PATH}" "*.gcda")
-message("GCDA files:")
+message("Process GCDA files:")
+message("===============================")
 
 # Get a list of all the object directories needed by gcov
 # (The directories the .gcda files and .o files are found in)
 # and run gcov on those.
 foreach(GCDA ${GCDA_FILES})
-	message("Process: ${GCDA}")
-	message("------------------------------------------------------------------------------")
 	get_filename_component(GCDA_DIR ${GCDA} PATH)
 
 	#
@@ -135,7 +134,7 @@ foreach(GCDA ${GCDA_FILES})
 	# If -p is not specified then the file is named only "the_file.c.gcov"
 	#
 	execute_process(
-		COMMAND ${GCOV_EXECUTABLE} -p -o ${GCDA_DIR} ${GCDA}
+		COMMAND "${GCOV_EXECUTABLE} -p -o ${GCDA_DIR} ${GCDA} >/dev/null"
 		WORKING_DIRECTORY ${GCDA_DIR}
 	)
 endforeach()
@@ -383,7 +382,6 @@ foreach(NOT_COVERED_SRC ${COVERAGE_SRCS_REMAINING})
 	set(GCOV_FILE_COVERAGE "${GCOV_FILE_COVERAGE}]")
 
 	# Generate the final JSON for this file.
-	message("Generate JSON for non-gcov file: ${NOT_COVERED_SRC}...")
 	string(CONFIGURE ${SRC_FILE_TEMPLATE} FILE_JSON)
 	set(JSON_GCOV_FILES "${JSON_GCOV_FILES}${FILE_JSON}, ")
 endforeach()

cmake/external/protobuf.cmake

@@ -14,46 +14,58 @@
 
 INCLUDE(ExternalProject)
 
-SET(PROTOBUF_SOURCES_DIR ${THIRD_PARTY_PATH}/protobuf)
-SET(PROTOBUF_INSTALL_DIR ${THIRD_PARTY_PATH}/install/protobuf)
-SET(PROTOBUF_INCLUDE_DIR "${PROTOBUF_INSTALL_DIR}/include" CACHE PATH "protobuf include directory." FORCE)
+FIND_PACKAGE(Protobuf 3.1)
 
-INCLUDE_DIRECTORIES(${PROTOBUF_INCLUDE_DIR})
+IF(PROTOBUF_FOUND)
+    EXEC_PROGRAM(${PROTOBUF_PROTOC_EXECUTABLE} ARGS --version OUTPUT_VARIABLE PROTOBUF_VERSION)
+    STRING(REGEX MATCH "[0-9]+.[0-9]+" PROTOBUF_VERSION "${PROTOBUF_VERSION}")
+    IF (${PROTOBUF_VERSION} VERSION_LESS "3.1.0")
+        SET(PROTOBUF_FOUND OFF)
+    ENDIF()
+ENDIF(PROTOBUF_FOUND)
+
+IF(NOT PROTOBUF_FOUND)
+    SET(PROTOBUF_SOURCES_DIR ${THIRD_PARTY_PATH}/protobuf)
+    SET(PROTOBUF_INSTALL_DIR ${THIRD_PARTY_PATH}/install/protobuf)
+    SET(PROTOBUF_INCLUDE_DIR "${PROTOBUF_INSTALL_DIR}/include" CACHE PATH "protobuf include directory." FORCE)
+
+    IF(WIN32)
+        SET(PROTOBUF_LITE_LIBRARY
+            "${PROTOBUF_INSTALL_DIR}/lib/libprotobuf-lite.lib" CACHE FILEPATH "protobuf lite library." FORCE)
+        SET(PROTOBUF_LIBRARY
+            "${PROTOBUF_INSTALL_DIR}/lib/libprotobuf.lib" CACHE FILEPATH "protobuf library." FORCE)
+        SET(PROTOBUF_PROTOC_LIBRARY
+            "${PROTOBUF_INSTALL_DIR}/lib/libprotoc.lib" CACHE FILEPATH "protoc library." FORCE)
+        SET(PROTOBUF_PROTOC_EXECUTABLE "${PROTOBUF_INSTALL_DIR}/bin/protoc.exe" CACHE FILEPATH "protobuf executable." FORCE)
+    ELSE(WIN32)
+        SET(PROTOBUF_LITE_LIBRARY
+            "${PROTOBUF_INSTALL_DIR}/lib/libprotobuf-lite.a" CACHE FILEPATH "protobuf lite library." FORCE)
+        SET(PROTOBUF_LIBRARY
+            "${PROTOBUF_INSTALL_DIR}/lib/libprotobuf.a" CACHE FILEPATH "protobuf library." FORCE)
+        SET(PROTOBUF_PROTOC_LIBRARY
+            "${PROTOBUF_INSTALL_DIR}/lib/libprotoc.a" CACHE FILEPATH "protoc library." FORCE)
+        SET(PROTOBUF_PROTOC_EXECUTABLE "${PROTOBUF_INSTALL_DIR}/bin/protoc" CACHE FILEPATH "protobuf executable." FORCE)
+    ENDIF(WIN32)
 
-IF(WIN32)
-  SET(PROTOBUF_LITE_LIBRARY
-        "${PROTOBUF_INSTALL_DIR}/lib/libprotobuf-lite.lib" CACHE FILEPATH "protobuf lite library." FORCE)
-  SET(PROTOBUF_LIBRARY
-        "${PROTOBUF_INSTALL_DIR}/lib/libprotobuf.lib" CACHE FILEPATH "protobuf library." FORCE)
-  SET(PROTOBUF_PROTOC_LIBRARY
-        "${PROTOBUF_INSTALL_DIR}/lib/libprotoc.lib" CACHE FILEPATH "protoc library." FORCE)
-  SET(PROTOBUF_PROTOC_EXECUTABLE "${PROTOBUF_INSTALL_DIR}/bin/protoc.exe" CACHE FILEPATH "protobuf executable." FORCE)
-ELSE(WIN32)
-  SET(PROTOBUF_LITE_LIBRARY
-        "${PROTOBUF_INSTALL_DIR}/lib/libprotobuf-lite.a" CACHE FILEPATH "protobuf lite library." FORCE)
-  SET(PROTOBUF_LIBRARY
-        "${PROTOBUF_INSTALL_DIR}/lib/libprotobuf.a" CACHE FILEPATH "protobuf library." FORCE)
-  SET(PROTOBUF_PROTOC_LIBRARY
-        "${PROTOBUF_INSTALL_DIR}/lib/libprotoc.a" CACHE FILEPATH "protoc library." FORCE)
-  SET(PROTOBUF_PROTOC_EXECUTABLE "${PROTOBUF_INSTALL_DIR}/bin/protoc" CACHE FILEPATH "protobuf executable." FORCE)
-ENDIF(WIN32)
-
-ExternalProject_Add(
-  protobuf
-  ${EXTERNAL_PROJECT_LOG_ARGS}
-  PREFIX          ${PROTOBUF_SOURCES_DIR}
-  UPDATE_COMMAND  ""
-  DEPENDS         zlib
-  GIT_REPOSITORY  "https://github.com/google/protobuf.git"
-  GIT_TAG         "9f75c5aa851cd877fb0d93ccc31b8567a6706546"
-  CONFIGURE_COMMAND
-    ${CMAKE_COMMAND} ${PROTOBUF_SOURCES_DIR}/src/protobuf/cmake
-    -Dprotobuf_BUILD_TESTS=OFF
-    -DZLIB_ROOT:FILEPATH=${ZLIB_ROOT}
-    -DCMAKE_POSITION_INDEPENDENT_CODE=ON
-    -DCMAKE_BUILD_TYPE=Release
-    -DCMAKE_INSTALL_PREFIX=${PROTOBUF_INSTALL_DIR}
-    -DCMAKE_INSTALL_LIBDIR=lib
-)
-
-LIST(APPEND external_project_dependencies protobuf)
+    ExternalProject_Add(
+        protobuf
+        ${EXTERNAL_PROJECT_LOG_ARGS}
+        PREFIX          ${PROTOBUF_SOURCES_DIR}
+        UPDATE_COMMAND  ""
+        DEPENDS         zlib
+        GIT_REPOSITORY  "https://github.com/google/protobuf.git"
+        GIT_TAG         "9f75c5aa851cd877fb0d93ccc31b8567a6706546"
+        CONFIGURE_COMMAND
+        ${CMAKE_COMMAND} ${PROTOBUF_SOURCES_DIR}/src/protobuf/cmake
+        -Dprotobuf_BUILD_TESTS=OFF
+        -DZLIB_ROOT:FILEPATH=${ZLIB_ROOT}
+        -DCMAKE_POSITION_INDEPENDENT_CODE=ON
+        -DCMAKE_BUILD_TYPE=Release
+        -DCMAKE_INSTALL_PREFIX=${PROTOBUF_INSTALL_DIR}
+        -DCMAKE_INSTALL_LIBDIR=lib
+    )
+
+    LIST(APPEND external_project_dependencies protobuf)
+ENDIF(NOT PROTOBUF_FOUND)
+
+INCLUDE_DIRECTORIES(${PROTOBUF_INCLUDE_DIR})

cmake/external/python.cmake

@@ -221,7 +221,3 @@ ENDIF(PYTHONLIBS_FOUND AND PYTHONINTERP_FOUND)
 
 INCLUDE_DIRECTORIES(${PYTHON_INCLUDE_DIR})
 INCLUDE_DIRECTORIES(${PYTHON_NUMPY_INCLUDE_DIR})
-
-MESSAGE("[Paddle] Python Executable: ${PYTHON_EXECUTABLE}")
-MESSAGE("[Paddle] Python Include: ${PYTHON_INCLUDE_DIRS}")
-MESSAGE("[Paddle] Python Libraries: ${PYTHON_LIBRARIES}")

cmake/util.cmake

@@ -71,21 +71,10 @@ function(link_paddle_exe TARGET_NAME)
         generate_rdma_links()
     endif()
 
-    if(WITH_METRIC)
-        if(WITH_GPU)
-            set(METRIC_LIBS paddle_metric_learning paddle_dserver_lib metric metric_cpu)
-        else()
-            set(METRIC_LIBS paddle_metric_learning paddle_dserver_lib metric_cpu)
-        endif()
-    else()
-        set(METRIC_LIBS "")
-    endif()
-
     target_circle_link_libraries(${TARGET_NAME}
         ARCHIVE_START
         paddle_gserver
         paddle_function
-        ${METRIC_LIBS}
         ARCHIVE_END
         paddle_pserver
         paddle_trainer_lib
@@ -95,7 +84,6 @@ function(link_paddle_exe TARGET_NAME)
         paddle_parameter
         paddle_proto
         paddle_cuda
-        ${METRIC_LIBS}
         ${EXTERNAL_LIBS}
         ${CMAKE_THREAD_LIBS_INIT}
         ${CMAKE_DL_LIBS}

demo/image_classification/api_v2_resnet.py

+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# 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.
+
+import paddle.v2 as paddle
+
+__all__ = ['resnet_cifar10']
+
+
+def conv_bn_layer(input,
+                  ch_out,
+                  filter_size,
+                  stride,
+                  padding,
+                  active_type=paddle.activation.Relu(),
+                  ch_in=None):
+    tmp = paddle.layer.img_conv(
+        input=input,
+        filter_size=filter_size,
+        num_channels=ch_in,
+        num_filters=ch_out,
+        stride=stride,
+        padding=padding,
+        act=paddle.activation.Linear(),
+        bias_attr=False)
+    return paddle.layer.batch_norm(input=tmp, act=active_type)
+
+
+def shortcut(ipt, n_in, n_out, stride):
+    if n_in != n_out:
+        return conv_bn_layer(ipt, n_out, 1, stride, 0,
+                             paddle.activation.Linear())
+    else:
+        return ipt
+
+
+def basicblock(ipt, ch_out, stride):
+    ch_in = ch_out * 2
+    tmp = conv_bn_layer(ipt, ch_out, 3, stride, 1)
+    tmp = conv_bn_layer(tmp, ch_out, 3, 1, 1, paddle.activation.Linear())
+    short = shortcut(ipt, ch_in, ch_out, stride)
+    return paddle.layer.addto(input=[tmp, short], act=paddle.activation.Relu())
+
+
+def layer_warp(block_func, ipt, features, count, stride):
+    tmp = block_func(ipt, features, stride)
+    for i in range(1, count):
+        tmp = block_func(tmp, features, 1)
+    return tmp
+
+
+def resnet_cifar10(ipt, depth=32):
+    # depth should be one of 20, 32, 44, 56, 110, 1202
+    assert (depth - 2) % 6 == 0
+    n = (depth - 2) / 6
+    nStages = {16, 64, 128}
+    conv1 = conv_bn_layer(
+        ipt, ch_in=3, ch_out=16, filter_size=3, stride=1, padding=1)
+    res1 = layer_warp(basicblock, conv1, 16, n, 1)
+    res2 = layer_warp(basicblock, res1, 32, n, 2)
+    res3 = layer_warp(basicblock, res2, 64, n, 2)
+    pool = paddle.layer.img_pool(
+        input=res3, pool_size=8, stride=1, pool_type=paddle.pooling.Avg())
+    return pool

demo/image_classification/api_v2_train.py

+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# 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
+
+import sys
+
+import paddle.v2 as paddle
+
+from api_v2_vgg import vgg_bn_drop
+
+
+def main():
+    datadim = 3 * 32 * 32
+    classdim = 10
+
+    # PaddlePaddle init
+    paddle.init(use_gpu=False, trainer_count=1)
+
+    image = paddle.layer.data(
+        name="image", type=paddle.data_type.dense_vector(datadim))
+
+    # Add neural network config
+    # option 1. resnet
+    # net = resnet_cifar10(image, depth=32)
+    # option 2. vgg
+    net = vgg_bn_drop(image)
+
+    out = paddle.layer.fc(input=net,
+                          size=classdim,
+                          act=paddle.activation.Softmax())
+
+    lbl = paddle.layer.data(
+        name="label", type=paddle.data_type.integer_value(classdim))
+    cost = paddle.layer.classification_cost(input=out, label=lbl)
+
+    # Create parameters
+    parameters = paddle.parameters.create(cost)
+
+    # Create optimizer
+    momentum_optimizer = paddle.optimizer.Momentum(
+        momentum=0.9,
+        regularization=paddle.optimizer.L2Regularization(rate=0.0002 * 128),
+        learning_rate=0.1 / 128.0,
+        learning_rate_decay_a=0.1,
+        learning_rate_decay_b=50000 * 100,
+        learning_rate_schedule='discexp',
+        batch_size=128)
+
+    # End batch and end pass event handler
+    def event_handler(event):
+        if isinstance(event, paddle.event.EndIteration):
+            if event.batch_id % 100 == 0:
+                print "\nPass %d, Batch %d, Cost %f, %s" % (
+                    event.pass_id, event.batch_id, event.cost, event.metrics)
+            else:
+                sys.stdout.write('.')
+                sys.stdout.flush()
+        if isinstance(event, paddle.event.EndPass):
+            result = trainer.test(
+                reader=paddle.batch(
+                    paddle.dataset.cifar.test10(), batch_size=128),
+                feeding={'image': 0,
+                         'label': 1})
+            print "\nTest with Pass %d, %s" % (event.pass_id, result.metrics)
+
+    # Create trainer
+    trainer = paddle.trainer.SGD(cost=cost,
+                                 parameters=parameters,
+                                 update_equation=momentum_optimizer)
+    trainer.train(
+        reader=paddle.batch(
+            paddle.reader.shuffle(
+                paddle.dataset.cifar.train10(), buf_size=50000),
+            batch_size=128),
+        num_passes=5,
+        event_handler=event_handler,
+        feeding={'image': 0,
+                 'label': 1})
+
+
+if __name__ == '__main__':
+    main()

demo/image_classification/api_v2_vgg.py

+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# 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.
+
+import paddle.v2 as paddle
+
+__all__ = ['vgg_bn_drop']
+
+
+def vgg_bn_drop(input):
+    def conv_block(ipt, num_filter, groups, dropouts, num_channels=None):
+        return paddle.networks.img_conv_group(
+            input=ipt,
+            num_channels=num_channels,
+            pool_size=2,
+            pool_stride=2,
+            conv_num_filter=[num_filter] * groups,
+            conv_filter_size=3,
+            conv_act=paddle.activation.Relu(),
+            conv_with_batchnorm=True,
+            conv_batchnorm_drop_rate=dropouts,
+            pool_type=paddle.pooling.Max())
+
+    conv1 = conv_block(input, 64, 2, [0.3, 0], 3)
+    conv2 = conv_block(conv1, 128, 2, [0.4, 0])
+    conv3 = conv_block(conv2, 256, 3, [0.4, 0.4, 0])
+    conv4 = conv_block(conv3, 512, 3, [0.4, 0.4, 0])
+    conv5 = conv_block(conv4, 512, 3, [0.4, 0.4, 0])
+
+    drop = paddle.layer.dropout(input=conv5, dropout_rate=0.5)
+    fc1 = paddle.layer.fc(input=drop, size=512, act=paddle.activation.Linear())
+    bn = paddle.layer.batch_norm(
+        input=fc1,
+        act=paddle.activation.Relu(),
+        layer_attr=paddle.attr.Extra(drop_rate=0.5))
+    fc2 = paddle.layer.fc(input=bn, size=512, act=paddle.activation.Linear())
+    return fc2

demo/image_classification/prediction.py

@@ -126,7 +126,7 @@ class ImageClassifier():
         # For oversampling, average predictions across crops.
         # If not, the shape of output[name]: (1, class_number),
         # the mean is also applicable.
-        return output[output_layer].mean(0)
+        return output[output_layer]['value'].mean(0)
 
     def predict(self, image=None, output_layer=None):
         assert isinstance(image, basestring)

demo/introduction/api_train_v2.py

+import paddle.v2 as paddle
+import paddle.v2.dataset.uci_housing as uci_housing
+
+
+def main():
+    # init
+    paddle.init(use_gpu=False, trainer_count=1)
+
+    # network config
+    x = paddle.layer.data(name='x', type=paddle.data_type.dense_vector(13))
+    y_predict = paddle.layer.fc(input=x,
+                                param_attr=paddle.attr.Param(name='w'),
+                                size=1,
+                                act=paddle.activation.Linear(),
+                                bias_attr=paddle.attr.Param(name='b'))
+    y = paddle.layer.data(name='y', type=paddle.data_type.dense_vector(1))
+    cost = paddle.layer.mse_cost(input=y_predict, label=y)
+
+    # create parameters
+    parameters = paddle.parameters.create(cost)
+
+    # create optimizer
+    optimizer = paddle.optimizer.Momentum(momentum=0)
+
+    trainer = paddle.trainer.SGD(cost=cost,
+                                 parameters=parameters,
+                                 update_equation=optimizer)
+
+    # event_handler to print training and testing info
+    def event_handler(event):
+        if isinstance(event, paddle.event.EndIteration):
+            if event.batch_id % 100 == 0:
+                print "Pass %d, Batch %d, Cost %f" % (
+                    event.pass_id, event.batch_id, event.cost)
+
+        if isinstance(event, paddle.event.EndPass):
+            if (event.pass_id + 1) % 10 == 0:
+                result = trainer.test(
+                    reader=paddle.batch(
+                        uci_housing.test(), batch_size=2),
+                    feeding={'x': 0,
+                             'y': 1})
+                print "Test %d, %.2f" % (event.pass_id, result.cost)
+
+    # training
+    trainer.train(
+        reader=paddle.batch(
+            paddle.reader.shuffle(
+                uci_housing.train(), buf_size=500),
+            batch_size=2),
+        feeding={'x': 0,
+                 'y': 1},
+        event_handler=event_handler,
+        num_passes=30)
+
+
+if __name__ == '__main__':
+    main()

demo/introduction/trainer_config.py

@@ -34,5 +34,5 @@ y_predict = fc_layer(
     size=1,
     act=LinearActivation(),
     bias_attr=ParamAttr(name='b'))
-cost = regression_cost(input=y_predict, label=y)
+cost = mse_cost(input=y_predict, label=y)
 outputs(cost)

demo/mnist/.gitignore

@@ -5,3 +5,6 @@ plot.png
 train.log
 *pyc
 .ipynb_checkpoints
+params.pkl
+params.tar
+params.tar.gz

demo/mnist/api_train.py

@@ -6,25 +6,15 @@ passed to C++ side of Paddle.
 
