Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into infer_prog

5ecbba46 · wanghaoshuang · 026e88a3 · 750299f5 · 5ecbba46 · 5ecbba46
580 changed file
--- a/.copyright.hook
+++ b/.copyright.hook
@@ -49,10 +49,15 @@ def generate_copyright(template, lang='C'):
        LANG_COMMENT_MARK = "//"

    lines = template.split(NEW_LINE_MARK)
-    ans = LANG_COMMENT_MARK + " " + COPYRIGHT_HEADER + NEW_LINE_MARK
+    BLANK = " "
+    ans = LANG_COMMENT_MARK + BLANK + COPYRIGHT_HEADER + NEW_LINE_MARK
    for lino, line in enumerate(lines):
        if lino == 0 or lino == 1 or lino == len(lines) - 1: continue
-        ans += LANG_COMMENT_MARK + " " + line + NEW_LINE_MARK
+        if len(line)  == 0:
+            BLANK = ""
+        else:
+            BLANK = " "
+        ans += LANG_COMMENT_MARK + BLANK + line + NEW_LINE_MARK

    return ans + "\n"

@@ -62,6 +67,8 @@ def lang_type(filename):
        return "Python"
    elif filename.endswith(".h"):
        return "C"
+    elif filename.endswith(".c"):
+        return "C"
    elif filename.endswith(".hpp"):
        return "C"
    elif filename.endswith(".cc"):
@@ -77,10 +84,13 @@ def lang_type(filename):
    elif filename.endswith(".proto"):
        return "C"
    else:
-        print("Unsupported filetype")
+        print("Unsupported filetype %s", filename)
        exit(0)


+PYTHON_ENCODE = re.compile("^[ \t\v]*#.*?coding[:=][ \t]*([-_.a-zA-Z0-9]+)")
+
+
 def main(argv=None):
    parser = argparse.ArgumentParser(
        description='Checker for copyright declaration.')
@@ -89,9 +99,14 @@ def main(argv=None):

    retv = 0
    for filename in args.filenames:
-        first_line = io.open(filename).readline()
-        if "COPYRIGHT" in first_line.upper() : continue
-        original_contents = io.open(filename).read()
+        fd = io.open(filename, encoding="utf-8")
+        first_line = fd.readline()
+        second_line = fd.readline()
+        if "COPYRIGHT (C)" in first_line.upper(): continue
+        if first_line.startswith("#!") or PYTHON_ENCODE.match(
+                second_line) != None or PYTHON_ENCODE.match(first_line) != None:
+            continue
+        original_contents = io.open(filename, encoding="utf-8").read()
        new_contents = generate_copyright(
            COPYRIGHT, lang_type(filename)) + original_contents
        print('Auto Insert Copyright Header {}'.format(filename))

--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -31,9 +31,6 @@ if(NOT CMAKE_CROSSCOMPILING)
 endif(NOT CMAKE_CROSSCOMPILING)
 find_package(Git REQUIRED)
 find_package(Threads REQUIRED)
-if(NOT ANDROID AND NOT IOS)
-    find_package(Boost QUIET)
-endif()

 include(simd)

@@ -55,6 +52,8 @@ option(WITH_COVERAGE    "Compile PaddlePaddle with code coverage"       OFF)
 option(COVERALLS_UPLOAD "Package code coverage data to coveralls"       OFF)
 option(ON_TRAVIS        "Exclude special unit test on Travis CI"        OFF)
 option(WITH_C_API       "Compile PaddlePaddle with C-API(Prediction)"   OFF)
+# TODO: Only compile PaddlePaddle fluid version by WITH_FLUID option. 
+option(WITH_FLUID       "Compile PaddlePaddle fluid only(TODO)"         ON)
 option(WITH_GOLANG      "Compile PaddlePaddle with GOLANG"              OFF)
 option(GLIDE_INSTALL    "Download and install go dependencies "         ON)
 option(USE_NNPACK       "Compile PaddlePaddle with NNPACK library"      OFF)
@@ -107,6 +106,10 @@ if (WITH_C_API AND WITH_PYTHON)
    "different Python interpreter from compiling.")
 endif()