 The user api could be simpler and carefully designed.
 """
-import py_paddle.swig_paddle as api
-from py_paddle import DataProviderConverter
-import paddle.trainer.PyDataProvider2 as dp
-import numpy as np
 import random
-from mnist_util import read_from_mnist
-from paddle.trainer_config_helpers import *
-import paddle.v2
 
+import numpy as np
+import paddle.v2 as paddle_v2
+import py_paddle.swig_paddle as api
+from paddle.trainer_config_helpers import *
+from py_paddle import DataProviderConverter
 
-def network_config():
-    imgs = data_layer(name='pixel', size=784)
-    hidden1 = fc_layer(input=imgs, size=200)
-    hidden2 = fc_layer(input=hidden1, size=200)
-    inference = fc_layer(input=hidden2, size=10, act=SoftmaxActivation())
-    cost = classification_cost(
-        input=inference, label=data_layer(
-            name='label', size=10))
-    outputs(cost)
+from mnist_util import read_from_mnist
 
 
 def init_parameter(network):
@@ -67,7 +57,7 @@ def input_order_converter(generator):
 def main():
     api.initPaddle("-use_gpu=false", "-trainer_count=4")  # use 4 cpu cores
 
-    optimizer = paddle.v2.optimizer.Adam(
+    optimizer = paddle_v2.optimizer.Adam(
         learning_rate=1e-4,
         batch_size=1000,
         model_average=ModelAverage(average_window=0.5),
@@ -79,8 +69,20 @@ def main():
     updater = optimizer.create_local_updater()
     assert isinstance(updater, api.ParameterUpdater)
 
+    # define network
+    images = paddle_v2.layer.data(
+        name='pixel', type=paddle_v2.data_type.dense_vector(784))
+    label = paddle_v2.layer.data(
+        name='label', type=paddle_v2.data_type.integer_value(10))
+    hidden1 = paddle_v2.layer.fc(input=images, size=200)
+    hidden2 = paddle_v2.layer.fc(input=hidden1, size=200)
+    inference = paddle_v2.layer.fc(input=hidden2,
+                                   size=10,
+                                   act=paddle_v2.activation.Softmax())
+    cost = paddle_v2.layer.classification_cost(input=inference, label=label)
+
     # Create Simple Gradient Machine.
-    model_config = parse_network_config(network_config)
+    model_config = paddle_v2.layer.parse_network(cost)
     m = api.GradientMachine.createFromConfigProto(model_config,
                                                   api.CREATE_MODE_NORMAL,
                                                   optimizer.enable_types())
@@ -97,8 +99,7 @@ def main():
 
     # DataProvider Converter is a utility convert Python Object to Paddle C++
     # Input. The input format is as same as Paddle's DataProvider.
-    converter = DataProviderConverter(
-        input_types=[dp.dense_vector(784), dp.integer_value(10)])
+    converter = DataProviderConverter(input_types=[images.type, label.type])
 
     train_file = './data/raw_data/train'
     test_file = './data/raw_data/t10k'

demo/mnist/api_train_v2.py

+import paddle.v2 as paddle
+import gzip
+
+
+def softmax_regression(img):
+    predict = paddle.layer.fc(input=img,
+                              size=10,
+                              act=paddle.activation.Softmax())
+    return predict
+
+
+def multilayer_perceptron(img):
+    # The first fully-connected layer
+    hidden1 = paddle.layer.fc(input=img, size=128, act=paddle.activation.Relu())
+    # The second fully-connected layer and the according activation function
+    hidden2 = paddle.layer.fc(input=hidden1,
+                              size=64,
+                              act=paddle.activation.Relu())
+    # The thrid fully-connected layer, note that the hidden size should be 10,
+    # which is the number of unique digits
+    predict = paddle.layer.fc(input=hidden2,
+                              size=10,
+                              act=paddle.activation.Softmax())
+    return predict
+
+
+def convolutional_neural_network(img):
+    # first conv layer
+    conv_pool_1 = paddle.networks.simple_img_conv_pool(
+        input=img,
+        filter_size=5,
+        num_filters=20,
+        num_channel=1,
+        pool_size=2,
+        pool_stride=2,
+        act=paddle.activation.Tanh())
+    # second conv layer
+    conv_pool_2 = paddle.networks.simple_img_conv_pool(
+        input=conv_pool_1,
+        filter_size=5,
+        num_filters=50,
+        num_channel=20,
+        pool_size=2,
+        pool_stride=2,
+        act=paddle.activation.Tanh())
+    # The first fully-connected layer
+    fc1 = paddle.layer.fc(input=conv_pool_2,
+                          size=128,
+                          act=paddle.activation.Tanh())
+    # The softmax layer, note that the hidden size should be 10,
+    # which is the number of unique digits
+    predict = paddle.layer.fc(input=fc1,
+                              size=10,
+                              act=paddle.activation.Softmax())
+    return predict
+
+
+def main():
+    paddle.init(use_gpu=False, trainer_count=1)
+
+    # define network topology
+    images = paddle.layer.data(
+        name='pixel', type=paddle.data_type.dense_vector(784))
+    label = paddle.layer.data(
+        name='label', type=paddle.data_type.integer_value(10))
+
+    # Here we can build the prediction network in different ways. Please
+    # choose one by uncomment corresponding line.
+    predict = softmax_regression(images)
+    #predict = multilayer_perceptron(images)
+    #predict = convolutional_neural_network(images)
+
+    cost = paddle.layer.classification_cost(input=predict, label=label)
+
+    try:
+        with gzip.open('params.tar.gz', 'r') as f:
+            parameters = paddle.parameters.Parameters.from_tar(f)
+    except IOError:
+        parameters = paddle.parameters.create(cost)
+
+    optimizer = paddle.optimizer.Momentum(
+        learning_rate=0.1 / 128.0,
+        momentum=0.9,
+        regularization=paddle.optimizer.L2Regularization(rate=0.0005 * 128))
+
+    trainer = paddle.trainer.SGD(cost=cost,
+                                 parameters=parameters,
+                                 update_equation=optimizer)
+
+    lists = []
+
+    def event_handler(event):
+        if isinstance(event, paddle.event.EndIteration):
+            if event.batch_id % 1000 == 0:
+                print "Pass %d, Batch %d, Cost %f, %s" % (
+                    event.pass_id, event.batch_id, event.cost, event.metrics)
+
+                with gzip.open('params.tar.gz', 'w') as f:
+                    parameters.to_tar(f)
+
+        elif isinstance(event, paddle.event.EndPass):
+            result = trainer.test(reader=paddle.batch(
+                paddle.dataset.mnist.test(), batch_size=128))
+            print "Test with Pass %d, Cost %f, %s\n" % (
+                event.pass_id, result.cost, result.metrics)
+            lists.append((event.pass_id, result.cost,
+                          result.metrics['classification_error_evaluator']))
+
+    trainer.train(
+        reader=paddle.batch(
+            paddle.reader.shuffle(
+                paddle.dataset.mnist.train(), buf_size=8192),
+            batch_size=128),
+        event_handler=event_handler,
+        num_passes=100)
+
+    # find the best pass
+    best = sorted(lists, key=lambda list: float(list[1]))[0]
+    print 'Best pass is %s, testing Avgcost is %s' % (best[0], best[1])
+    print 'The classification accuracy is %.2f%%' % (100 - float(best[2]) * 100)
+
+    test_creator = paddle.dataset.mnist.test()
+    test_data = []
+    for item in test_creator():
+        test_data.append((item[0], ))
+        if len(test_data) == 100:
+            break
+
+    # output is a softmax layer. It returns probabilities.
+    # Shape should be (100, 10)
+    probs = paddle.infer(
+        output_layer=predict, parameters=parameters, input=test_data)
+    print probs.shape
+
+
+if __name__ == '__main__':
+    main()

demo/model_zoo/resnet/classify.py

@@ -156,7 +156,7 @@ class ImageClassifier():
             # For oversampling, average predictions across crops.
             # If not, the shape of output[name]: (1, class_number),
             # the mean is also applicable.
-            res[name] = output[name].mean(0)
+            res[name] = output[name]['value'].mean(0)
 
         return res
 

demo/recommendation/api_train_v2.py

+import paddle.v2 as paddle
+import cPickle
+import copy
+
+
+def main():
+    paddle.init(use_gpu=False)
+    movie_title_dict = paddle.dataset.movielens.get_movie_title_dict()
+    uid = paddle.layer.data(
+        name='user_id',
+        type=paddle.data_type.integer_value(
+            paddle.dataset.movielens.max_user_id() + 1))
+    usr_emb = paddle.layer.embedding(input=uid, size=32)
+
+    usr_gender_id = paddle.layer.data(
+        name='gender_id', type=paddle.data_type.integer_value(2))
+    usr_gender_emb = paddle.layer.embedding(input=usr_gender_id, size=16)
+
+    usr_age_id = paddle.layer.data(
+        name='age_id',
+        type=paddle.data_type.integer_value(
+            len(paddle.dataset.movielens.age_table)))
+    usr_age_emb = paddle.layer.embedding(input=usr_age_id, size=16)
+
+    usr_job_id = paddle.layer.data(
+        name='job_id',
+        type=paddle.data_type.integer_value(paddle.dataset.movielens.max_job_id(
+        ) + 1))
+
+    usr_job_emb = paddle.layer.embedding(input=usr_job_id, size=16)
+
+    usr_combined_features = paddle.layer.fc(
+        input=[usr_emb, usr_gender_emb, usr_age_emb, usr_job_emb],
+        size=200,
+        act=paddle.activation.Tanh())
+
+    mov_id = paddle.layer.data(
+        name='movie_id',
+        type=paddle.data_type.integer_value(
+            paddle.dataset.movielens.max_movie_id() + 1))
+    mov_emb = paddle.layer.embedding(input=mov_id, size=32)
+
+    mov_categories = paddle.layer.data(
+        name='category_id',
+        type=paddle.data_type.sparse_binary_vector(
+            len(paddle.dataset.movielens.movie_categories())))
+
+    mov_categories_hidden = paddle.layer.fc(input=mov_categories, size=32)
+
+    mov_title_id = paddle.layer.data(
+        name='movie_title',
+        type=paddle.data_type.integer_value_sequence(len(movie_title_dict)))
+    mov_title_emb = paddle.layer.embedding(input=mov_title_id, size=32)
+    mov_title_conv = paddle.networks.sequence_conv_pool(
+        input=mov_title_emb, hidden_size=32, context_len=3)
+
+    mov_combined_features = paddle.layer.fc(
+        input=[mov_emb, mov_categories_hidden, mov_title_conv],
+        size=200,
+        act=paddle.activation.Tanh())
+
+    inference = paddle.layer.cos_sim(
+        a=usr_combined_features, b=mov_combined_features, size=1, scale=5)
+    cost = paddle.layer.mse_cost(
+        input=inference,
+        label=paddle.layer.data(
+            name='score', type=paddle.data_type.dense_vector(1)))
+
+    parameters = paddle.parameters.create(cost)
+
+    trainer = paddle.trainer.SGD(cost=cost,
+                                 parameters=parameters,
+                                 update_equation=paddle.optimizer.Adam(
+                                     learning_rate=1e-4))
+    feeding = {
+        'user_id': 0,
+        'gender_id': 1,
+        'age_id': 2,
+        'job_id': 3,
+        'movie_id': 4,
+        'category_id': 5,
+        'movie_title': 6,
+        'score': 7
+    }
+
+    def event_handler(event):
+        if isinstance(event, paddle.event.EndIteration):
+            if event.batch_id % 100 == 0:
+                print "Pass %d Batch %d Cost %.2f" % (
+                    event.pass_id, event.batch_id, event.cost)
+
+    trainer.train(
+        reader=paddle.batch(
+            paddle.reader.shuffle(
+                paddle.dataset.movielens.train(), buf_size=8192),
+            batch_size=256),
+        event_handler=event_handler,
+        feeding=feeding,
+        num_passes=1)
+
+    user_id = 234
+    movie_id = 345
+
+    user = paddle.dataset.movielens.user_info()[user_id]
+    movie = paddle.dataset.movielens.movie_info()[movie_id]
+
+    feature = user.value() + movie.value()
+
+    def reader():
+        yield feature
+
+    infer_dict = copy.copy(feeding)
+    del infer_dict['score']
+
+    prediction = paddle.infer(
+        output=inference,
+        parameters=parameters,
+        reader=paddle.batch(
+            reader, batch_size=32),
+        feeding=infer_dict)
+    print(prediction + 5) / 2
+
+
+if __name__ == '__main__':
+    main()

demo/recommendation/trainer_config.py

@@ -86,10 +86,7 @@ movie_feature = construct_feature("movie")
 user_feature = construct_feature("user")
 similarity = cos_sim(a=movie_feature, b=user_feature)
 if not is_predict:
-    outputs(
-        regression_cost(
-            input=similarity, label=data_layer(
-                'rating', size=1)))
+    outputs(mse_cost(input=similarity, label=data_layer('rating', size=1)))
 
     define_py_data_sources2(
         'data/train.list',

demo/semantic_role_labeling/api_train_v2.py

+import sys
+import math
+import numpy as np
+import paddle.v2 as paddle
+import paddle.v2.dataset.conll05 as conll05
+
+
+def db_lstm():
+    word_dict, verb_dict, label_dict = conll05.get_dict()
+    word_dict_len = len(word_dict)
+    label_dict_len = len(label_dict)
+    pred_len = len(verb_dict)
+
+    mark_dict_len = 2
+    word_dim = 32
+    mark_dim = 5
+    hidden_dim = 512
+    depth = 8
+
+    #8 features
+    def d_type(size):
+        return paddle.data_type.integer_value_sequence(size)
+
+    word = paddle.layer.data(name='word_data', type=d_type(word_dict_len))
+    predicate = paddle.layer.data(name='verb_data', type=d_type(pred_len))
+
+    ctx_n2 = paddle.layer.data(name='ctx_n2_data', type=d_type(word_dict_len))
+    ctx_n1 = paddle.layer.data(name='ctx_n1_data', type=d_type(word_dict_len))
+    ctx_0 = paddle.layer.data(name='ctx_0_data', type=d_type(word_dict_len))
+    ctx_p1 = paddle.layer.data(name='ctx_p1_data', type=d_type(word_dict_len))
+    ctx_p2 = paddle.layer.data(name='ctx_p2_data', type=d_type(word_dict_len))
+    mark = paddle.layer.data(name='mark_data', type=d_type(mark_dict_len))
+
+    target = paddle.layer.data(name='target', type=d_type(label_dict_len))
+
+    default_std = 1 / math.sqrt(hidden_dim) / 3.0
+
+    emb_para = paddle.attr.Param(name='emb', initial_std=0., learning_rate=0.)
+    std_0 = paddle.attr.Param(initial_std=0.)
+    std_default = paddle.attr.Param(initial_std=default_std)
+
+    predicate_embedding = paddle.layer.embedding(
+        size=word_dim,
+        input=predicate,
+        param_attr=paddle.attr.Param(
+            name='vemb', initial_std=default_std))
+    mark_embedding = paddle.layer.embedding(
+        size=mark_dim, input=mark, param_attr=std_0)
+
+    word_input = [word, ctx_n2, ctx_n1, ctx_0, ctx_p1, ctx_p2]
+    emb_layers = [
+        paddle.layer.embedding(
+            size=word_dim, input=x, param_attr=emb_para) for x in word_input
+    ]
+    emb_layers.append(predicate_embedding)
+    emb_layers.append(mark_embedding)
+
+    hidden_0 = paddle.layer.mixed(
+        size=hidden_dim,
+        bias_attr=std_default,
+        input=[
+            paddle.layer.full_matrix_projection(
+                input=emb, param_attr=std_default) for emb in emb_layers
+        ])
+
+    mix_hidden_lr = 1e-3
+    lstm_para_attr = paddle.attr.Param(initial_std=0.0, learning_rate=1.0)
+    hidden_para_attr = paddle.attr.Param(
+        initial_std=default_std, learning_rate=mix_hidden_lr)
+
+    lstm_0 = paddle.layer.lstmemory(
+        input=hidden_0,
+        act=paddle.activation.Relu(),
+        gate_act=paddle.activation.Sigmoid(),
+        state_act=paddle.activation.Sigmoid(),
+        bias_attr=std_0,
+        param_attr=lstm_para_attr)
+
+    #stack L-LSTM and R-LSTM with direct edges
+    input_tmp = [hidden_0, lstm_0]
+
+    for i in range(1, depth):
+        mix_hidden = paddle.layer.mixed(
+            size=hidden_dim,
+            bias_attr=std_default,
+            input=[
+                paddle.layer.full_matrix_projection(
+                    input=input_tmp[0], param_attr=hidden_para_attr),
+                paddle.layer.full_matrix_projection(
+                    input=input_tmp[1], param_attr=lstm_para_attr)
+            ])
+
+        lstm = paddle.layer.lstmemory(
+            input=mix_hidden,
+            act=paddle.activation.Relu(),
+            gate_act=paddle.activation.Sigmoid(),
+            state_act=paddle.activation.Sigmoid(),
+            reverse=((i % 2) == 1),
+            bias_attr=std_0,
+            param_attr=lstm_para_attr)
+
+        input_tmp = [mix_hidden, lstm]
+
+    feature_out = paddle.layer.mixed(
+        size=label_dict_len,
+        bias_attr=std_default,
+        input=[
+            paddle.layer.full_matrix_projection(
+                input=input_tmp[0], param_attr=hidden_para_attr),
+            paddle.layer.full_matrix_projection(
+                input=input_tmp[1], param_attr=lstm_para_attr)
+        ], )
+
+    crf_cost = paddle.layer.crf(size=label_dict_len,
+                                input=feature_out,
+                                label=target,
+                                param_attr=paddle.attr.Param(
+                                    name='crfw',
+                                    initial_std=default_std,
+                                    learning_rate=mix_hidden_lr))
+
+    crf_dec = paddle.layer.crf_decoding(
+        name='crf_dec_l',
+        size=label_dict_len,
+        input=feature_out,
+        label=target,
+        param_attr=paddle.attr.Param(name='crfw'))
+
+    return crf_cost, crf_dec
+
+
+def load_parameter(file_name, h, w):
+    with open(file_name, 'rb') as f:
+        f.read(16)  # skip header.
+        return np.fromfile(f, dtype=np.float32).reshape(h, w)
+
+
+def main():
+    paddle.init(use_gpu=False, trainer_count=1)
+
+    # define network topology
+    crf_cost, crf_dec = db_lstm()
+
+    # create parameters
+    parameters = paddle.parameters.create([crf_cost, crf_dec])
+
+    # create optimizer
+    optimizer = paddle.optimizer.Momentum(
+        momentum=0,
+        learning_rate=2e-2,
+        regularization=paddle.optimizer.L2Regularization(rate=8e-4),
+        model_average=paddle.optimizer.ModelAverage(
+            average_window=0.5, max_average_window=10000), )
+
+    def event_handler(event):
+        if isinstance(event, paddle.event.EndIteration):
+            if event.batch_id % 100 == 0:
+                print "Pass %d, Batch %d, Cost %f, %s" % (
+                    event.pass_id, event.batch_id, event.cost, event.metrics)
+
+    trainer = paddle.trainer.SGD(cost=crf_cost,
+                                 parameters=parameters,
+                                 update_equation=optimizer)
+    parameters.set('emb', load_parameter(conll05.get_embedding(), 44068, 32))
+
+    trn_reader = paddle.batch(
+        paddle.reader.shuffle(
+            conll05.test(), buf_size=8192), batch_size=10)
+
+    feeding = {
+        'word_data': 0,
+        'ctx_n2_data': 1,
+        'ctx_n1_data': 2,
+        'ctx_0_data': 3,
+        'ctx_p1_data': 4,
+        'ctx_p2_data': 5,
+        'verb_data': 6,
+        'mark_data': 7,
+        'target': 8
+    }
+
+    trainer.train(
+        reader=trn_reader,
+        event_handler=event_handler,
+        num_passes=10000,
+        feeding=feeding)
+
+
+if __name__ == '__main__':
+    main()

demo/sentiment/dataprovider.py

@@ -32,4 +32,6 @@ def process(settings, file_name):
             word_slot = [
                 settings.word_dict[w] for w in words if w in settings.word_dict
             ]
+            if not word_slot:
+                continue
             yield word_slot, label

demo/sentiment/predict.py

@@ -138,7 +138,11 @@ def main():
 
     batch = []
     for line in sys.stdin:
-        batch.append([predict.get_index(line)])
+        words = predict.get_index(line)
+        if words:
+            batch.append([words])
+        else:
+            print('All the words in [%s] are not in the dictionary.' % line)
         if len(batch) == batch_size:
             predict.batch_predict(batch)
             batch = []

demo/sentiment/train_v2.py

+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
+#
+# 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.
+
+import sys
+import paddle.v2 as paddle
+
+
+def convolution_net(input_dim, class_dim=2, emb_dim=128, hid_dim=128):
+    data = paddle.layer.data("word",
+                             paddle.data_type.integer_value_sequence(input_dim))
+    emb = paddle.layer.embedding(input=data, size=emb_dim)
+    conv_3 = paddle.networks.sequence_conv_pool(
+        input=emb, context_len=3, hidden_size=hid_dim)
+    conv_4 = paddle.networks.sequence_conv_pool(
+        input=emb, context_len=4, hidden_size=hid_dim)
+    output = paddle.layer.fc(input=[conv_3, conv_4],
+                             size=class_dim,
+                             act=paddle.activation.Softmax())
+    lbl = paddle.layer.data("label", paddle.data_type.integer_value(2))
+    cost = paddle.layer.classification_cost(input=output, label=lbl)
+    return cost
+
+
+def stacked_lstm_net(input_dim,
+                     class_dim=2,
+                     emb_dim=128,
+                     hid_dim=512,
+                     stacked_num=3):
+    """
+    A Wrapper for sentiment classification task.
+    This network uses bi-directional recurrent network,
+    consisting three LSTM layers. This configure is referred to
+    the paper as following url, but use fewer layrs.
+        http://www.aclweb.org/anthology/P15-1109
+
+    input_dim: here is word dictionary dimension.
+    class_dim: number of categories.
+    emb_dim: dimension of word embedding.
+    hid_dim: dimension of hidden layer.
+    stacked_num: number of stacked lstm-hidden layer.
+    """
+    assert stacked_num % 2 == 1
+
+    layer_attr = paddle.attr.Extra(drop_rate=0.5)
+    fc_para_attr = paddle.attr.Param(learning_rate=1e-3)
+    lstm_para_attr = paddle.attr.Param(initial_std=0., learning_rate=1.)
+    para_attr = [fc_para_attr, lstm_para_attr]
+    bias_attr = paddle.attr.Param(initial_std=0., l2_rate=0.)
+    relu = paddle.activation.Relu()
+    linear = paddle.activation.Linear()
+
+    data = paddle.layer.data("word",
+                             paddle.data_type.integer_value_sequence(input_dim))
+    emb = paddle.layer.embedding(input=data, size=emb_dim)
+
+    fc1 = paddle.layer.fc(input=emb,
+                          size=hid_dim,
+                          act=linear,
+                          bias_attr=bias_attr)
+    lstm1 = paddle.layer.lstmemory(
+        input=fc1, act=relu, bias_attr=bias_attr, layer_attr=layer_attr)
+
+    inputs = [fc1, lstm1]
+    for i in range(2, stacked_num + 1):
+        fc = paddle.layer.fc(input=inputs,
+                             size=hid_dim,
+                             act=linear,
+                             param_attr=para_attr,
+                             bias_attr=bias_attr)
+        lstm = paddle.layer.lstmemory(
+            input=fc,
+            reverse=(i % 2) == 0,
+            act=relu,
+            bias_attr=bias_attr,
+            layer_attr=layer_attr)
+        inputs = [fc, lstm]
+
+    fc_last = paddle.layer.pooling(
+        input=inputs[0], pooling_type=paddle.pooling.Max())
+    lstm_last = paddle.layer.pooling(
+        input=inputs[1], pooling_type=paddle.pooling.Max())
+    output = paddle.layer.fc(input=[fc_last, lstm_last],
+                             size=class_dim,
+                             act=paddle.activation.Softmax(),
+                             bias_attr=bias_attr,
+                             param_attr=para_attr)
+
+    lbl = paddle.layer.data("label", paddle.data_type.integer_value(2))
+    cost = paddle.layer.classification_cost(input=output, label=lbl)
+    return cost
+
+
+if __name__ == '__main__':
+    # init
+    paddle.init(use_gpu=False)
+
+    #data
+    print 'load dictionary...'
+    word_dict = paddle.dataset.imdb.word_dict()
+    dict_dim = len(word_dict)
+    class_dim = 2
+    train_reader = paddle.batch(
+        paddle.reader.shuffle(
+            lambda: paddle.dataset.imdb.train(word_dict), buf_size=1000),
+        batch_size=100)
+    test_reader = paddle.batch(
+        lambda: paddle.dataset.imdb.test(word_dict), batch_size=100)
+
+    feeding = {'word': 0, 'label': 1}
+
+    # network config
+    # Please choose the way to build the network
+    # by uncommenting the corresponding line.
+    cost = convolution_net(dict_dim, class_dim=class_dim)
+    # cost = stacked_lstm_net(dict_dim, class_dim=class_dim, stacked_num=3)
+
+    # create parameters
+    parameters = paddle.parameters.create(cost)
+
+    # create optimizer
+    adam_optimizer = paddle.optimizer.Adam(
+        learning_rate=2e-3,
+        regularization=paddle.optimizer.L2Regularization(rate=8e-4),
+        model_average=paddle.optimizer.ModelAverage(average_window=0.5))
+
+    # End batch and end pass event handler
+    def event_handler(event):
+        if isinstance(event, paddle.event.EndIteration):
+            if event.batch_id % 100 == 0:
+                print "\nPass %d, Batch %d, Cost %f, %s" % (
+                    event.pass_id, event.batch_id, event.cost, event.metrics)
+            else:
+                sys.stdout.write('.')
+                sys.stdout.flush()
+        if isinstance(event, paddle.event.EndPass):
+            result = trainer.test(reader=test_reader, feeding=feeding)
+            print "\nTest with Pass %d, %s" % (event.pass_id, result.metrics)
+
+    # create trainer
+    trainer = paddle.trainer.SGD(cost=cost,
+                                 parameters=parameters,
+                                 update_equation=adam_optimizer)
+
+    trainer.train(
+        reader=train_reader,
+        event_handler=event_handler,
+        feeding=feeding,
+        num_passes=2)

demo/seqToseq/api_train_v2.py

+import sys
+import paddle.v2 as paddle
+
+
+def seqToseq_net(source_dict_dim, target_dict_dim):
+    ### Network Architecture
+    word_vector_dim = 512  # dimension of word vector
+    decoder_size = 512  # dimension of hidden unit in GRU Decoder network
+    encoder_size = 512  # dimension of hidden unit in GRU Encoder network
+
+    #### Encoder
+    src_word_id = paddle.layer.data(
+        name='source_language_word',
+        type=paddle.data_type.integer_value_sequence(source_dict_dim))
+    src_embedding = paddle.layer.embedding(
+        input=src_word_id,
+        size=word_vector_dim,
+        param_attr=paddle.attr.ParamAttr(name='_source_language_embedding'))
+    src_forward = paddle.networks.simple_gru(
+        input=src_embedding, size=encoder_size)
+    src_backward = paddle.networks.simple_gru(
+        input=src_embedding, size=encoder_size, reverse=True)
+    encoded_vector = paddle.layer.concat(input=[src_forward, src_backward])
+
+    #### Decoder
+    with paddle.layer.mixed(size=decoder_size) as encoded_proj:
+        encoded_proj += paddle.layer.full_matrix_projection(
+            input=encoded_vector)
+
+    backward_first = paddle.layer.first_seq(input=src_backward)
+
+    with paddle.layer.mixed(
+            size=decoder_size, act=paddle.activation.Tanh()) as decoder_boot:
+        decoder_boot += paddle.layer.full_matrix_projection(
+            input=backward_first)
+
+    def gru_decoder_with_attention(enc_vec, enc_proj, current_word):
+
+        decoder_mem = paddle.layer.memory(
+            name='gru_decoder', size=decoder_size, boot_layer=decoder_boot)
+
+        context = paddle.networks.simple_attention(
+            encoded_sequence=enc_vec,
+            encoded_proj=enc_proj,
+            decoder_state=decoder_mem)
+
+        with paddle.layer.mixed(size=decoder_size * 3) as decoder_inputs:
+            decoder_inputs += paddle.layer.full_matrix_projection(input=context)
+            decoder_inputs += paddle.layer.full_matrix_projection(
+                input=current_word)
+
+        gru_step = paddle.layer.gru_step(
+            name='gru_decoder',
+            input=decoder_inputs,
+            output_mem=decoder_mem,
+            size=decoder_size)
+
+        with paddle.layer.mixed(
+                size=target_dict_dim,
+                bias_attr=True,
+                act=paddle.activation.Softmax()) as out:
+            out += paddle.layer.full_matrix_projection(input=gru_step)
+        return out
+
+    decoder_group_name = "decoder_group"
+    group_input1 = paddle.layer.StaticInputV2(input=encoded_vector, is_seq=True)
+    group_input2 = paddle.layer.StaticInputV2(input=encoded_proj, is_seq=True)
+    group_inputs = [group_input1, group_input2]
+
+    trg_embedding = paddle.layer.embedding(
+        input=paddle.layer.data(
+            name='target_language_word',
+            type=paddle.data_type.integer_value_sequence(target_dict_dim)),
+        size=word_vector_dim,
+        param_attr=paddle.attr.ParamAttr(name='_target_language_embedding'))
+    group_inputs.append(trg_embedding)
+
+    # For decoder equipped with attention mechanism, in training,
+    # target embeding (the groudtruth) is the data input,
+    # while encoded source sequence is accessed to as an unbounded memory.
+    # Here, the StaticInput defines a read-only memory
+    # for the recurrent_group.
+    decoder = paddle.layer.recurrent_group(
+        name=decoder_group_name,
+        step=gru_decoder_with_attention,
+        input=group_inputs)
+
+    lbl = paddle.layer.data(
+        name='target_language_next_word',
+        type=paddle.data_type.integer_value_sequence(target_dict_dim))
+    cost = paddle.layer.classification_cost(input=decoder, label=lbl)
+
+    return cost
+
+
+def main():
+    paddle.init(use_gpu=False, trainer_count=1)
+
+    # source and target dict dim.
+    dict_size = 30000
+    source_dict_dim = target_dict_dim = dict_size
+
+    # define network topology
+    cost = seqToseq_net(source_dict_dim, target_dict_dim)
+    parameters = paddle.parameters.create(cost)
+
+    # define optimize method and trainer
+    optimizer = paddle.optimizer.Adam(
+        learning_rate=5e-5,
+        regularization=paddle.optimizer.L2Regularization(rate=1e-3))
+    trainer = paddle.trainer.SGD(cost=cost,
+                                 parameters=parameters,
+                                 update_equation=optimizer)
+
+    # define data reader
+    feeding = {
+        'source_language_word': 0,
+        'target_language_word': 1,
+        'target_language_next_word': 2
+    }
+
+    wmt14_reader = paddle.batch(
+        paddle.reader.shuffle(
+            paddle.dataset.wmt14.train(dict_size=dict_size), buf_size=8192),
+        batch_size=5)
+
+    # define event_handler callback
+    def event_handler(event):
+        if isinstance(event, paddle.event.EndIteration):
+            if event.batch_id % 10 == 0:
+                print "\nPass %d, Batch %d, Cost %f, %s" % (
+                    event.pass_id, event.batch_id, event.cost, event.metrics)
+            else:
+                sys.stdout.write('.')
+                sys.stdout.flush()
+
+    # start to train
+    trainer.train(
+        reader=wmt14_reader,
+        event_handler=event_handler,
+        num_passes=10000,
+        feeding=feeding)
+
+
+if __name__ == '__main__':
+    main()

demo/word2vec/train_v2.py

+import math
+
+import paddle.v2 as paddle
+
+dictsize = 1953
+embsize = 32
+hiddensize = 256
+N = 5
+
+
+def wordemb(inlayer):
+    wordemb = paddle.layer.table_projection(
+        input=inlayer,
+        size=embsize,
+        param_attr=paddle.attr.Param(
+            name="_proj",
+            initial_std=0.001,
+            learning_rate=1,
+            l2_rate=0, ))
+    return wordemb
+
+
+def main():
+    paddle.init(use_gpu=False, trainer_count=1)
+    word_dict = paddle.dataset.imikolov.build_dict()
+    dict_size = len(word_dict)
+    firstword = paddle.layer.data(
+        name="firstw", type=paddle.data_type.integer_value(dict_size))
+    secondword = paddle.layer.data(
+        name="secondw", type=paddle.data_type.integer_value(dict_size))
+    thirdword = paddle.layer.data(
+        name="thirdw", type=paddle.data_type.integer_value(dict_size))
+    fourthword = paddle.layer.data(
+        name="fourthw", type=paddle.data_type.integer_value(dict_size))
+    nextword = paddle.layer.data(
+        name="fifthw", type=paddle.data_type.integer_value(dict_size))
+
+    Efirst = wordemb(firstword)
+    Esecond = wordemb(secondword)
+    Ethird = wordemb(thirdword)
+    Efourth = wordemb(fourthword)
+
+    contextemb = paddle.layer.concat(input=[Efirst, Esecond, Ethird, Efourth])
+    hidden1 = paddle.layer.fc(input=contextemb,
+                              size=hiddensize,
+                              act=paddle.activation.Sigmoid(),
+                              layer_attr=paddle.attr.Extra(drop_rate=0.5),
+                              bias_attr=paddle.attr.Param(learning_rate=2),
+                              param_attr=paddle.attr.Param(
+                                  initial_std=1. / math.sqrt(embsize * 8),
+                                  learning_rate=1))
+    predictword = paddle.layer.fc(input=hidden1,
+                                  size=dict_size,
+                                  bias_attr=paddle.attr.Param(learning_rate=2),
+                                  act=paddle.activation.Softmax())
+
+    def event_handler(event):
+        if isinstance(event, paddle.event.EndIteration):
+            if event.batch_id % 100 == 0:
+                result = trainer.test(
+                    paddle.batch(
+                        paddle.dataset.imikolov.test(word_dict, N), 32))
+                print "Pass %d, Batch %d, Cost %f, %s, Testing metrics %s" % (
+                    event.pass_id, event.batch_id, event.cost, event.metrics,
+                    result.metrics)
+
+    cost = paddle.layer.classification_cost(input=predictword, label=nextword)
+    parameters = paddle.parameters.create(cost)
+    adam_optimizer = paddle.optimizer.Adam(
+        learning_rate=3e-3,
+        regularization=paddle.optimizer.L2Regularization(8e-4))
+    trainer = paddle.trainer.SGD(cost, parameters, adam_optimizer)
+    trainer.train(
+        paddle.batch(paddle.dataset.imikolov.train(word_dict, N), 32),
+        num_passes=30,
+        event_handler=event_handler)
+
+
+if __name__ == '__main__':
+    main()

doc/api/index_cn.rst

-API中文手册
-============
-
-DataProvider API
-----------------
-
-..  toctree::
-    :maxdepth: 1
-
-    data_provider/dataprovider_cn.rst
-    data_provider/pydataprovider2_cn.rst
-
-..  _api_trainer_config:
-
-Model Config API
-----------------
-
-..  toctree::
-    :maxdepth: 1
-
-    trainer_config_helpers/optimizers.rst
-    trainer_config_helpers/data_sources.rst
-    trainer_config_helpers/layers.rst
-    trainer_config_helpers/activations.rst 
-    trainer_config_helpers/poolings.rst
-    trainer_config_helpers/networks.rst
-    trainer_config_helpers/evaluators.rst
-    trainer_config_helpers/attrs.rst
-
-
-Applications API
-----------------
+API
+===
 
 ..  toctree::
     :maxdepth: 1
 
-    predict/swig_py_paddle_cn.rst
+    模型配置 <v2/model_configs.rst>
+    数据访问 <v2/data.rst>
+    训练与应用 <v2/run_logic.rst>

doc/api/index_en.rst

 API
 ===
 
-DataProvider API
-----------------
-
-..  toctree::
-    :maxdepth: 1
-
-    data_provider/dataprovider_en.rst
-    data_provider/pydataprovider2_en.rst
-
-..  _api_trainer_config:
-
-Model Config API
-----------------
-
-..  toctree::
-    :maxdepth: 1
-
-    trainer_config_helpers/optimizers.rst
-    trainer_config_helpers/data_sources.rst
-    trainer_config_helpers/layers.rst
-    trainer_config_helpers/activations.rst 
-    trainer_config_helpers/poolings.rst
-    trainer_config_helpers/networks.rst
-    trainer_config_helpers/evaluators.rst
-    trainer_config_helpers/attrs.rst
-
-
-Applications API
-----------------
-
 ..  toctree::
     :maxdepth: 1
 
-    predict/swig_py_paddle_en.rst
+    v2/model_configs.rst
+    v2/data.rst
+    v2/run_logic.rst

doc/api/v1/index_cn.rst

+API中文手册
+============
+
+DataProvider API
+----------------
+
+..  toctree::
+    :maxdepth: 1
+
+    data_provider/dataprovider_cn.rst
+    data_provider/pydataprovider2_cn.rst
+
+..  _api_trainer_config:
+
+Model Config API
+----------------
+
+..  toctree::
+    :maxdepth: 1
+
+    trainer_config_helpers/optimizers.rst
+    trainer_config_helpers/data_sources.rst
+    trainer_config_helpers/layers.rst
+    trainer_config_helpers/activations.rst 
+    trainer_config_helpers/poolings.rst
+    trainer_config_helpers/networks.rst
+    trainer_config_helpers/evaluators.rst
+    trainer_config_helpers/attrs.rst
+
+
+Applications API
+----------------
+
+..  toctree::
+    :maxdepth: 1
+
+    predict/swig_py_paddle_cn.rst

doc/api/v1/index_en.rst

+API
+===
+
+DataProvider API
+----------------
+
+..  toctree::
+    :maxdepth: 1
+
+    data_provider/dataprovider_en.rst
+    data_provider/pydataprovider2_en.rst
+
+..  _api_trainer_config:
+
+Model Config API
+----------------
+
+..  toctree::
+    :maxdepth: 1
+
+    trainer_config_helpers/optimizers.rst
+    trainer_config_helpers/data_sources.rst
+    trainer_config_helpers/layers.rst
+    trainer_config_helpers/activations.rst 
+    trainer_config_helpers/poolings.rst
+    trainer_config_helpers/networks.rst
+    trainer_config_helpers/evaluators.rst
+    trainer_config_helpers/attrs.rst
+
+
+Applications API
+----------------
+
+..  toctree::
+    :maxdepth: 1
+
+    predict/swig_py_paddle_en.rst

doc/api/trainer_config_helpers/layers.rst → doc/api/v1/trainer_config_helpers/layers.rst