+if (WITH_C_API)
+  set(WITH_FLUID OFF CACHE STRING "Disable install fluid when compile the C_API" FORCE)
+endif()
+
 if(MOBILE_INFERENCE)
    set(THIRD_PARTY_BUILD_TYPE MinSizeRel)
 else()
@@ -134,6 +137,7 @@ include(external/openblas)  # download, build, install openblas
 include(external/mkldnn)    # download, build, install mkldnn
 include(external/swig)      # download, build, install swig
 include(external/warpctc)   # download, build, install warpctc
+include(external/boost)     # download, build, install boost
 include(external/any)       # download libn::any
 include(external/eigen)     # download eigen3
 include(external/pybind11)  # download pybind11
@@ -158,7 +162,6 @@ include_directories("${PADDLE_SOURCE_DIR}")
 include_directories("${PADDLE_SOURCE_DIR}/paddle/cuda/include")
 include_directories("${CMAKE_CURRENT_BINARY_DIR}/proto")
 include_directories("${CMAKE_CURRENT_BINARY_DIR}/go/pserver/client/c")
-include_directories(${Boost_INCLUDE_DIRS})

 set(EXTERNAL_LIBS
    ${GFLAGS_LIBRARIES}

--- a/CODE_OF_CONDUCT_cn.md
+++ b/CODE_OF_CONDUCT_cn.md
-# 貢獻者公約
+# 参与者公约

-## 我們的承諾
+## 我们的保证

-為了促進一個開放透明且受歡迎的環境，我們作為貢獻者和維護者保證，無論年齡、種族、民族、性別認同和表達、體型、殘疾、經驗水平、國籍、個人表現、宗教或性別取向，在我們的專案以及社群的參與者都有不被騷擾的體驗。
+为了促进一个开放透明且友好的环境，我们作为贡献者和维护者保证：无论年龄、种族、民族、性别认同和表达（方式）、体型、身体健全与否、经验水平、国籍、个人表现、宗教或性别取向，参与者在我们项目和社区中都免于骚扰。

-## 我們的準則
+## 我们的标准

-舉例來說有助於創造正面環境的行為包括：
-* 使用歡迎和包容性語言
-* 尊重不同的觀點和經驗
-* 優雅地接受建設性批評
-* 關注在對於社群最好的事情上
-* 對其他社群成員的表現友善
+有助于创造正面环境的行为包括但不限于：
+* 使用友好和包容性语言
+* 尊重不同的观点和经历
+* 耐心地接受建设性批评
+* 关注对社区最有利的事情
+* 友善对待其他社区成员

-舉例來說身為參與者不能接受的行為包括：
-* 使用與性有關的言語或是圖像，以及不受歡迎的性騷擾
-* 酸民/反串/釣魚行為或進行侮辱/貶損的評論，人身攻擊及政治攻擊
-* 公開或私下的騷擾
-* 未經許可地發布他人的個人資料，例如住址或是電子地址
-* 其他可以被合理地認定為不恰當或者違反職業操守的行為
+身为参与者不能接受的行为包括但不限于：
+* 使用与性有关的言语或是图像，以及不受欢迎的性骚扰
+* 捣乱/煽动/造谣的行为或进行侮辱/贬损的评论，人身攻击及政治攻击
+* 公开或私下的骚扰
+* 未经许可地发布他人的个人资料，例如住址或是电子地址
+* 其他可以被合理地认定为不恰当或者违反职业操守的行为

-## 我們的責任
+## 我们的责任

-專案維護者有責任為"可接受的行為"準則做出詮釋，以及對已發生的不被接受的行為採取恰當且公平的糾正措施。
+项目维护者有责任为「可接受的行为」标准做出诠释，以及对已发生的不被接受的行为采取恰当且公平的纠正措施。

-專案維護者有權力及責任去刪除、編輯、拒絕與本行為準則有所違背的評論(comments)、提交(commits)、程式碼、wiki 編輯、問題(issues)和其他貢獻，以及專案維護者可暫時或永久性的禁止任何他們認為有不適當、威脅、冒犯、有害行為的貢獻者。
+项目维护者有权利及责任去删除、编辑、拒绝与本行为标准有所违背的评论(comments)、提交(commits)、代码、wiki 编辑、问题(issues)和其他贡献，以及项目维护者可暂时或永久性的禁止任何他们认为有不适当、威胁、冒犯、有害行为的贡献者。

-## 使用範圍
+## 使用范围

-當一個人代表該專案或是其社群時，本行為準則適用於其專案平台和公共平台。
+当一个人代表该项目或是其社区时，本行为标准适用于其项目平台和公共平台。

-代表專案或是社群的情況，舉例來說包括使用官方專案的電子郵件地址、通過官方的社群媒體帳號發布或線上或線下事件中擔任指定代表。
+代表项目或是社区的情况，举例来说包括使用官方项目的电子邮件地址、通过官方的社区媒体账号发布或线上或线下事件中担任指定代表。

-該專案的呈現方式可由其專案維護者進行進一步的定義及解釋。
+该项目的呈现方式可由其项目维护者进行进一步的定义及解释。

-## 強制執行
+## 强制执行

-可以透過paddle-dev@baidu.com，來聯繫專案團隊來報告濫用、騷擾或其他不被接受的行為。
+可以通过paddle-dev@baidu.com，来联系项目团队来举报滥用、骚扰或其他不被接受的行为。

-任何維護團隊認為有必要且適合的所有投訴都將進行審查及調查，並做出相對應的回應。專案小組有對事件回報者有保密的義務。具體執行的方針近一步細節可能會單獨公佈。
+任何维护团队认为有必要且适合的所有投诉都将进行审查及调查，并做出相对应的回应。项目小组有对事件回报者有保密的义务。具体执行的方针近一步细节可能会单独公布。

-沒有真誠的遵守或是執行本行為準則的專案維護人員，可能會因專案領導人或是其他成員的決定，暫時或是永久的取消其身份。
+没有切实地遵守或是执行本行为标准的项目维护人员，可能会因项目领导人或是其他成员的决定，暂时或是永久地取消其参与资格。

-## 來源
+## 来源

-本行為準則改編自[貢獻者公約][首頁]，版本 1.4
-可在此觀看https://www.contributor-covenant.org/zh-tw/version/1/4/code-of-conduct.html
+本行为标准改编自[贡献者公约][主页]，版本 1.4
+可在此观看https://www.contributor-covenant.org/zh-cn/version/1/4/code-of-conduct.html

-[首頁]: https://www.contributor-covenant.org
+[主页]: https://www.contributor-covenant.org
--- a/Dockerfile
+++ b/Dockerfile
@@ -27,7 +27,7 @@ RUN apt-get update && \
    curl sed grep graphviz libjpeg-dev zlib1g-dev  \
    python-matplotlib gcc-4.8 g++-4.8 \
    automake locales clang-format swig doxygen cmake  \
-    liblapack-dev liblapacke-dev libboost-dev \
+    liblapack-dev liblapacke-dev \
    clang-3.8 llvm-3.8 libclang-3.8-dev \
    net-tools libtool && \
    apt-get clean -y