@@ -139,24 +139,12 @@ lstmemory
     :members: lstmemory
     :noindex:
 
-lstm_step_layer
----------------
-..  automodule:: paddle.trainer_config_helpers.layers
-    :members: lstm_step_layer
-    :noindex:
-
 grumemory
 ---------
 ..  automodule:: paddle.trainer_config_helpers.layers
     :members: grumemory
     :noindex:
 
-gru_step_layer
----------------
-..  automodule:: paddle.trainer_config_helpers.layers
-    :members: gru_step_layer
-    :noindex:
-
 Recurrent Layer Group
 =====================
 
@@ -172,6 +160,18 @@ recurrent_group
     :members: recurrent_group
     :noindex:
     
+lstm_step_layer
+---------------
+..  automodule:: paddle.trainer_config_helpers.layers
+    :members: lstm_step_layer
+    :noindex:
+
+gru_step_layer
+---------------
+..  automodule:: paddle.trainer_config_helpers.layers
+    :members: gru_step_layer
+    :noindex:
+
 beam_search
 ------------
 ..  automodule:: paddle.trainer_config_helpers.layers
@@ -279,6 +279,12 @@ concat_layer
     :members: concat_layer
     :noindex:
 
+seq_concat_layer
+----------------
+..  automodule:: paddle.trainer_config_helpers.layers
+    :members: seq_concat_layer
+    :noindex:
+
 Reshaping Layers
 ================
 
@@ -302,6 +308,18 @@ repeat_layer
     :members: repeat_layer
     :noindex:
 
+rotate_layer
+------------
+..  automodule:: paddle.trainer_config_helpers.layers
+    :members: rotate_layer
+    :noindex:
+
+seq_reshape_layer
+-----------------
+..  automodule:: paddle.trainer_config_helpers.layers
+    :members: seq_reshape_layer
+    :noindex:
+
 Math Layers
 ===========
 
@@ -414,6 +432,12 @@ multi_binary_label_cross_entropy
     :members: multi_binary_label_cross_entropy
     :noindex:
 
+mse_cost
+---------
+..  automodule:: paddle.trainer_config_helpers.layers
+    :members: mse_cost
+    :noindex:
+
 huber_cost
 ----------
 ..  automodule:: paddle.trainer_config_helpers.layers
@@ -432,6 +456,12 @@ rank_cost
     :members: rank_cost
     :noindex:
 
+sum_cost
+---------
+..  automodule:: paddle.trainer_config_helpers.layers
+    :members: sum_cost
+    :noindex:
+
 crf_layer
 -----------------
 ..  automodule:: paddle.trainer_config_helpers.layers
@@ -450,6 +480,12 @@ ctc_layer
     :members: ctc_layer
     :noindex:
 
+warp_ctc_layer
+--------------
+..  automodule:: paddle.trainer_config_helpers.layers
+    :members: warp_ctc_layer
+    :noindex:
+
 nce_layer
 -----------
 ..  automodule:: paddle.trainer_config_helpers.layers
@@ -462,12 +498,6 @@ hsigmoid
     :members: hsigmoid
     :noindex:
 
-sum_cost
----------
-..  automodule:: paddle.trainer_config_helpers.layers
-    :members: sum_cost
-    :noindex:
-
 Check Layer 
 ============
 

doc/api/v2/config/activation.rst

+===========
+Activation
+===========
+
+Abs
+===
+
+..  automodule:: paddle.v2.activation
+    :members: Abs
+    :noindex:
+    
+Exp
+===
+
+..  automodule:: paddle.v2.activation
+    :members: Exp
+    :noindex:
+    
+Identity
+========
+
+..  automodule:: paddle.v2.activation
+    :members: Identity
+    :noindex:
+    
+Linear
+======
+
+..  automodule:: paddle.v2.activation
+    :members: Linear
+    :noindex:
+
+Log
+===
+
+..  automodule:: paddle.v2.activation
+    :members: Log
+    :noindex:
+    
+Square
+======
+
+..  automodule:: paddle.v2.activation
+    :members: Square
+    :noindex:
+    
+Sigmoid
+=======
+
+..  automodule:: paddle.v2.activation
+    :members: Sigmoid
+    :noindex:
+    
+Softmax
+=======
+
+..  automodule:: paddle.v2.activation
+    :members: Softmax
+    :noindex:
+    
+SequenceSoftmax
+===============
+
+..  automodule:: paddle.v2.activation
+    :members: SequenceSoftmax
+    :noindex:
+    
+Relu
+====
+
+..  automodule:: paddle.v2.activation
+    :members: Relu
+    :noindex:
+    
+BRelu
+=====
+
+..  automodule:: paddle.v2.activation
+    :members: BRelu
+    :noindex:
+    
+SoftRelu
+========
+
+..  automodule:: paddle.v2.activation
+    :members: SoftRelu
+    :noindex:
+    
+Tanh
+====
+
+..  automodule:: paddle.v2.activation
+    :members: Tanh
+    :noindex:
+    
+STanh
+=====
+
+..  automodule:: paddle.v2.activation
+    :members: STanh
+    :noindex:

doc/api/v2/config/attr.rst

+Parameter Attribute
+===================
+
+..  automodule:: paddle.v2.attr
+    :members:
+    :noindex:

doc/api/v2/config/layer.rst

+..  _api_v2.layer:
+
+======
+Layers
+======
+
+Data layer
+===========
+
+..  _api_v2.layer_data:
+
+data
+----
+..  automodule:: paddle.v2.layer
+    :members: data
+    :noindex:
+
+Fully Connected Layers
+======================
+
+..  _api_v2.layer_fc:
+
+fc
+--
+..  automodule:: paddle.v2.layer
+    :members: fc
+    :noindex:
+
+selective_fc
+------------
+..  automodule:: paddle.v2.layer
+    :members: selective_fc
+    :noindex:
+
+Conv Layers
+===========
+
+conv_operator
+-------------
+..  automodule:: paddle.v2.layer
+    :members: conv_operator
+    :noindex:
+
+conv_projection
+---------------
+..  automodule:: paddle.v2.layer
+    :members: conv_projection
+    :noindex:
+
+conv_shift
+----------
+..  automodule:: paddle.v2.layer
+    :members: conv_shift
+    :noindex:
+
+img_conv
+--------
+..  automodule:: paddle.v2.layer
+    :members: img_conv
+    :noindex:
+
+..  _api_v2.layer_context_projection:
+
+context_projection 
+------------------
+..  automodule:: paddle.v2.layer
+    :members: context_projection
+    :noindex:
+
+Image Pooling Layer
+===================
+
+img_pool
+--------
+..  automodule:: paddle.v2.layer
+    :members: img_pool
+    :noindex:   
+
+spp
+---
+..  automodule:: paddle.v2.layer
+    :members: spp
+    :noindex:
+
+maxout
+------
+..  automodule:: paddle.v2.layer
+    :members: maxout
+    :noindex:
+
+Norm Layer
+==========
+
+img_cmrnorm
+-----------
+..  automodule:: paddle.v2.layer
+    :members: img_cmrnorm
+    :noindex:
+
+batch_norm
+----------
+..  automodule:: paddle.v2.layer
+    :members: batch_norm
+    :noindex:
+
+sum_to_one_norm
+---------------
+..  automodule:: paddle.v2.layer
+    :members: sum_to_one_norm
+    :noindex:
+    
+Recurrent Layers
+================
+
+recurrent
+---------
+..  automodule:: paddle.v2.layer
+    :members: recurrent
+    :noindex:
+
+lstmemory
+---------
+..  automodule:: paddle.v2.layer
+    :members: lstmemory
+    :noindex:
+
+grumemory
+---------
+..  automodule:: paddle.v2.layer
+    :members: grumemory
+    :noindex:
+
+Recurrent Layer Group
+=====================
+
+memory
+------
+..  automodule:: paddle.v2.layer
+    :members: memory
+    :noindex:
+
+recurrent_group
+---------------
+..  automodule:: paddle.v2.layer
+    :members: recurrent_group
+    :noindex:
+    
+lstm_step
+---------
+..  automodule:: paddle.v2.layer
+    :members: lstm_step
+    :noindex:
+
+gru_step
+--------
+..  automodule:: paddle.v2.layer
+    :members: gru_step
+    :noindex:
+
+beam_search
+------------
+..  automodule:: paddle.v2.layer
+    :members: beam_search
+    :noindex:
+    
+get_output
+----------
+..  automodule:: paddle.v2.layer
+    :members: get_output
+    :noindex:
+    
+Mixed Layer
+===========
+
+..  _api_v2.layer_mixed:
+
+mixed
+-----
+..  automodule:: paddle.v2.layer
+    :members: mixed
+    :noindex:
+
+..  _api_v2.layer_embedding:
+
+embedding
+---------
+..  automodule:: paddle.v2.layer
+    :members: embedding
+    :noindex:
+
+scaling_projection
+------------------
+..  automodule:: paddle.v2.layer
+    :members: scaling_projection
+    :noindex:
+
+dotmul_projection
+-----------------
+..  automodule:: paddle.v2.layer
+    :members: dotmul_projection
+    :noindex:
+
+dotmul_operator
+---------------
+..  automodule:: paddle.v2.layer
+    :members: dotmul_operator
+    :noindex:
+
+full_matrix_projection
+----------------------
+..  automodule:: paddle.v2.layer
+    :members: full_matrix_projection
+    :noindex:
+
+identity_projection
+-------------------
+..  automodule:: paddle.v2.layer
+    :members: identity_projection
+    :noindex:
+
+
+table_projection
+----------------
+..  automodule:: paddle.v2.layer
+    :members: table_projection
+    :noindex:
+
+trans_full_matrix_projection
+----------------------------
+..  automodule:: paddle.v2.layer
+    :members: trans_full_matrix_projection
+    :noindex:
+    
+Aggregate Layers
+================
+
+..  _api_v2.layer_pooling:
+
+pooling
+-------
+..  automodule:: paddle.v2.layer
+    :members: pooling
+    :noindex:
+
+..  _api_v2.layer_last_seq:
+
+last_seq
+--------
+..  automodule:: paddle.v2.layer
+    :members: last_seq
+    :noindex:
+
+..  _api_v2.layer_first_seq:
+
+first_seq
+---------
+..  automodule:: paddle.v2.layer
+    :members: first_seq
+    :noindex:
+
+concat
+------
+..  automodule:: paddle.v2.layer
+    :members: concat
+    :noindex:
+
+seq_concat
+----------
+..  automodule:: paddle.v2.layer
+    :members: seq_concat
+    :noindex:
+
+Reshaping Layers
+================
+
+block_expand
+------------
+..  automodule:: paddle.v2.layer
+    :members: block_expand
+    :noindex:
+
+..  _api_v2.layer_expand:
+
+expand
+------
+..  automodule:: paddle.v2.layer
+    :members: expand
+    :noindex:
+
+repeat
+------
+..  automodule:: paddle.v2.layer
+    :members: repeat
+    :noindex:
+
+rotate
+------
+..  automodule:: paddle.v2.layer
+    :members: rotate
+    :noindex:
+
+seq_reshape
+-----------
+..  automodule:: paddle.v2.layer
+    :members: seq_reshape
+    :noindex:
+
+Math Layers
+===========
+
+addto
+-----
+..  automodule:: paddle.v2.layer
+    :members: addto
+    :noindex:
+
+linear_comb
+-----------
+..  automodule:: paddle.v2.layer
+    :members: linear_comb
+    :noindex:
+
+interpolation
+-------------
+..  automodule:: paddle.v2.layer
+    :members: interpolation
+    :noindex:
+
+bilinear_interp
+---------------
+..  automodule:: paddle.v2.layer
+    :members: bilinear_interp
+    :noindex:
+
+power
+-----
+..  automodule:: paddle.v2.layer
+    :members: power
+    :noindex:
+
+scaling
+-------
+..  automodule:: paddle.v2.layer
+    :members: scaling
+    :noindex:
+
+slope_intercept
+---------------
+..  automodule:: paddle.v2.layer
+    :members: slope_intercept
+    :noindex:
+
+tensor
+------
+..  automodule:: paddle.v2.layer
+    :members: tensor
+    :noindex:
+
+..  _api_v2.layer_cos_sim:
+
+cos_sim
+-------
+..  automodule:: paddle.v2.layer
+    :members: cos_sim
+    :noindex:
+
+trans
+-----
+..  automodule:: paddle.v2.layer
+    :members: trans
+    :noindex:
+
+Sampling Layers
+===============
+
+maxid
+-----
+..  automodule:: paddle.v2.layer
+    :members: maxid
+    :noindex:
+
+sampling_id
+-----------
+..  automodule:: paddle.v2.layer
+    :members: sampling_id
+    :noindex:
+
+Slicing and Joining Layers
+==========================
+
+pad
+----
+..  automodule:: paddle.v2.layer
+    :members: pad
+    :noindex:
+
+..  _api_v2.layer_costs:
+
+Cost Layers
+===========
+
+cross_entropy_cost
+------------------
+..  automodule:: paddle.v2.layer
+    :members: cross_entropy_cost
+    :noindex:
+
+cross_entropy_with_selfnorm_cost
+--------------------------------
+..  automodule:: paddle.v2.layer
+    :members: cross_entropy_with_selfnorm_cost
+    :noindex:
+
+multi_binary_label_cross_entropy_cost
+-------------------------------------
+..  automodule:: paddle.v2.layer
+    :members: multi_binary_label_cross_entropy_cost
+    :noindex:
+
+huber_cost
+----------
+..  automodule:: paddle.v2.layer
+    :members: huber_cost
+    :noindex:
+
+lambda_cost
+-----------
+..  automodule:: paddle.v2.layer
+    :members: lambda_cost
+    :noindex:
+
+rank_cost
+---------
+..  automodule:: paddle.v2.layer
+    :members: rank_cost
+    :noindex:
+
+sum_cost
+---------
+..  automodule:: paddle.v2.layer
+    :members: sum_cost
+    :noindex:
+
+crf
+---
+..  automodule:: paddle.v2.layer
+    :members: crf
+    :noindex:
+
+crf_decoding
+------------
+..  automodule:: paddle.v2.layer
+    :members: crf_decoding
+    :noindex:
+
+ctc
+---
+..  automodule:: paddle.v2.layer
+    :members: ctc
+    :noindex:
+
+warp_ctc
+--------
+..  automodule:: paddle.v2.layer
+    :members: warp_ctc
+    :noindex:
+
+nce
+---
+..  automodule:: paddle.v2.layer
+    :members: nce
+    :noindex:
+
+hsigmoid
+---------
+..  automodule:: paddle.v2.layer
+    :members: hsigmoid
+    :noindex:
+
+Check Layer 
+============
+
+eos
+---
+..  automodule:: paddle.v2.layer
+    :members: eos
+    :noindex:

doc/api/v2/config/networks.rst

+========
+Networks
+========
+
+The v2.networks module contains pieces of neural network that combine multiple layers.
+
+NLP
+===
+
+sequence_conv_pool
+------------------
+..  automodule:: paddle.v2.networks
+    :members: sequence_conv_pool
+    :noindex:
+
+..  _api_trainer_config_helpers_network_text_conv_pool:
+
+text_conv_pool
+--------------
+..  automodule:: paddle.v2.networks
+    :members: text_conv_pool
+    :noindex:
+
+Images
+======
+
+img_conv_bn_pool
+----------------
+..  automodule:: paddle.v2.networks
+    :members: img_conv_bn_pool
+    :noindex:
+
+img_conv_group
+--------------
+..  automodule:: paddle.v2.networks
+    :members: img_conv_group
+    :noindex:
+
+..  _api_trainer_config_helpers_network_simple_img_conv_pool:
+
+simple_img_conv_pool
+--------------------
+..  automodule:: paddle.v2.networks
+    :members: simple_img_conv_pool
+    :noindex:
+
+vgg_16_network
+---------------
+..  automodule:: paddle.v2.networks
+    :members: vgg_16_network
+    :noindex:
+
+Recurrent
+=========
+
+LSTM
+----
+
+lstmemory_unit
+``````````````
+..  automodule:: paddle.v2.networks
+    :members: lstmemory_unit
+    :noindex:
+
+lstmemory_group
+```````````````
+..  automodule:: paddle.v2.networks
+    :members: lstmemory_group
+    :noindex:
+
+simple_lstm
+```````````
+..  automodule:: paddle.v2.networks
+    :members: simple_lstm
+    :noindex:
+
+bidirectional_lstm
+``````````````````
+..  automodule:: paddle.v2.networks
+    :members: bidirectional_lstm
+    :noindex:
+
+GRU
+---
+
+gru_unit
+````````
+..  automodule:: paddle.v2.networks
+    :members: gru_unit
+    :noindex:
+
+gru_group
+`````````
+..  automodule:: paddle.v2.networks
+    :members: gru_group
+    :noindex:
+
+simple_gru
+``````````
+..  automodule:: paddle.v2.networks
+    :members: simple_gru
+    :noindex:
+
+simple_attention
+----------------
+..  automodule:: paddle.v2.networks
+    :members: simple_attention
+    :noindex:
+
+Miscs
+=====
+
+dropout_layer
+--------------
+..  automodule:: paddle.v2.networks
+    :members: dropout_layer
+    :noindex:

doc/api/v2/config/optimizer.rst

+..  _api_v2.optimizer:
+
+==========
+Optimizer
+==========
+
+Momentum
+========
+..  automodule:: paddle.v2.optimizer
+    :members: Momentum
+    :noindex:
+
+Adam
+====
+..  automodule:: paddle.v2.optimizer
+    :members: Adam
+    :noindex:
+
+Adamax
+======
+..  automodule:: paddle.v2.optimizer
+    :members: Adamax
+    :noindex:
+
+AdaGrad
+=======
+..  automodule:: paddle.v2.optimizer
+    :members: AdaGrad
+    :noindex:
+
+DecayedAdaGrad
+==============
+..  automodule:: paddle.v2.optimizer
+    :members: DecayedAdaGrad
+    :noindex:
+
+AdaDelta
+========
+..  automodule:: paddle.v2.optimizer
+    :members: AdaDelta
+    :noindex:
+
+RMSProp
+=======
+..  automodule:: paddle.v2.optimizer
+    :members: RMSProp
+    :noindex:

doc/api/v2/config/pooling.rst

+=======
+Pooling
+=======
+
+BasePool
+========
+..  automodule:: paddle.v2.pooling
+    :members: BasePool
+    :noindex:
+
+Avg
+===
+..  automodule:: paddle.v2.pooling
+    :members: Avg
+    :noindex:
+
+Max
+===
+..  automodule:: paddle.v2.pooling
+    :members: Max
+    :noindex:
+
+Sum
+===
+..  automodule:: paddle.v2.pooling
+    :members: Sum
+    :noindex:
+
+SquareRootN
+===========
+..  automodule:: paddle.v2.pooling
+    :members: SquareRootN
+    :noindex:
+
+CudnnAvg
+========
+..  automodule:: paddle.v2.pooling
+    :members: CudnnAvg
+    :noindex:
+
+CudnnMax
+========
+..  automodule:: paddle.v2.pooling
+    :members: CudnnMax
+    :noindex:
+