--- a/adversarial/README.md
+++ b/adversarial/README.md
-# Advbox
-
-Advbox is a Python toolbox to create adversarial examples that fool neural networks. It requires Python and paddle.
-
-## How to use
-
-1. train a model and save it's parameters. (like fluid_mnist.py)
-2. load the parameters which is trained in step1, then reconstruct the model.(like mnist_tutorial_fgsm.py)
-3. use advbox to generate the adversarial sample.
--- a/adversarial/advbox/__init__.py
+++ b/adversarial/advbox/__init__.py
-# Copyright (c) 2017 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.
-"""
-   A set of tools for generating adversarial example on paddle platform 
-"""
--- a/adversarial/advbox/attacks/base.py
+++ b/adversarial/advbox/attacks/base.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
-"""
-The base model of the model.
-"""
-from abc import ABCMeta, abstractmethod
-
-
-class Attack(object):
-    """
-    Abstract base class for adversarial attacks. `Attack` represent an adversarial attack
-    which search an adversarial example. subclass should implement the _apply() method.
-
-    Args:
-        model(Model): an instance of the class advbox.base.Model.
-
-    """
-    __metaclass__ = ABCMeta
-
-    def __init__(self, model):
-        self.model = model
-
-    def __call__(self, image_label):
-        """
-        Generate the adversarial sample.
-
-        Args:
-        image_label(list): The image and label tuple list with one element.
-        """
-        adv_img = self._apply(image_label)
-        return adv_img
-
-    @abstractmethod
-    def _apply(self, image_label):
-        """
-        Search an adversarial example.
-
-        Args:
-        image_batch(list): The image and label tuple list with one element.
-        """
-        raise NotImplementedError
--- a/adversarial/advbox/attacks/gradientsign.py
+++ b/adversarial/advbox/attacks/gradientsign.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
-"""
-This module provide the attack method for FGSM's implement.
-"""
-from __future__ import division
-import numpy as np
-from collections import Iterable
-from .base import Attack
-
-
-class GradientSignAttack(Attack):
-    """
-    This attack was originally implemented by Goodfellow et al. (2015) with the
-    infinity norm (and is known as the "Fast Gradient Sign Method"). This is therefore called
-    the Fast Gradient Method.
-    Paper link: https://arxiv.org/abs/1412.6572
-    """
-
-    def _apply(self, image_label, epsilons=1000):
-        assert len(image_label) == 1
-        pre_label = np.argmax(self.model.predict(image_label))
-
-        min_, max_ = self.model.bounds()
-        gradient = self.model.gradient(image_label)
-        gradient_sign = np.sign(gradient) * (max_ - min_)
-
-        if not isinstance(epsilons, Iterable):
-            epsilons = np.linspace(0, 1, num=epsilons + 1)
-
-        for epsilon in epsilons:
-            adv_img = image_label[0][0].reshape(
-                gradient_sign.shape) + epsilon * gradient_sign
-            adv_img = np.clip(adv_img, min_, max_)
-            adv_label = np.argmax(self.model.predict([(adv_img, 0)]))
-            if pre_label != adv_label:
-                return adv_img
-
-
-FGSM = GradientSignAttack
-
-
-class IteratorGradientSignAttack(Attack):
-    """
-    This attack was originally implemented by Alexey Kurakin(Google Brain).
-    Paper link: https://arxiv.org/pdf/1607.02533.pdf
-    """
-
-    def _apply(self, image_label, epsilons=100, steps=10):
-        """
-        Apply the iterative gradient sign attack.
-        Args:
-            image_label(list): The image and label tuple list of one element.
-            epsilons(list|tuple|int): The epsilon (input variation parameter).
-            steps(int): The number of iterator steps.
-        Return:
-            numpy.ndarray: The adversarail sample generated by the algorithm.
-        """
-        assert len(image_label) == 1
-        pre_label = np.argmax(self.model.predict(image_label))
-        gradient = self.model.gradient(image_label)
-        min_, max_ = self.model.bounds()
-
-        if not isinstance(epsilons, Iterable):
-            epsilons = np.linspace(0, 1, num=epsilons + 1)
-
-        for epsilon in epsilons:
-            adv_img = image_label[0][0].reshape(gradient.shape)