doc/api/v2/data.rst

+========
+Datasets
+========
+
+
+DataTypes
+=========
+
+..  automodule:: paddle.v2.data_type
+    :members:
+    :noindex:
+
+DataFeeder
+==========
+
+..  automodule:: paddle.v2.data_feeder
+    :members:
+    :noindex:
+
+Reader
+======
+
+..  automodule:: paddle.v2.reader
+    :members:
+    :noindex:
+
+..  automodule:: paddle.v2.reader.creator
+    :members:
+    :noindex:
+
+minibatch
+=========
+
+..  automodule:: paddle.v2.minibatch
+    :members:
+    :noindex:
+
+Dataset
+=======
+
+..  automodule:: paddle.v2.dataset
+    :members:
+    :noindex:
+
+mnist
++++++
+
+..  automodule:: paddle.v2.dataset.mnist
+    :members:
+    :noindex:
+
+
+cifar
++++++
+
+..  automodule:: paddle.v2.dataset.cifar
+    :members:
+    :noindex:
+
+conll05
++++++++
+
+..  automodule:: paddle.v2.dataset.conll05
+    :members:
+    :noindex:
+
+imdb
+++++
+
+..  automodule:: paddle.v2.dataset.imdb
+    :members:
+    :noindex:
+
+imikolov
+++++++++
+
+..  automodule:: paddle.v2.dataset.imikolov
+    :members:
+    :noindex:
+
+movielens
++++++++++
+
+..  automodule:: paddle.v2.dataset.movielens
+    :members:
+    :noindex:
+
+sentiment
++++++++++
+
+..  automodule:: paddle.v2.dataset.sentiment
+    :members:
+    :noindex:
+
+uci_housing
++++++++++++
+
+..  automodule:: paddle.v2.dataset.uci_housing
+    :members:
+    :noindex:
+
+wmt14
++++++
+
+..  automodule:: paddle.v2.dataset.uci_housing
+    :members:
+    :noindex:
+

doc/api/v2/model_configs.rst

+Model Configuration
+===================
+
+..  toctree::
+    :maxdepth: 1
+
+    config/activation.rst
+    config/layer.rst
+    config/optimizer.rst
+    config/pooling.rst
+    config/networks.rst
+    config/attr.rst

doc/api/v2/run_logic.rst

+======================
+Training and Inference
+======================
+
+Parameters
+==========
+
+..  automodule:: paddle.v2.parameters
+    :noindex:
+
+Trainer
+=======
+
+..  automodule:: paddle.v2.trainer
+    :noindex:
+
+Event
+=====
+
+..  automodule:: paddle.v2.event
+    :noindex:
+
+Inference
+=========
+
+..  autofunction:: paddle.v2.infer
+    :noindex:

doc/design/api.md

@@ -2,140 +2,148 @@
 
 ## Ingredients
 
-As the first step of our design, we list important concepts in deep
-learning and try to figure their relationship, as shown below:
+As our design principle is starting from the essence: how could we
+allow users to express and solve their problems at neural networks.
+Some essential concepts that our API have to provide include:
 
-```
-Model = {topology, parameters}
+1. A *topology* is an expression of *layers*.
 
-Evaluator = {Model*, activations}
-- forward
-- test(cost, ...)
+1. A layer could be any kind of computation, including *cost*.
 
-GradientMachine = {Evaluator*, gradients}
-- backward
+1. Some layers have parameters, some don't. Most costs don't have
+   parameters.
 
-Optimizer = {GradientMachine*}
-- train(cost, ...)
-- update
-- checkpoint
-```
+1. In some topologies, layers share parameters.  For
+   example,
+   [the network for training a ranking model](https://github.com/PaddlePaddle/Paddle/issues/1311#issuecomment-279121850).
 
-where the pair of curly braces `{` and `}` indicate *composition*, `*`
-indicates a *reference*, and `-` marks a "class method".
+1. At programming time, users specify topologies and possible sharing
+   of parameters.  PaddlePaddle can figure out and create parameters
+   required (and possibly shared) by one or more topologies.
 
 
-### Model
+## Starting from Examples
 
-We used to think that parameters are part of the topology (or layers).
-But that is not true because multiple layers could share the same
-parameter matrix.  An example is a network that compares two text
-segments in a semantic space:
+As a summarization
+of
+[our disucssion](https://github.com/PaddlePaddle/Paddle/issues/1315),
+let us present two examples here:
 
-```
-          semantic
-text A -> projection ---\
-          layer A        \
-                          cosine
-                          similarity -> output
-                          layer
-          semantic       /
-text B -> projection ---/
-          layer B
-```
 
-In this network, the two semantic projection layers (A and B) share
-the same parameter matrix.
+### Example 1. Sharing Parameters between Layers
 
-For more information about our API that specifies topology and
-parameter sharing, please refer to [TODO: API].
+We use
+the
+[3-branch ranking](https://github.com/PaddlePaddle/Paddle/issues/1311#issuecomment-279121850) model
+in this example.  For your convenience, I copy-a-paste the model's
+topology as follows:
 
+```
+A -> f -\
+Q -> f --> cost
+B -> f -/
+```
 
-### Evaluator
-
-Supposed that we have a trained ranking model, we should be able to
-use it in our search engine.  The search engine's Web server is a
-concurrent program so to serve many HTTP requests simultaneously.  It
-doesn't make sense for each of these threads to have its own copy of the model because that would duplicate topologies and parameters.
-However, each thread should be able to record layer outputs, i.e.,
-activations, computed from an input, derived from the request.  With
-*Evaluator* that saves activations, we can write the over-simplified
-server program as:
+The following program trains the topology including the cost, and then
+use the sub-network in the trained topology in inference:
 
 ```python
-m = paddle.model.load("trained.model")
-
-http.handle("/",
-            lambda req:
-                e = paddle.evaluator.create(m)
-                e.forward(req)
-                e.activation(layer="output")) # returns activations of layer "output"
+def f(in):
+    e = paddle.layer.embedding(in, parameter_name="embedding")
+    o = paddle.layer.softmax(e, parameter_name="semantic")
+    return o
+
+# Create 3 topologies (subnets), they share parameters because all
+# correspoinding layers have the same parameter names.
+fA = f(paddle.layer.data(input_name="A"))
+fB = f(paddle.layer.data(input_name="B"))
+fQ = f(paddle.layer.data(input_name="Q"))
+
+topology = paddle.layer.less_than(
+               paddle.layer.cross_entropy(fA, fQ),
+               paddle.layer.corss_entropy(fB, fQ))
+
+# Derive parameters required in topology and create them in model.
+parameters = paddle.parameters.create(topology)
+
+# Estimate parameters used in topology from data.
+paddle.train(topology, parameters, reader=read_ranking_model_data)
+
+# Inference using fA (or fB or fC, as they share their parameters).
+[testA, testB, testQ] = read_ranking_model_data()
+print "The sematic-vector of testA: ", paddle.infer(fA, parameters, testA)
 ```
 
-### GradientMachine
-
-Similar to the evaluation, the training needs to compute gradients so
-to update model parameters.  Because an [optimizer](#optimizer) might
-run multiple simultaneous threads to update the same model, gradients
-should be separated from the model.  Because gradients are only used
-in training, but not serving, they should be separate from Evaluator.
-Hence the `GradientMachine`.
-
-### Optimizer
-
-None of Model, Evaluator, nor GradientMachine implements the training
-loop, hence Optimizer.  We can define a concurrent optimizer that runs
-multiple simultaneous threads to train a model -- just let each
-thread has its own GradientMachine object.
 
-Most models should be able to be trained using the
-`paddle.optimizer.SGD` by calling its `train` method.  Many
-customizations to the SGD algorithm happens with the update equation,
-e.g., momentum and the Adam SGD algorithm.  We make `train` calls
-`update` to do an update, so that we can derive a `paddle.optimizer.Adam`
-from `paddle.optimizer.SGD` by overrides only the `update` method.
+### Example 2. Sharing Parameters between "Models"
 
+We use [GAN](https://github.com/PaddlePaddle/book/tree/develop/gan) in
+this example.  In the following example program, `d0` and `d1`
+correspond to the two networks in the following figure:
 
-## Programming Interface
-
-A fictive example of PaddlePaddle program looks like the following:
+<img src="https://github.com/wangyang59/book/raw/00036f4b0da5225041a6824587c1a01cf20159b1/gan/image/gan_ig.png" width=400 />
 
 ```python
-import paddle
+def G(in):
+    # over-simplified example as G has only one layers:
+    return paddle.layer.fc(in, parameter_name="G")
+
+def D(in);
+    # again, over-simplified:
+    return paddle.layer.fc(in, parameter_name="D")
+
+# Construct the first topology, which contains both D and G.
+# By learning this topology, we update parameters of G.
+d0 = paddle.layer.should_be_false(D(G(paddle.layer.data())))
+
+# Construct a second topology d1, which contains only D. By
+# training this topology, we update parameters of D.  Note
+# that d1 share parameters with d0.
+d1 = paddle.layer.should_be_true(D(paddle.layer.data()))
+
+# Create parameters from a list of multiple topologies (models) for
+# the chance to share parameters between these topologies.
+parameters = paddle.parameters.create([d0, d1])
+
+# Iterative training of GAN.
+for ...:
+    train(d0, parameters, reader=read_from_rng, immutable_parameters={"D"})
+    train(d1, parameters, reader=read_from_realistic_images)
+
+# Use d1 for inference:
+print "D thinks a batch of images are realistic ", infer(d1, parameters, read_mnist_images)
+```
 
-def read(args):
-    f = open_file(args["filename"])
-    mb = read_a_minibatch(f)
-    end_pass = eof(f)
-    if end_pass:
-       f = open_file(args["filename"]) # rewind for reading again
-    yield mb, end_pass
 
-input = paddle.layer.data(...)
-intermediate = paddle.layers.fc(input)
-output = paddle.layer.softmax(intermediate)
+### Summarization
 
-model = paddle.model.create(output)
 
-paddle.train(model, data_provider=read)
-```
+Above two programs reveal some important design concerns:
 
-This shows some important part of a program:
+1. Users describe a topology as an expression of layers.  Every layer
+   has a *parameter name*.  If the users don't specify it explicitly, it's automatically generated as a unique name.  By
+   specifying the parameter name, users can specify the sharing of
+   parameters between layers and even between topologies.
 
-1. Define how to read (and augment) data by defining a function, in
-   this example, `read`, that `yields` a minibatch and a boolean flag
-   `eof_of_pass`.
+1. `paddle.parameters.create` figures out parameters required by one
+   or more topologies from parameter names of layers.  It creates these
+   parameters and returns a `ParameterSet` object, which is in essence
+   a map from *parameter names* to *parameters*.
 
-1. Define the topology, `input`, `intermediate`, and `output` in this
-   example.
+1. At training and inference time, `paddle.train` and `paddle.infer`
+   requires both a topology and the parameter set that holds the parameters of that topology.  There are some reasons:
 
-1. Create parameters from the topology thus forms the model by calling
-   `paddel.model.create`.
+   1. This prevents users from forgetting to call
+      `paddle.parameters.create`.
+   1. `paddle.train` needs to know which parameter set to update.
+   1. Users could load another (pre-trained) parameter set and use it
+      with a topology in `train.infer`.
 
-1. Train the model by calling `paddle.train`.
+1. By specifying the `immutable_parameters` parameter of
+   `paddle.train`, we can forbid the update of these parameters.
 
 
-### Reader
+## Reader
 
 Not all programming frameworks allow users to define I/O functions.
 An example is Google MapReduce, which can only read from text,
@@ -145,91 +153,67 @@ readers and writers by deriving from base classes `Reader` and
 decide to provide the flexibility to users to define their readers.
 
 
-#### A Synthetic Data Reader
+There are some open questions here:
 
-Sometimes we want to test a topology and/or a training algorithm using
-synthetic data.  We can do this by defining the reader a synthesizer:
+1. **Should a reader return a Python dictionary?**
 
-```python
-def read(args):
-    x = sample_from_uniform(0.0, 1.0)
-    y = sample_from_gauss(2 * x, sigma)
-    yield {x, y}, False # no end-of-file so no end-of-pass
-```
+1. **How to map multiple outputs from a reader to multiple data layers?**
 
-#### A Reader for Online Learning
+1. **How to easily compose some existing readers to read more data and
+   feed a topology with more data layers?**
 
-Readers can also read an infinite data stream, e.g., a log stream from
-a search engine and collected by Kafka:
 
-```python
-def read(args):
-    log_stream = kafka.open_channel(args["kafka channel name"])
-    yeild log_stream.read(), False # no end-of-pass in online learning
-```
+## Training
 
-### Topology
-
-By default, layers don't have names.  But if we want to refer to a
-layer later some time, for example, when we do serving using the model
-and wants activations/outputs of a layer, we should give it a name.
+The recommended way to training a model is to call `paddle.train`,
+which simply calls `paddle.trainer.Default`, a global variable of
+type `paddle.trainer.SGD`.  Equivalently, we can do
 
 ```python
-input = paddle.layer.data(...)
-intermediate = paddle.layer.fc(input, name="inter", ...)
-output = paddle.layer.softmax(intermediate, name="output", ...)
-
-m = paddle.model.create(output)
-e = paddle.evaluator.create(model)
-e.forward(read_an_input()) # compute activations of all layers.
-print e.activations(layer="inter")  # retrieve the activations of layer "inter"
-print e.activations(layer="output") # retrieve the activations of layer "output"
+opt = paddle.trainer.SGD(..., paddle.updater.Adam(...))
+opt.train(topology, parameters, reader=read, ...)
 ```
 
-#### Sharing Parameters
+### Updater
 
-In [above section](#model) we shows a network whose two layers share
-the same parameter matrix.  To specify such cases, we give "parameter
-names" to layers.  If some layers have the same paraemter names,
-`paddle.model.create` creates a single parameter matrix for these
-layers:
+Please be aware that a trainer can accept an updater as its data
+member, where an updater is a class derived from
+`paddle.trainer.Updater`.  This is to make it easier to customize
+trainers, as discussed
+[here](https://github.com/PaddlePaddle/Paddle/issues/1319).
 
-```python
-text1 = paddle.layer.data(...)
-sematic1 = paddle.layer.fc(text1, ..., parameter_name="sematic_projection")
-text2 = paddle.layer.data(...)
-sematic2 = paddle.layer.fc(text2, ..., parameter_name="sematic_projection")
-out = paddle.layer.cosine(semantic1, semantic2)
-```
+### Event Handler
 
-We can also share parameter matrices between layers in different
-models.  To do this, we need an additional parameter that refers to a
-model:
+`paddle.train` and `paddle.trainer.XXX.train` take an optional
+parameter `event_handler`, which should be either `None` or a function
+that handle some events:
 
-```python
-model1_input = paddle.layer.data(...)
-model1_output = paddle.layer.softmax(model1_input, ...,
-                                     parameter_name="a_parameter_matrix")
-model1 = paddle.model.create(model1_output)
-
-# Another model
-model2_semantic = paddle.layer.fc(text2, ...,
-                                  parameter_name="a_parameter_matrix",
-                                  parameter_model=model1)
-```
+1. BeginTraining
+1. EndTraining
+1. BeginIteration
+1. EndIteration
+1. BeginPass
+1. EndPass
 
-### Training
+where EndPass is sent if and only if the reader yields
+`end_pass=True`.
 
-The recommended way to training a model is to call `paddle.train`,
-which simply calls `paddle.optimizer.Default`, a global variable of
-type `paddle.optimizer.SGD`.  Equivalently, we can do
+An example as follows:
 
 ```python
-opt = paddle.optimizer.SGD(...)
-opt.train(model, reader=read, ...)
+def event_handler(event):
+    if ininstance(event, paddle.event.EndIteration):
+        print paddle.test(...)
+
+paddle.train(topology, parameters, reader, event_handler)
 ```
 
-#### Distributed Training
+If we are writing a PaddlePaddle program in and for iPython/Jypyter,
+we can use metaplotlib in the event handler to plot a curve of
+cost/error versus iterations, as shown
+[here](https://blog.dominodatalab.com/interactive-dashboards-in-jupyter/).
+
+### Distributed Training
 
 If users want to do distributed training on a cluster, s/he should
 call `paddle.dist_train` and provides access tokens to the cluster as
@@ -240,8 +224,9 @@ access a Kubernetes cluster, s/he should be able to call
 
 ```python
 paddle.dist_train(model,
+                  trainer=paddle.trainer.SGD(...,
+                                             paddle.updater.Adam(...)),
                   reader=read,
-                  optimizer=paddle.optimizer.SGDOptimizer(...),
                   k8s_user="yi",
                   k8s_token="kube_cluster_tls.pem",
                   k8s_job="hello",
@@ -251,7 +236,7 @@ paddle.dist_train(model,
 The pseudo code if `paddle.dist_train` is as follows:
 
 ```python
-def dist_train():
+def dist_train(topology, parameters, trainer, reader, ...):
     if os.getenv("KUBERNETES_SERVICE_HOST") == None:
         image_name = k8s_user + '/' + k8s_job
         docker_build(image_name)
@@ -264,13 +249,13 @@ def dist_train():
         elif rank < 15:
             parameter_server()
         else:
-            optimizer.train(model, reader=read)
+            trainer.train(model, reader=read)
 ```
 
 Please be aware that if a process is running on the Kubernetes
 cluster, it will have some environment variables pre-defined.
 
-If `dist_train` doesn't see these environment variables, it knowns
+If `dist_train` doesn't see these environment variables, it knows
 that it's running on users' personal computer, and it should work as a
 *launcher*.  Otherwise, it knows that it's running on the cluster and
 need to figure out its role as either the master, or a trainer, or a

doc/design/multi_language_interface/why_plain_c.md

+# Paddle多语言接口实现
+## 背景
+
+Paddle需要一个多语言接口，这个接口需要做到:
+
+* 有标准的，良好的文档
+    * 例如Python可以使用[Sphinx](http://www.sphinx-doc.org/en/stable/)生成API文档，golang可以使用[GoDoc](https://godoc.org/golang.org/x/tools/cmd/godoc)生成文档。这都需要这个接口按照约定俗成的规则来注释完备。
+* 不同语言的接口适应不同语言的特性
+    * 例如Java与Python的错误处理是直接扔出来Exception，而对于golang错误处理应该使用返回值。
+
+## 基本要求
+
+Paddle的多语言接口实现包括一下几个方面:
+
+* 我们使用动态库来分发Paddle。在这个动态库中不嵌入任何其他语言的解释器，也不使用其他动态库。
+* 这个动态库使用C99标准的头文件导出一些函数，不使用/导出C++符号。
+* 不导出Paddle内部的结构体、类，仅仅使用`void*`指针作为类型的句柄(handler)。
+* 不使用SWIG这种代码生成器，而是手写多语言绑定。
+
+
+## 原因
+
+### 使用动态库来分发Paddle
+
+* Paddle的链接方式比较复杂
+    * 如果用户要把Paddle的静态库（libpaddle.a）链接到自己的程序里，得使用 `--whole-archive` (for GCC) 或者 `--force_load` (for Clang) 参数，来确保把 libpaddle.a 里所有的符号都写入自己的程序的二进制文件里。这是因为 Paddle 的源码里使用了[object factory design pattern](http://stackoverflow.com/a/1310326/724872)。
+* 编译型语言，例如C/C++使用静态库和动态库难度差不多。但是解释性语言，例如[Python](http://stackoverflow.com/questions/19560594/how-to-import-static-library-in-python)或者[Java](http://stackoverflow.com/questions/24493337/linking-static-library-with-jni)，只能调用Paddle的动态库，否则得把Paddle静态库链接到解释器里。
+    * 解释性语言实际运行的二进制是解释器本身，如果调用静态库只能将静态库与解释器链接。例如对于Java来说，便是将静态库加入JVM中。这对于通常的Java的开发者来说，是不常见的做法。
+
+### 动态库中不嵌入任何其他语言的解释器
+
+* 目前Paddle的进程模型是C++内部驱动Python解释器进行模型配置解析和数据读取
+* 我们最终的动态库中不嵌入Python或者其他任何语言的解释器。模型配置解析，数据读取均交由其他语言完成
+
+现阶段Paddle有一个问题是，Paddle内嵌的Python解释器和外部使用的Python如果版本不同，会直接报错退出。
+
+### Paddle动态库中，不引用其他动态库
+
+* 即这个动态库是不依赖于其他任何文件的，可以在任何机器上执行的。
+
+###  这个动态库使用C99标准的头文件导出一些函数，不使用/导出C++符号
+
+* 由于C++编译器没有[名字修饰](https://en.wikipedia.org/wiki/Name_mangling#C.2B.2B)的规范，不同版本的编译器之间，对于同一段C++代码生成的符号可能不一致。而多语言接口需要直接读取生成的二进制(动态库)，需要有稳定的导出符号。
+* C语言是有导出符号的标准的，并且在常见的平台上，都是ABI调用标准的。
+* 大多数语言都支持使用C语言API
+* 使用C99而不使用C89，是因为C99支持[Fixed-width integer types](https://en.wikipedia.org/wiki/C_data_types#Fixed-width_integer_types)和[Boolean type](https://en.wikipedia.org/wiki/C_data_types#Boolean_type)。
+* 使用C99而不使用C11的原因是，[C11](https://en.wikipedia.org/wiki/C11_(C_standard_revision))并没有Paddle特别需要的特性，且C99相对于C11使用更加广泛。
+
+### 不导出Paddle内部的结构体、类，仅仅使用`void*`指针作为类型的句柄(handler)
+
+* Paddle内部的类为C++书写，直接导出到C的接口比较困难。
+* 在C-API中使用`void*`来表示Paddle内部类。再在每一个API中自己检查类型。
+
+在C的头文件 `paddle_matrix.h` 中:
+
+```C
+typedef void* paddle_matrix;
+typedef int paddle_error;
+
+extern "C"
+paddle_error paddle_matrix_shape(paddle_matrix matrix,
+                                 uint64_t* width,
+                                 uint64_t* height);
+```
+而在CPP里面实现这个C的接口，文件 `paddle_matrix.cpp`
+
+```cpp
+#include "paddle/math/matrix.hpp"
+extern "C"
+paddle_error paddle_matrix_shape(paddle_matrix matrix,
+                                 uint64_t *width,
+                                 uint64_t *height) {
+  auto m = (paddle::math::matrix*)(matrix);
+  *width = m->width();
+  *height = m->height();
+}
+```
+
+其中`paddle/math/matrix.hpp`文件内容为:
+
+```cpp
+namespace paddle {
+namespace math {  
+
+class Matrix {
+  //...
+};
+
+}  // namespace math
+}  // namespace paddle
+```
+
+### 不使用SWIG这种代码生成器，而是手写多语言绑定
+
+* [SWIG](http://www.swig.org/)是一个多语言接口的代码生成器。他的目标是使用C/C++写代码，SWIG直接读取C/C++的头文件，生成各种语言的绑定代码。
+    * 对于多语言接口，SWIG需要写一个interface文件。这个文件具有独特的语法，学习成本高。且增加一个第三方语言，就需要对这个第三方语言增加一些定义。有的时候，interface文件的写法非常[tricky](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/api/Paddle.swig#L36)。社区贡献代码学习成本高。
+    * SWIG暴露的接口保留了C++的接口样式，很难保证多语言代码风格的一致性。(函数命名，错误处理)
+        * 因为SWIG在第三方语言中暴露的函数名，类名和C++中完全一致。C++的命名风格并不能适应其他第三方语言。如果使用SWIG我们需要将在interface文件里，将大量的`SomeCppClass`重命名成`some_python_class`，或者`SomeGoTypes`。
+        * 对于不同语言，错误处理的方式也不尽相同。例如对于Java或者Python，最常见的错误处理方式是Exception，而对于Golang，错误处理方式是返回值。而SWIG只能简单的暴露C++接口，无法做到对于各种语言错误处理方式的适配。
+    * 对于大多数语言，直接使用C语言的.h并不困难。例如Python的[cffi](https://cffi.readthedocs.io/en/latest/overview.html#simple-example-abi-level-in-line)或者[Cython](http://cython.org/), golang的[cgo](https://golang.org/cmd/cgo/)。
+    * SWIG支持的语言或者解释器有局限。例如对于Python，使用SWIG只支持CPython解释器，而不支持PyPy解释器。
+
+
+## 原因列表
+
+| 结论 | 对比 | 原因 |
+|---| --- | --- |
+| 使用动态库 | 不使用静态库 | 解释型语言只能调用动态库，Paddle静态库链接复杂 |
+| 不嵌入其他语言解释器 | 不嵌入Python解释器 | Paddle C++目前嵌入Python解释器，会导致不同版本Python在一个进程里的bug |
+| 不引用其他动态库 | | Paddle一个动态库可以在任何Linux系统上运行 |
+| 使用C99做接口 | 不使用C++做接口 | C有标准的ABI，C99是目前C最广泛的使用标准，且C99支持bool类型和定长整数(uint64_t等)类型 |
+| 使用void*作为类句柄 | 不显示的写每个类具体包含什么| 实现简单，并且让接口脱离实现细节 |
+| 手写多语言绑定 | 不使用SWIG | 使用SWIG需要多语言绑定的开发人员熟练掌握SWIG配置，社区参与困难。SWIG生成的代码不能保证多语言代码风格的一致性 |
+
+
+## 简单实现
+
+TBD

doc/design/reader/README.md

+# Python Data Reader Design Doc
+
+At training and testing time, PaddlePaddle programs need to read data. To ease the users' work to write data reading code, we define that
+
+- A *reader* is a function that reads data (from file, network, random number generator, etc) and yields data items.
+- A *reader creator* is a function that returns a reader function.
+- A *reader decorator* is a function, which accepts one or more readers, and returns a reader.
+- A *batch reader* is a function that reads data (from *reader*, file, network, random number generator, etc) and yields a batch of data items.
+
+and provide function which converts reader to batch reader, frequently used reader creators and reader decorators.
+
+## Data Reader Interface
+
+Indeed, *data reader* doesn't have to be a function that reads and yields data items. It can be any function with no parameter that creates a iterable (anything can be used in `for x in iterable`):
+
+```
+iterable = data_reader()
+```
+
+Element produced from the iterable should be a **single** entry of data, **not** a mini batch. That entry of data could be a single item, or a tuple of items. Item should be of [supported type](http://www.paddlepaddle.org/doc/ui/data_provider/pydataprovider2.html?highlight=dense_vector#input-types) (e.g., numpy 1d array of float32, int, list of int)
+
+An example implementation for single item data reader creator:
+
+```python
+def reader_creator_random_image(width, height):
+    def reader():
+        while True:
+            yield numpy.random.uniform(-1, 1, size=width*height)
+    return reader
+```
+
+An example implementation for multiple item data reader creator:
+```python
+def reader_creator_random_image_and_label(width, height, label):
+    def reader():
+        while True:
+            yield numpy.random.uniform(-1, 1, size=width*height), label
+    return reader
+```
+
+## Batch Reader Interface
+
+*batch reader* can be any function with no parameter that creates a iterable (anything can be used in `for x in iterable`). The output of the iterable should be a batch (list) of data items. Each item inside the list must be a tuple.
+
+Here are valid outputs:
+```python
+# a mini batch of three data items. Each data item consist three columns of data, each of which is 1.
+[(1, 1, 1),
+(2, 2, 2),
+(3, 3, 3)]
+
+# a mini batch of three data items, each data item is a list (single column).
+[([1,1,1],),
+([2,2,2],),
+([3,3,3],),
+```
+
+Please note that each item inside the list must be a tuple, below is an invalid output:
+```python
+ # wrong, [1,1,1] needs to be inside a tuple: ([1,1,1],).
+ # Otherwise it's ambiguous whether [1,1,1] means a single column of data [1, 1, 1],
+ # or three column of datas, each of which is 1.
+[[1,1,1],
+[2,2,2],
+[3,3,3]]
+```
+
+It's easy to convert from reader to batch reader:
+```python
+mnist_train = paddle.dataset.mnist.train()
+mnist_train_batch_reader = paddle.batch(mnist_train, 128)
+```
+
+Also easy to create custom batch reader:
+```python
+def custom_batch_reader():
+    while True:
+        batch = []
+        for i in xrange(128):
+            batch.append((numpy.random.uniform(-1, 1, 28*28),)) # note that it's a tuple being appended.
+        yield batch
+
+mnist_random_image_batch_reader = custom_batch_reader
+```
+
+## Usage
+
+batch reader, mapping from item(s) read to data layer, batch size and number of total pass will be passed into `paddle.train`:
+
+```python
+# two data layer is created:
+image_layer = paddle.layer.data("image", ...)
+label_layer = paddle.layer.data("label", ...)
+
+# ...
+batch_reader = paddle.batch(paddle.dataset.mnist.train(), 128)
+paddle.train(batch_reader, {"image":0, "label":1}, 128, 10, ...)
+```
+
+## Data Reader Decorator
+
+*Data reader decorator* takes a single or multiple data reader, returns a new data reader. It is similar to a [python decorator](https://wiki.python.org/moin/PythonDecorators), but it does not use `@` syntax.
+
+Since we have a strict interface for data readers (no parameter, return a single data item). Data reader can be used flexiable via data reader decorators. Following are a few examples:
+
+### Prefetch Data
+
+Since reading data may take time and training can not proceed without data. It is generally a good idea to prefetch data.
+
+Use `paddle.reader.buffered` to prefetch data:
+
+```python
+buffered_reader = paddle.reader.buffered(paddle.dataset.mnist.train(), 100)
+```
+
+`buffered_reader` will try to buffer (prefetch) `100` data entries.
+
+### Compose Multiple Data Readers
+
+For example, we want to use a source of real images (reusing mnist dataset), and a source of random images as input for [Generative Adversarial Networks](https://arxiv.org/abs/1406.2661).
+
+We can do:
+
+```python
+def reader_creator_random_image(width, height):
+    def reader():
+        while True:
+            yield numpy.random.uniform(-1, 1, size=width*height)
+    return reader
+
+def reader_creator_bool(t):
+    def reader:
+        while True:
+            yield t
+    return reader
+
+true_reader = reader_creator_bool(True)
+false_reader = reader_creator_bool(False)
+
+reader = paddle.reader.compose(paddle.dataset.mnist.train(), data_reader_creator_random_image(20, 20), true_reader, false_reader)
+# Skipped 1 because paddle.dataset.mnist.train() produces two items per data entry.
+# And we don't care second item at this time.
+paddle.train(paddle.batch(reader, 128), {"true_image":0, "fake_image": 2, "true_label": 3, "false_label": 4}, ...)
+```
+
+### Shuffle
+
+Given shuffle buffer size `n`, `paddle.reader.shuffle` will return a data reader that buffers `n` data entries and shuffle them before a data entry is read.
+
+Example:
+```python
+reader = paddle.reader.shuffle(paddle.dataset.mnist.train(), 512)
+```
+
+## Q & A
+
+### Why reader return only a single entry, but not a mini batch?
+
+Always returning a single entry make reusing existing data readers much easier (e.g., if existing reader return not a single entry but 3 entries, training code will be more complex because it need to handle cases like batch size 2).
+
+We provide function `paddle.batch` to turn (single entry) reader into batch reader.
+
+### Why do we need batch reader, isn't train take reader and batch_size as arguments sufficient?
+
+In most of the case, train taking reader and batch_size as arguments would be sufficent. However sometimes user want to customize order of data entries inside a mini batch. Or even change batch size dynamically.
+
+### Why use a dictionary but not a list to provide mapping?
+
+We decided to use dictionary (`{"image":0, "label":1}`) instead of list (`["image", "label"]`) is because that user can easily resue item (e.g., using `{"image_a":0, "image_b":0, "label":1}`) or skip item (e.g., using `{"image_a":0, "label":2}`).
+
+### How to create custom data reader creator
+
+```python
+def image_reader_creator(image_path, label_path, n):
+    def reader():
+        f = open(image_path)
+        l = open(label_path)
+        images = numpy.fromfile(
+            f, 'ubyte', count=n * 28 * 28).reshape((n, 28 * 28)).astype('float32')
+        images = images / 255.0 * 2.0 - 1.0
+        labels = numpy.fromfile(l, 'ubyte', count=n).astype("int")
+        for i in xrange(n):
+            yield images[i, :], labels[i] # a single entry of data is created each time
+        f.close()
+        l.close()
+    return reader
+
+# images_reader_creator creates a reader
+reader = image_reader_creator("/path/to/image_file", "/path/to/label_file", 1024)
+paddle.train(paddle.batch(reader, 128), {"image":0, "label":1}, ...)
+```
+
+### How is `paddle.train` implemented
+
+An example implementation of paddle.train could be:
+
+```python
+def train(batch_reader, mapping, batch_size, total_pass):
+    for pass_idx in range(total_pass):
+        for mini_batch in batch_reader(): # this loop will never end in online learning.
+            do_forward_backward(mini_batch, mapping)
+```

doc/faq/index_cn.rst

@@ -286,3 +286,16 @@ PaddlePaddle的参数使用名字 :code:`name` 作为参数的ID，相同名字
 ..      code-block:: bash
 
         paddle train --use_gpu=true --trainer_count=2 --gpu_id=2
+
+
+12. 训练过程中出现 :code:`Floating point exception`, 训练因此退出怎么办?
+------------------------------------------------------------------------
+
+Paddle二进制在运行时捕获了浮点数异常，只要出现浮点数异常(即训练过程中出现NaN或者Inf)，立刻退出。浮点异常通常的原因是浮点数溢出、除零等问题。
+主要原因包括两个方面:
+
+* 训练过程中参数或者训练过程中的梯度尺度过大，导致参数累加，乘除等时候，导致了浮点数溢出。
+* 模型一直不收敛，发散到了一个数值特别大的地方。
+* 训练数据有问题，导致参数收敛到了一些奇异的情况。或者输入数据尺度过大，有些特征的取值达到数百万，这时进行矩阵乘法运算就可能导致浮点数溢出。
+
+主要的解决办法是减小学习律或者对数据进行归一化处理。

doc/getstarted/basic_usage/index_cn.rst

@@ -55,7 +55,7 @@ PaddlePaddle是源于百度的一个深度学习平台。这份简短的介绍
     # 线性计算网络层: ȳ = wx + b
     ȳ = fc_layer(input=x, param_attr=ParamAttr(name='w'), size=1, act=LinearActivation(), bias_attr=ParamAttr(name='b'))
     # 计算误差函数，即  ȳ 和真实 y 之间的距离
-    cost = regression_cost(input= ȳ, label=y)
+    cost = mse_cost(input= ȳ, label=y)
     outputs(cost)
 
 
@@ -69,7 +69,7 @@ PaddlePaddle是源于百度的一个深度学习平台。这份简短的介绍
     
     - **数据层**：数据层 `data_layer` 是神经网络的入口，它读入数据并将它们传输到接下来的网络层。这里数据层有两个，分别对应于变量 `x` 和 `y`。
     - **全连接层**：全连接层 `fc_layer` 是基础的计算单元，这里利用它建模变量之间的线性关系。计算单元是神经网络的核心，PaddlePaddle支持大量的计算单元和任意深度的网络连接，从而可以拟合任意的函数来学习复杂的数据关系。
-    - **回归误差代价层**：回归误差代价层 `regression_cost` 是众多误差代价函数层的一种，它们在训练过程作为网络的出口，用来计算模型的误差，是模型参数优化的目标函数。
+    - **回归误差代价层**：回归误差代价层 `mse_cost` 是众多误差代价函数层的一种，它们在训练过程作为网络的出口，用来计算模型的误差，是模型参数优化的目标函数。
 
 定义了网络结构并保存为 `trainer_config.py` 之后，运行以下训练命令：
 

doc/getstarted/basic_usage/index_en.rst

@@ -49,7 +49,7 @@ To recover this relationship between ``X`` and ``Y``, we use a neural network wi
         x = data_layer(name='x', size=1)
         y = data_layer(name='y', size=1)
         y_predict = fc_layer(input=x, param_attr=ParamAttr(name='w'), size=1, act=LinearActivation(), bias_attr=ParamAttr(name='b'))
-        cost = regression_cost(input=y_predict, label=y)
+        cost = mse_cost(input=y_predict, label=y)
         outputs(cost)
 
 Some of the most fundamental usages of PaddlePaddle are demonstrated:

doc/getstarted/build_and_install/docker_install_cn.rst

-安装PaddlePaddle的Docker镜像
-============================
+PaddlePaddle的Docker容器使用方式
+================================
 
-PaddlePaddle项目提供官方 `Docker <https://www.docker.com/>`_ 镜像。Docker镜像是我们目前唯一官方支持的部署和运行方式。
+PaddlePaddle目前唯一官方支持的运行的方式是Docker容器。因为Docker能在所有主要操作系统（包括Linux，Mac OS X和Windows）上运行。 请注意，您需要更改 `Dockers设置 <https://github.com/PaddlePaddle/Paddle/issues/627>`_ 才能充分利用Mac OS X和Windows上的硬件资源。
 
-下述内容将分为如下几个类别描述。
 
-* PaddlePaddle提供的Docker镜像版本
-* 下载和运行Docker镜像
-* 注意事项
+纯CPU和GPU的docker镜像使用说明
+------------------------------
 
-PaddlePaddle提供的Docker镜像版本
---------------------------------
+对于每一个PaddlePaddle版本，我们都会发布两个Docker镜像：纯CPU的和GPU的。
+我们通过设置 `dockerhub.com <https://hub.docker.com/r/paddledev/paddle/>`_ 自动生成最新的docker镜像：
+`paddledev/paddle:0.10.0rc1-cpu` 和 `paddledev/paddle:0.10.0rc1-gpu`。
 
-我们提供了12个 `Docker image <https://hub.docker.com/r/paddledev/paddle/tags/>`_ ，他们的image name都是 :code:`paddledev/paddle` ，tag分别为
+以交互容器方式运行纯CPU的镜像：
 
-+-----------------+------------------+------------------------+-----------------------+
-|                 |   normal         |           devel        |          demo         |
-+=================+==================+========================+=======================+
-|       CPU       | cpu-latest       | cpu-devel-latest       | cpu-demo-latest       |
-+-----------------+------------------+------------------------+-----------------------+
-|       GPU       | gpu-latest       | gpu-devel-latest       | gpu-demo-latest       |
-+-----------------+------------------+------------------------+-----------------------+
-| CPU WITHOUT AVX | cpu-noavx-latest | cpu-noavx-devel-latest | cpu-noavx-demo-latest |
-+-----------------+------------------+------------------------+-----------------------+
-| GPU WITHOUT AVX | gpu-noavx-latest | gpu-noavx-devel-latest | gpu-noavx-demo-latest |
-+-----------------+------------------+------------------------+-----------------------+
+.. code-block:: bash
 
-其中，横向包括三个版本，normal，devel和demo。
+    docker run -it --rm paddledev/paddle:0.10.0rc1-cpu /bin/bash
 
-* Normal: 正常的Docker image，只包括paddle的二进制
-* Devel: 包括Paddle的二进制、编译环境和源代码
-* Demo: 包括Paddle运行demo所需要的依赖
+或者，可以以后台进程方式运行容器：
 
-纵向包括四个版本，他们是。
+.. code-block:: bash
 
-* CPU: CPU版本。需要支持AVX指令集的CPU
-* GPU: GPU版本。需要支持AVX指令集的CPU
-* CPU WITHOUT AVX: CPU版本，不支持AVX指令集的CPU也可以运行
-* GPU WITHOUT AVX: GPU版本，不需要AVX指令集的CPU也可以运行。
+    docker run -d -p 2202:22 -p 8888:8888 paddledev/paddle:0.10.0rc1-cpu
 
-用户可以选择对应版本的docker image。使用如下脚本可以确定本机的CPU是否支持 :code:`AVX` 指令集\:
+然后用密码 :code:`root` SSH进入容器：
 
-..  code-block:: bash
+.. code-block:: bash
 
-    if cat /proc/cpuinfo | grep -q avx ; then echo "Support AVX"; else echo "Not support AVX"; fi
+    ssh -p 2202 root@localhost
 
-如果输出 :code:`Support AVX`，则可以选择上表中的AVX版本PaddlePaddle。否则需要选择非AVX的PaddlePaddle。选择普通CPU版本的devel版本的image，则可以使用 :code:`paddledev/paddle:cpu-devel-latest` 来引用这个image。
+SSH方式的一个优点是我们可以从多个终端进入容器。比如，一个终端运行vi，另一个终端运行Python。另一个好处是我们可以把PaddlePaddle容器运行在远程服务器上，并在笔记本上通过SSH与其连接。
 
-PaddlePaddle提供的镜像并不包含任何命令运行，想要运行PaddlePaddle，您需要进入镜像运行PaddlePaddle
-程序或者自定义一个含有启动脚本的image。具体请参考注意事项中的 :code:`使用ssh访问PaddlePaddle镜像`
 
-下载和运行Docker镜像
---------------------
+以上方法在GPU镜像里也能用－只是请不要忘记按装CUDA驱动，以及告诉Docker：
 
-为了运行PaddlePaddle的docker镜像，您需要在机器中安装好Docker。安装Docker需要您的机器
-至少具有3.10以上的linux kernel。安装方法请参考
-`Docker的官方文档 <https://docs.docker.com/engine/installation/>`_ 。如果您使用
-mac osx或者是windows机器，请参考 
-`mac osx的安装文档 <https://docs.docker.com/engine/installation/mac/>`_ 和
-`windows 的安装文档 <https://docs.docker.com/engine/installation/windows/>`_ 。
+.. code-block:: bash
 
-您可以使用 :code:`docker pull` 命令预先下载镜像，也可以直接执行 
-:code:`docker run` 命令运行镜像。执行方法如下:
+    export CUDA_SO="$(\ls /usr/lib64/libcuda* | xargs -I{} echo '-v {}:{}') $(\ls /usr/lib64/libnvidia* | xargs -I{} echo '-v {}:{}')"
+    export DEVICES=$(\ls /dev/nvidia* | xargs -I{} echo '--device {}:{}')
+    docker run ${CUDA_SO} ${DEVICES} -it paddledev/paddle:0.10.0rc1-gpu
 
-..  code-block:: bash
+
+运行PaddlePaddle书籍
+---------------------
+
+Jupyter Notebook是一个开源的web程序，大家可以通过它制作和分享带有代码、公式、图表、文字的交互式文档。用户可以通过网页浏览文档。
+
+PaddlePaddle书籍是为用户和开发者制作的一个交互式的Jupyter Nodebook。
+如果您想要更深入了解deep learning，PaddlePaddle书籍一定是您最好的选择。
+
+当您进入容器内之后，只用运行以下命令：
+
+.. code-block:: bash
+        
+    jupyter notebook
+
+然后在浏览器中输入以下网址：
     
-    $ docker run -it paddledev/paddle:cpu-latest
+.. code-block:: text
 
-即可启动和进入PaddlePaddle的container。如果运行GPU版本的PaddlePaddle，则需要先将
-cuda相关的Driver和设备映射进container中，脚本类似于
+    http://localhost:8888/
 
-..  code-block:: bash
+就这么简单，享受您的旅程！
 
-    $ export CUDA_SO="$(\ls /usr/lib64/libcuda* | xargs -I{} echo '-v {}:{}') $(\ls /usr/lib64/libnvidia* | xargs -I{} echo '-v {}:{}')"
-    $ export DEVICES=$(\ls /dev/nvidia* | xargs -I{} echo '--device {}:{}')
-    $ docker run ${CUDA_SO} ${DEVICES} -it paddledev/paddle:gpu-latest
 
-进入Docker container后，运行 :code:`paddle version` 即可打印出PaddlePaddle的版本和构建
-信息。安装完成的PaddlePaddle主体包括三个部分， :code:`paddle` 脚本， python的
-:code:`paddle` 包和 :code:`py_paddle` 包。其中\:
+非AVX镜像
+---------
 
-* :code:`paddle` 脚本和 :code:`paddle` 的python包是PaddlePaddle的训练主要程序。使用 
-  :code:`paddle` 脚本可以启动PaddlePaddle的训练进程和pserver。而 :code:`paddle` 脚本
-  中的二进制使用了 :code:`paddle` 的python包来做配置文件解析等工作。
-* python包 :code:`py_paddle` 是一个swig封装的PaddlePaddle包，用来做预测和简单的定制化
-  训练。
+纯CPU镜像以及GPU镜像都会用到AVX指令集，但是2008年之前生产的旧电脑不支持AVX。以下指令能检查Linux电脑是否支持AVX：
 
-注意事项
---------
+.. code-block:: bash
 
-性能问题
-++++++++
+   if cat /proc/cpuinfo | grep -i avx; then echo Yes; else echo No; fi
 
-由于Docker是基于容器的轻量化虚拟方案，所以在CPU的运算性能上并不会有严重的影响。
-而GPU的驱动和设备全部映射到了容器内，所以GPU在运算性能上也不会有严重的影响。
+如果输出是No，我们就需要手动编译一个非AVX版本的镜像：
 
-但是如果使用了高性能的网卡，例如RDMA网卡(RoCE 40GbE 或者 IB 56GbE)，或者高性能的
-以太网卡 (10GbE)。推荐使用将本地网卡，即 "--net=host" 来进行训练。而不使用docker
-的网桥来进行网络通信。
+.. code-block:: bash
 
-远程访问问题和二次开发
-++++++++++++++++++++++
+   cd ~
+   git clone https://github.com/PaddlePaddle/Paddle.git
+   cd Paddle
+   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 .
 
-由于PaddlePaddle的Docker镜像并不包含任何预定义的运行命令。所以如果想要在后台启用ssh
-远程访问，则需要进行一定的二次开发，将ssh装入系统内并开启远程访问。二次开发可以
-使用Dockerfile构建一个全新的docker image。需要参考 
-`Dockerfile的文档 <https://docs.docker.com/engine/reference/builder/>`_ 和
-`Dockerfile的最佳实践 <https://docs.docker.com/engine/userguide/eng-image/dockerfile_best-practices/>`_ 
-两个文档。
 
-简单的含有ssh的Dockerfile如下：
+通过Docker容器开发PaddlePaddle
+------------------------------
 
-..  code-block:: bash
+开发人员可以在Docker中开发PaddlePaddle。这样开发人员可以以一致的方式在不同的平台上工作 - Linux，Mac OS X和Windows。
 
-    FROM paddledev/paddle:cpu-latest
+1. 将开发环境构建为Docker镜像
+   
+   .. code-block:: bash
 
-    MAINTAINER PaddlePaddle dev team <paddle-dev@baidu.com>
+      git clone --recursive https://github.com/PaddlePaddle/Paddle
+      cd Paddle
+      docker build -t paddle:dev -f paddle/scripts/docker/Dockerfile .
 
-    RUN apt-get update
-    RUN apt-get install -y openssh-server
-    RUN mkdir /var/run/sshd
-    RUN echo 'root:root' | chpasswd
 
-    RUN sed -ri 's/^PermitRootLogin\s+.*/PermitRootLogin yes/' /etc/ssh/sshd_config
-    RUN sed -ri 's/UsePAM yes/#UsePAM yes/g' /etc/ssh/sshd_config
+   请注意，默认情况下，:code:`docker build` 不会将源码导入到镜像中并编译它。如果我们想这样做，需要设置一个参数：
 
-    EXPOSE 22
+   .. code-block:: bash
 
-    CMD    ["/usr/sbin/sshd", "-D"]
+      docker build -t paddle:dev -f paddle/scripts/docker/Dockerfile --build-arg BUILD_AND_INSTALL=ON .
 
 
-使用该Dockerfile构建出镜像，然后运行这个container即可。相关命令为\:
+2. 运行开发环境
 
-..  code-block:: bash
+   当我们编译好了 :code:`paddle:dev`， 我们可以在docker容器里做开发，源代码可以通过挂载本地文件来被载入Docker的开发环境里面：
+   
+   .. code-block:: bash
 
-    # cd到含有Dockerfile的路径中
-    $ docker build . -t paddle_ssh
-    # 运行这个container，将宿主机的8022端口映射到container的22端口上
-    $ docker run -d -p 8022:22  --name paddle_ssh_machine paddle_ssh
+      docker run -d -p 2202:22 -v $PWD:/paddle paddle:dev
 
-执行如下命令即可以关闭这个container，并且删除container中的数据\:
+   以上代码会启动一个带有PaddlePaddle开发环境的docker容器，源代码会被挂载到 :code:`/paddle` 。
 
-..  code-block:: bash
-    
-    # 关闭container
-    $ docker stop paddle_ssh_machine
-    # 删除container
-    $ docker rm paddle_ssh_machine
+   请注意， :code:`paddle:dev` 的默认入口是 :code:`sshd` 。以上的 :code:`docker run` 命令其实会启动一个在2202端口监听的SSHD服务器。这样，我们就能SSH进入我们的开发容器了：
+   
+   .. code-block:: bash
+
+      ssh root@localhost -p 2202
+
+3. 在Docker开发环境中编译与安装PaddlPaddle代码
+
+   当在容器里面的时候，可以用脚本 :code:`paddle/scripts/docker/build.sh` 来编译、安装与测试PaddlePaddle：
+   
+   .. code-block:: bash
+		      
+      /paddle/paddle/scripts/docker/build.sh
+
+   以上指令会在 :code:`/paddle/build` 中编译PaddlePaddle。通过以下指令可以运行单元测试：
+   
+   .. code-block:: bash
+
+      cd /paddle/build
+      ctest
+
+
+文档
+----
+
+Paddle的Docker镜像带有一个通过 `woboq code browser
+<https://github.com/woboq/woboq_codebrowser>`_ 生成的HTML版本的C++源代码，便于用户浏览C++源码。
 
-如果想要在外部机器访问这个container，即可以使用ssh访问宿主机的8022端口。用户名为
-root，密码也是root。命令为\:
+只要在Docker里启动PaddlePaddle的时候给它一个名字，就可以再运行另一个Nginx Docker镜像来服务HTML代码：
 
-..  code-block:: bash
+.. code-block:: bash
 
-    $ ssh -p 8022 root@YOUR_HOST_MACHINE
+   docker run -d --name paddle-cpu-doc paddle:0.10.0rc1-cpu
+   docker run -d --volumes-from paddle-cpu-doc -p 8088:80 nginx
 
-至此，您就可以远程的使用PaddlePaddle啦。
+接着我们就能够打开浏览器在 http://localhost:8088/paddle/ 浏览代码。

doc/getstarted/build_and_install/docker_install_en.rst

@@ -9,6 +9,100 @@ Please be aware that you will need to change `Dockers settings
 of your hardware resource on Mac OS X and Windows.
 
 
+Usage of CPU-only and GPU Images
+----------------------------------
+
+For each version of PaddlePaddle, we release 2 Docker images, a
+CPU-only one and a CUDA GPU one.  We do so by configuring
+`dockerhub.com <https://hub.docker.com/r/paddledev/paddle/>`_
+automatically generate the latest docker images `paddledev/paddle:0.10.0rc1-cpu`
+and `paddledev/paddle:0.10.0rc1-gpu`.
+
+To run the CPU-only image as an interactive container:
+
+.. code-block:: bash
+
+    docker run -it --rm paddledev/paddle:0.10.0rc1-cpu /bin/bash
+
+or, we can run it as a daemon container
+
+.. code-block:: bash
+
+    docker run -d -p 2202:22 -p 8888:8888 paddledev/paddle:0.10.0rc1-cpu
+
+and SSH to this container using password :code:`root`:
+
+.. code-block:: bash
+
+    ssh -p 2202 root@localhost
+
+An advantage of using SSH is that we can connect to PaddlePaddle from
+more than one terminals.  For example, one terminal running vi and
+another one running Python interpreter.  Another advantage is that we
+can run the PaddlePaddle container on a remote server and SSH to it
+from a laptop.
+
+Above methods work with the GPU image too -- just please don't forget
+to install CUDA driver and let Docker knows about it:
+
+.. code-block:: bash
+
+    export CUDA_SO="$(\ls /usr/lib64/libcuda* | xargs -I{} echo '-v {}:{}') $(\ls /usr/lib64/libnvidia* | xargs -I{} echo '-v {}:{}')"
+    export DEVICES=$(\ls /dev/nvidia* | xargs -I{} echo '--device {}:{}')
+    docker run ${CUDA_SO} ${DEVICES} -it paddledev/paddle:0.10.0rc1-gpu
+
+
+PaddlePaddle Book
+------------------
+
+The Jupyter Notebook is an open-source web application that allows
+you to create and share documents that contain live code, equations,
+visualizations and explanatory text in a single browser.
+
+PaddlePaddle Book is an interactive Jupyter Notebook for users and developers. 
+We already exposed port 8888 for this book. If you want to
+dig deeper into deep learning, PaddlePaddle Book definitely is your best choice.
+
+Once you are inside the container, simply issue the command:
+
+.. code-block:: bash
+        
+    jupyter notebook
+
+Then, you would back and paste the address into the local browser:
+    
+.. code-block:: text
+
+    http://localhost:8888/
+
+That's all. Enjoy your journey!
+
+
+Non-AVX Images
+--------------
+
+Please be aware that the CPU-only and the GPU images both use the AVX
+instruction set, but old computers produced before 2008 do not support
+AVX.  The following command checks if your Linux computer supports
+AVX:
+
+.. code-block:: bash
+
+   if cat /proc/cpuinfo | grep -i avx; then echo Yes; else echo No; fi
+
+
+If it doesn't, we will need to build non-AVX images manually from
+source code:
+
+.. code-block:: bash
+
+   cd ~
+   git clone https://github.com/PaddlePaddle/Paddle.git
+   cd Paddle
+   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 .
+
+
 Development Using Docker
 ------------------------
 
@@ -42,7 +136,7 @@ Windows -- in a consistent way.
 
    .. code-block:: bash
 
-      docker run -d -p 2202:22 -v $PWD:/paddle paddle:dev
+      docker run -d -p 2202:22 -p 8888:8888 -v $PWD:/paddle paddle:dev
 
    This runs a container of the development environment Docker image
    with the local source tree mounted to :code:`/paddle` of the
@@ -83,80 +177,6 @@ Windows -- in a consistent way.
       ctest
 
 
-CPU-only and GPU Images
------------------------
-
-For each version of PaddlePaddle, we release 2 Docker images, a
-CPU-only one and a CUDA GPU one.  We do so by configuring
-`dockerhub.com <https://hub.docker.com/r/paddledev/paddle/>`_
-automatically runs the following commands:
-
-.. code-block:: bash
-
-   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:
-
-.. code-block:: bash
-
-    docker run -it --rm paddledev/paddle:cpu-latest /bin/bash
-
-or, we can run it as a daemon container
-
-.. code-block:: bash
-
-    docker run -d -p 2202:22 paddledev/paddle:cpu-latest
-
-and SSH to this container using password :code:`root`:
-
-.. code-block:: bash
-
-    ssh -p 2202 root@localhost
-
-An advantage of using SSH is that we can connect to PaddlePaddle from
-more than one terminals.  For example, one terminal running vi and
-another one running Python interpreter.  Another advantage is that we
-can run the PaddlePaddle container on a remote server and SSH to it
-from a laptop.
-
-
-Above methods work with the GPU image too -- just please don't forget
-to install CUDA driver and let Docker knows about it:
-
-.. code-block:: bash
-
-    export CUDA_SO="$(\ls /usr/lib64/libcuda* | xargs -I{} echo '-v {}:{}') $(\ls /usr/lib64/libnvidia* | xargs -I{} echo '-v {}:{}')"
-    export DEVICES=$(\ls /dev/nvidia* | xargs -I{} echo '--device {}:{}')
-    docker run ${CUDA_SO} ${DEVICES} -it paddledev/paddle:gpu-latest
-
-
-Non-AVX Images
---------------
-
-Please be aware that the CPU-only and the GPU images both use the AVX
-instruction set, but old computers produced before 2008 do not support
-AVX.  The following command checks if your Linux computer supports
-AVX:
-
-.. code-block:: bash
-
-   if cat /proc/cpuinfo | grep -i avx; then echo Yes; else echo No; fi
-
-
-If it doesn't, we will need to build non-AVX images manually from
-source code:
-
-.. code-block:: bash
-
-   cd ~
-   git clone https://github.com/PaddlePaddle/Paddle.git
-   cd Paddle
-   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
 -------------
 
@@ -171,7 +191,7 @@ container:
 
 .. code-block:: bash
 
-   docker run -d --name paddle-cpu-doc paddle:cpu
+   docker run -d --name paddle-cpu-doc paddle:0.10.0rc1-cpu
    docker run -d --volumes-from paddle-cpu-doc -p 8088:80 nginx
 
 

doc/howto/index_cn.rst

@@ -10,6 +10,7 @@
   usage/cmd_parameter/index_cn.rst
   usage/concepts/use_concepts_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
 

doc/howto/usage/cluster/cluster_train_cn.md

@@ -6,7 +6,7 @@
 
 在本文中，我们将阐释如何在集群上运行分布式 Paddle 训练作业。我们将以[推荐系统](https://github.com/baidu/Paddle/tree/develop/demo/recommendation)为例创建分布式的单进程训练。
 
-在本文中使用的[脚本](https://github.com/baidu/Paddle/tree/develop/paddle/scripts/cluster_train)通过 SSH 运行分布式作业。 它们还可以供那些运行更复杂的集群管理系统（如 MPI 和 [Kubernetes](https://github.com/PaddlePaddle/Paddle/tree/develop/doc/howto/usage/cluster/k8s) ）的用户参考。
+在本文中使用的[脚本](https://github.com/baidu/Paddle/tree/develop/paddle/scripts/cluster_train)通过 SSH 运行分布式作业。 它们还可以供那些运行更复杂的集群管理系统（如 MPI 和 [Kubernetes](https://github.com/PaddlePaddle/Paddle/tree/develop/doc/howto/usage/k8s) ）的用户参考。
 
 ## 前提条件
 

doc/howto/usage/cluster/cluster_train_en.md

@@ -2,7 +2,7 @@
 
 In this article, we explain how to run distributed Paddle training jobs on clusters.  We will create the distributed version of the single-process training example, [recommendation](https://github.com/baidu/Paddle/tree/develop/demo/recommendation).
 
-[Scripts](https://github.com/baidu/Paddle/tree/develop/paddle/scripts/cluster_train) used in this article launch distributed jobs via SSH.  They also work as a reference for users running more sophisticated cluster management systems like MPI and [Kubernetes](https://github.com/PaddlePaddle/Paddle/tree/develop/doc/howto/usage/cluster/k8s).
+[Scripts](https://github.com/baidu/Paddle/tree/develop/paddle/scripts/cluster_train) used in this article launch distributed jobs via SSH.  They also work as a reference for users running more sophisticated cluster management systems like MPI and [Kubernetes](https://github.com/PaddlePaddle/Paddle/tree/develop/doc/howto/usage/k8s).
 
 ## Prerequisite
 

doc/howto/usage/cmd_parameter/arguments_cn.md

@@ -228,16 +228,6 @@
 <td class="left"></td><td class="left"></td><td class="left">√</td><td class="left">√</td>
 </tr>
 
-<tr>
-<td class="left" rowspan = "2">度量学习(metric learning)</td><td class="left">external</td>
-<td class="left">√</td><td class="left">√</td><td class="left">√</td><td class="left">√</td>
-</tr>
-
-<tr>
-<td class="left">data_server_port</td>
-<td class="left"></td><td class="left">√</td><td class="left"></td><td class="left">√</td>
-</tr>
-
 <tr>
 <td class="left" rowspan = "16">参数服务器(PServer)</td><td class="left">start_pserver</td>
 <td class="left"></td><td class="left">√</td><td class="left"></td><td class="left">√</td>

doc/howto/usage/cmd_parameter/arguments_en.md

@@ -228,16 +228,6 @@ It looks like there are a lot of arguments. However, most of them are for develo
 <td class="left"></td><td class="left"></td><td class="left">√</td><td class="left">√</td>
 </tr>
 
-<tr>
-<td class="left" rowspan = "2">metric learning</td><td class="left">external</td>
-<td class="left">√</td><td class="left">√</td><td class="left">√</td><td class="left">√</td>
-</tr>
-
-<tr>
-<td class="left">data_server_port</td>
-<td class="left"></td><td class="left">√</td><td class="left"></td><td class="left">√</td>
-</tr>
-
 <tr>
 <td class="left" rowspan = "16">PServer</td><td class="left">start_pserver</td>
 <td class="left"></td><td class="left">√</td><td class="left"></td><td class="left">√</td>

doc/howto/usage/cmd_parameter/detail_introduction_cn.md

@@ -180,15 +180,6 @@
   - 用户可以自定义beam search的方法，编译成动态库，供PaddlePaddle加载。 该参数用于指定动态库路径.
   - 类型: string (默认: "", null).
 
-## 度量学习(Metric Learning)
-* `--external`
-   - 指示是否使用外部机器进行度量学习.
-   - 类型: bool (默认: 0).
-
-* `--data_server_port`
-  - 数据服务器(data server)的监听端口，主要用在度量学习中.
-  - 类型: int32 (默认: 21134).
-
 ## 数据支持(DataProvider)
 
 * `--memory_threshold_on_load_data`

doc/howto/usage/cmd_parameter/detail_introduction_en.md

@@ -184,15 +184,6 @@
   - Specify shared dynamic library. It can be defined out of paddle by user.
   - type: string (default: "", null).
 
-## Metric Learning
-* `--external`
-   - Whether to use external machine for metric learning.
-   - type: bool (default: 0).
-
-* `--data_server_port`
-  - Listening port for dserver (data server), dserver is mainly used in metric learning.
-  - type: int32 (default: 21134).
-
 ## DataProvider
 
 * `--memory_threshold_on_load_data`

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/howto/usage/k8s/src/k8s_train/start_paddle.py

@@ -132,7 +132,8 @@ def startPaddle(idMap={}, train_args_dict=None):
     logDir = JOB_PATH_OUTPUT + "/node_" + str(trainerId)
     if not os.path.exists(JOB_PATH_OUTPUT):
         os.makedirs(JOB_PATH_OUTPUT)
-    os.mkdir(logDir)
+    if not os.path.exists(logDir):
+        os.mkdir(logDir)
     copyCommand = 'cp -rf ' + JOB_PATH + \
         "/" + str(trainerId) + "/data/*" + " ./data/"
     os.system(copyCommand)

doc/templates/conf.py.cn.in

@@ -15,13 +15,19 @@ import sys
 import os, subprocess
 import shlex
 from recommonmark import parser, transform
+try:
+   import py_paddle
+   import paddle
+   import paddle.v2
+except ImportError:
+   print("Must install paddle python package before generating documentation")
+   sys.exit(1)
 
 MarkdownParser = parser.CommonMarkParser
 AutoStructify = transform.AutoStructify
 # If extensions (or modules to document with autodoc) are in another directory,
 # 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_theme/templates"]
 
 # -- General configuration ------------------------------------------------

doc/templates/conf.py.en.in

@@ -15,14 +15,20 @@ import sys
 import os, subprocess
 import shlex
 from recommonmark import parser, transform
+try:
+   import py_paddle
+   import paddle
+   import paddle.v2
+except ImportError:
+   print("Must install paddle python package before generating documentation")
+   sys.exit(1)
+
 
 MarkdownParser = parser.CommonMarkParser
 AutoStructify = transform.AutoStructify
 # If extensions (or modules to document with autodoc) are in another directory,
 # 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_theme/templates"]
 
 # -- General configuration ------------------------------------------------

doc/tutorials/quick_start/index_en.md

@@ -156,14 +156,14 @@ define_py_data_sources2(train_list='data/train.list',
                         obj="process",
                         args={"dictionary": word_dict})
 ```
-You can refer to the following link for more detailed examples and data formats: <a href = "../../api/data_provider/pydataprovider2_en.html">PyDataProvider2</a>.
+You can refer to the following link for more detailed examples and data formats: <a href = "../../api/v1/data_provider/pydataprovider2_en.html">PyDataProvider2</a>.
 
 ## Network Architecture
 We will describe four kinds of network architectures in this section.
 <center> ![](./src/PipelineNetwork_en.jpg) </center>
 
 First, you will build a logistic regression model. Later, you will also get chance to build other more powerful network architectures.
-For more detailed documentation, you could refer to: <a href = "../../api/trainer_config_helpers/layers.html">layer documentation</a>. All configuration files are in `demo/quick_start` directory.
+For more detailed documentation, you could refer to: <a href = "../../api/v1/trainer_config_helpers/layers.html">layer documentation</a>. All configuration files are in `demo/quick_start` directory.
 
 ### Logistic Regression
 The architecture is illustrated in the following picture:
@@ -366,7 +366,7 @@ You can use single layer LSTM model with Dropout for our text classification pro
 <br>
 
 ## Optimization Algorithm
-<a href = "../../api/trainer_config_helpers/optimizers.html">Optimization algorithms</a> include Momentum, RMSProp, AdaDelta, AdaGrad, Adam, and Adamax. You can use Adam optimization method here, with L2 regularization and gradient clipping, because Adam has been proved to work very well for training recurrent neural network.
+<a href = "../../api/v1/trainer_config_helpers/optimizers.html">Optimization algorithms</a> include Momentum, RMSProp, AdaDelta, AdaGrad, Adam, and Adamax. You can use Adam optimization method here, with L2 regularization and gradient clipping, because Adam has been proved to work very well for training recurrent neural network.
 
 ```python
 settings(batch_size=128,
@@ -407,7 +407,7 @@ paddle train \
 --init_model_path=./output/pass-0000x
 ```
 
-We will give an example of performing prediction using Recurrent model on a dataset with no labels. You can refer to <a href = "../../api/predict/swig_py_paddle_en.html">Python Prediction API</a> tutorial，or other <a href = "../../tutorials/index_en.html">demo</a> for the prediction process using Python. You can also use the following script for inference or evaluation.
+We will give an example of performing prediction using Recurrent model on a dataset with no labels. You can refer to <a href = "../../api/v1/predict/swig_py_paddle_en.html">Python Prediction API</a> tutorial，or other <a href = "../../tutorials/index_en.html">demo</a> for the prediction process using Python. You can also use the following script for inference or evaluation.
 
 inference script (predict.sh)：
 

doc_theme/static/css/override.css

+* {
+    font-family:"Roboto","Lato","proxima-nova","Helvetica Neue",Arial,sans-serif;
+}
 body {
     padding-top: 80px;
     background-image: none !important;

paddle/api/Arguments.cpp

@@ -38,6 +38,13 @@ Arguments* Arguments::createByPaddleArgumentVector(void* ptr) {
   return args;
 }
 
+Arguments* Arguments::createByPaddleArgument(const void* ptr) {
+  auto p = (paddle::Argument*)(ptr);
+  auto args = new Arguments();
+  args->m->outputs.push_back(*p);
+  return args;
+}
+
 Matrix* Arguments::getSlotValue(size_t idx) const throw(RangeError) {
   auto& a = m->getArg(idx);
   return Matrix::createByPaddleMatrixPtr(&a.value);
@@ -137,9 +144,7 @@ void Arguments::setSlotSequenceDim(size_t idx, IVector* vec) throw(RangeError) {
   a.cpuSequenceDims = m->cast<paddle::IVector>(vec->getSharedPtr());
 }
 
-float Arguments::sumCosts() const {
-  return paddle::Argument::sumCosts(m->outputs);
-}
+float Arguments::sum() const { return paddle::Argument::sum(m->outputs); }
 
 int64_t Arguments::getBatchSize(size_t idx) const throw(RangeError) {
   auto& a = m->getArg(idx);

paddle/api/Evaluator.cpp

@@ -27,3 +27,18 @@ std::string Evaluator::toString() {
   m->rawPtr->printStats(sout);
   return sout.str();
 }
+
+std::vector<std::string> Evaluator::getNames() const {
+  std::vector<std::string> retv;
+  m->rawPtr->getNames(&retv);
+  return retv;
+}
+
+double Evaluator::getValue(const std::string name) const {
+  paddle::Error err;
+  double v = m->rawPtr->getValue(name, &err);
+  if (err) {
+    throw std::runtime_error(err.msg());
+  }
+  return v;
+}

paddle/api/GradientMachine.cpp

@@ -142,14 +142,28 @@ Parameter* GradientMachine::getParameter(size_t i) throw(RangeError) {
   }
 }
 
+size_t GradientMachine::getNonStaticParameterSize() const {
+  return m->machine->getNonStaticParameters().size();
+}
+
+Parameter* GradientMachine::getNonStaticParameter(size_t i) throw(RangeError) {
+  auto params = m->machine->getNonStaticParameters();
+  if (i < params.size()) {
+    return Parameter::createFromSharedPtr(
+        &m->machine->getNonStaticParameters()[i]);
+  } else {
+    throw RangeError();
+  }
+}
+
 void GradientMachine::randParameters() { m->machine->randParameters(); }
 
-Matrix* GradientMachine::getLayerOutput(const std::string& layerName) const
+Arguments* GradientMachine::getLayerOutput(const std::string& layerName) const
     throw(UnsupportError) {
-  auto nn = std::dynamic_pointer_cast<paddle::NeuralNetwork>(m->machine);
+  auto nn = m->machine;
   if (nn) {
-    auto mat = nn->getLayerOutput(layerName);
-    return Matrix::createByPaddleMatrixPtr(&mat);
+    auto arg = nn->getLayerOutput(layerName);
+    return Arguments::createByPaddleArgument(&arg);
   } else {
     throw UnsupportError();
   }

paddle/api/PaddleAPI.h

@@ -47,6 +47,9 @@ void setUseGpu(bool useGpu);
 /// Return true if this py_paddle is compiled in GPU Version
 bool isGpuVersion();
 
+/// Return FLAGS_trainer_count
+int getTrainerCount();
+
 /// The Error of IO Operation. Such as file not found, etc.
 class IOError {};
 
@@ -450,10 +453,11 @@ public:
                                         IVector* vec) throw(RangeError);
   void setSlotSequenceDim(size_t idx, IVector* vec) throw(RangeError);
 
-  float sumCosts() const;
+  float sum() const;
 
 private:
   static Arguments* createByPaddleArgumentVector(void* ptr);
+  static Arguments* createByPaddleArgument(const void* ptr);
   void* getInternalArgumentsPtr() const;
 
 private:
@@ -767,9 +771,12 @@ public:
   size_t getParameterSize() const;
   Parameter* getParameter(size_t i) throw(RangeError);
 
+  size_t getNonStaticParameterSize() const;
+  Parameter* getNonStaticParameter(size_t i) throw(RangeError);
+
   void randParameters();
 
-  Matrix* getLayerOutput(const std::string& layerName) const
+  Arguments* getLayerOutput(const std::string& layerName) const
       throw(UnsupportError);
 
   /**
@@ -900,6 +907,10 @@ public:
    */
   std::string toString();
 
+  std::vector<std::string> getNames() const;
+
+  double getValue(const std::string name) const;
+
 private:
   EvaluatorPrivate* m;
 
@@ -952,7 +963,7 @@ public:
 
   Arguments* getForwardOutput();
 
-  Matrix* getLayerOutput(const std::string& layerName);
+  Arguments* getLayerOutput(const std::string& layerName) const;
 };
 
 /// the N-Best results generated from one input sequence.

paddle/api/Trainer.cpp

@@ -131,12 +131,11 @@ void Trainer::testOneDataBatch(size_t batchSize, const Arguments& args) {
 void TrainerPrivate::finishTestPeriod() { tester_->finishTestPeriod(); }
 void Trainer::finishTestPeriod() { m->finishTestPeriod(); }
 
-Matrix* Trainer::getLayerOutput(const std::string& layerName) {
-  auto nn = std::dynamic_pointer_cast<paddle::NeuralNetwork>(
-      this->m->getGradientMachine());
+Arguments* Trainer::getLayerOutput(const std::string& layerName) const {
+  auto nn = this->m->getGradientMachine();
   CHECK(nn) << "trainerInternal_.getGradientMachine() is not NeuralNetwork";
-  auto m = nn->getLayerOutput(layerName);
-  return Matrix::createByPaddleMatrixPtr(&m);
+  auto arg = nn->getLayerOutput(layerName);
+  return Arguments::createByPaddleArgument(&arg);
 }
 
 void Trainer::forwardOneBatch(size_t batchSize) {

paddle/api/Util.cpp

@@ -54,5 +54,7 @@ bool isGpuVersion() {
 #endif
 }
 
+int getTrainerCount() { return FLAGS_trainer_count; }
+
 static_assert(NUM_PARAMETER_TYPES == paddle::NUM_PARAMETER_TYPES,
               "The Parameter Type should be same in core/api and core/common");

paddle/api/test/testArguments.py

@@ -22,7 +22,7 @@ class TestArguments(unittest.TestCase):
         args = swig_paddle.Arguments.createArguments(1)
         args.setSlotValue(0, m)
 
-        self.assertAlmostEqual(27.0, args.sumCosts())
+        self.assertAlmostEqual(27.0, args.sum())
 
         mat = args.getSlotValue(0)
         assert isinstance(mat, swig_paddle.Matrix)

paddle/api/test/testMatrix.py

@@ -68,7 +68,7 @@ class TestMatrix(unittest.TestCase):
 
     def test_numpyCpu(self):
         numpy_mat = np.matrix([[1, 2], [3, 4], [5, 6]], dtype="float32")
-        m = swig_paddle.Matrix.createCpuDenseFromNumpy(numpy_mat, copy=False)
+        m = swig_paddle.Matrix.createCpuDenseFromNumpy(numpy_mat, False)
         self.assertEqual((int(m.getHeight()), int(m.getWidth())),
                          numpy_mat.shape)
 

paddle/api/test/testTrain.py

@@ -89,9 +89,14 @@ def main():
             except Exception as e:
                 print e
 
+        ev = m.makeEvaluator()
+        ev.start()
         m.forwardBackward(inArgs, outArgs, swig_paddle.PASS_TRAIN,
                           update_callback)
-
+        m.eval(ev)
+        ev.finish()
+        for name in ev.getNames():
+            print name, ev.getValue(name)
         for optimizer in optimizers:
             optimizer.finishBatch()
 

paddle/api/test/testVector.py

@@ -43,7 +43,7 @@ class TestIVector(unittest.TestCase):
 
     def test_cpu_numpy(self):
         vec = np.array([1, 3, 4, 65, 78, 1, 4], dtype="int32")
-        iv = swig_paddle.IVector.createCpuVectorFromNumpy(vec, copy=False)
+        iv = swig_paddle.IVector.createCpuVectorFromNumpy(vec, False)
         self.assertEqual(vec.shape[0], int(iv.__len__()))
         vec[4] = 832
         for i in xrange(len(iv)):
@@ -106,7 +106,7 @@ class TestVector(unittest.TestCase):
 
     def testCpuNumpy(self):
         numpy_arr = np.array([1.2, 2.3, 3.4, 4.5], dtype="float32")
-        vec = swig_paddle.Vector.createCpuVectorFromNumpy(numpy_arr, copy=False)
+        vec = swig_paddle.Vector.createCpuVectorFromNumpy(numpy_arr, False)
         assert isinstance(vec, swig_paddle.Vector)
         numpy_arr[0] = 0.1
         for n, v in zip(numpy_arr, vec):

paddle/cuda/include/hl_matrix.h

@@ -69,19 +69,6 @@ extern void hl_sequence_softmax_forward(real* A_d,
                                         const int* index,
                                         int numSequence);
 
-/**
- * @brief   Matrix classification error.
- *
- * @param[in]   A_d     input matrix (M x N).
- * @param[in]   B_d     input vector (M x 1).
- * @param[out]  C_d     output vector (M x 1).
- * @param[in]   dimM    matrix height.
- * @param[in]   dimN    matrix width.
- *
- */
-extern void hl_matrix_classification_error(
-    real* A_d, int* B_d, real* C_d, int dimM, int dimN);
-
 /**
  * @brief   Matrix cross entropy.
  *

paddle/cuda/include/hl_top_k.h

@@ -58,4 +58,30 @@ extern void hl_sparse_matrix_top_k(real* topVal,
                                    int beamSize,
                                    int numSamples);
 
-#endif /* HL_TOP_K_H_ */
+/**
+ * @brief   Matrix classification error.
+ *
+ * @param[out]  topVal         top k element.
+ * @param[in]   ldv            leading dimension of topVal.
+ * @param[out]  topIds         top k index.
+ * @param[in]   src            input value.
+ * @param[in]   lds            leading dimension of src.
+ * @param[in]   dim            width of input value.
+ * @param[in]   topkSize       size of top k element.
+ * @param[in]   numSamples     height of input value.
+ * @param[in]   label          ground truth label.
+ * @param[out]  recResult      top-k classification error.
+ *
+ */
+extern void hl_matrix_classification_error(real* topVal,
+                                           int ldv,
+                                           int* topIds,
+                                           real* src,
+                                           int lds,
+                                           int dim,
+                                           int topkSize,
+                                           int numSamples,
+                                           int* label,
+                                           real* recResult);
+
+#endif  // HL_TOP_K_H_

paddle/cuda/include/stub/hl_matrix_stub.h

@@ -35,8 +35,16 @@ inline void hl_sequence_softmax_forward(real* A_d,
 inline void hl_matrix_softmax_derivative(
     real* grad_d, real* output_d, real* sftmaxSum_d, int dimM, int dimN) {}
 
-inline void hl_matrix_classification_error(
-    real* A_d, int* B_d, real* C_d, int dimM, int dimN) {}
+inline void hl_matrix_classification_error(real* topVal,
+                                           int ldv,
+                                           int* topIds,
+                                           real* src,
+                                           int lds,
+                                           int dim,
+                                           int topkSize,
+                                           int numSamples,
+                                           int* label,
+                                           real* recResult) {}
 
 inline void hl_matrix_cross_entropy(
     real* A_d, real* C_d, int* label_d, int dimM, int dimN) {}