-            for _ in range(steps):
-                gradient = self.model.gradient([(adv_img, image_label[0][1])])
-                gradient_sign = np.sign(gradient) * (max_ - min_)
-                adv_img = adv_img + epsilon * gradient_sign
-                adv_img = np.clip(adv_img, min_, max_)
-                adv_label = np.argmax(self.model.predict([(adv_img, 0)]))
-                if pre_label != adv_label:
-                    return adv_img
--- a/adversarial/advbox/models/__init__.py
+++ b/adversarial/advbox/models/__init__.py
-# Copyright (c) 2017 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.
-"""
-Paddle model for target of attack 
-"""
--- a/adversarial/advbox/models/base.py
+++ b/adversarial/advbox/models/base.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
-"""
-The base model of the model.
-"""
-from abc import ABCMeta
-import abc
-
-abstractmethod = abc.abstractmethod
-
-
-class Model(object):
-    """
-    Base class of model to provide attack.
-
-
-    Args:
-        bounds(tuple): The lower and upper bound for the image pixel.
-        channel_axis(int): The index of the axis that represents the color channel.
-        preprocess(tuple): Two element tuple used to preprocess the input. First
-            substract the first element, then divide the second element.
-    """
-    __metaclass__ = ABCMeta
-
-    def __init__(self, bounds, channel_axis, preprocess=None):
-        assert len(bounds) == 2
-        assert channel_axis in [0, 1, 2, 3]
-
-        if preprocess is None:
-            preprocess = (0, 1)
-        self._bounds = bounds
-        self._channel_axis = channel_axis
-        self._preprocess = preprocess
-
-    def bounds(self):
-        """
-        Return the upper and lower bounds of the model.
-        """
-        return self._bounds
-
-    def channel_axis(self):
-        """
-        Return the channel axis of the model.
-        """
-        return self._channel_axis
-
-    def _process_input(self, input_):
-        res = input_
-        sub, div = self._preprocess
-        if sub != 0:
-            res = input_ - sub
-        assert div != 0
-        if div != 1:
-            res /= div
-        return res
-
-    @abstractmethod
-    def predict(self, image_batch):
-        """
-        Calculate the prediction of the image batch.
-
-        Args:
-            image_batch(numpy.ndarray): image batch of shape (batch_size, height, width, channels).
-
-        Return:
-            numpy.ndarray: predictions of the images with shape (batch_size, num_of_classes).
-        """
-        raise NotImplementedError
-
-    @abstractmethod
-    def num_classes(self):
-        """
-        Determine the number of the classes
-
-        Return:
-            int: the number of the classes
-        """
-        raise NotImplementedError
-
-    @abstractmethod
-    def gradient(self, image_batch):
-        """
-        Calculate the gradient of the cross-entropy loss w.r.t the image.
-
-        Args:
-            image_batch(list): The image and label tuple list.
-
-        Return:
-            numpy.ndarray: gradient of the cross-entropy loss w.r.t the image with
-                the shape (height, width, channel).
-        """
-        raise NotImplementedError
--- a/adversarial/advbox/models/paddle.py
+++ b/adversarial/advbox/models/paddle.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
-from __future__ import absolute_import
-
-import numpy as np
-import paddle.v2 as paddle
-import paddle.v2.fluid as fluid
-from paddle.v2.fluid.framework import program_guard
-
-from .base import Model
-
-
-class PaddleModel(Model):
-    """
-    Create a PaddleModel instance.
-    When you need to generate a adversarial sample, you should construct an instance of PaddleModel.
-
-    Args:
-        program(paddle.v2.fluid.framework.Program): The program of the model which generate the adversarial sample.
-        input_name(string): The name of the input.
-        logits_name(string): The name of the logits.
-        predict_name(string): The name of the predict.
-        cost_name(string): The name of the loss in the program.
-    """
-
-    def __init__(self,
-                 program,
-                 input_name,
-                 logits_name,
-                 predict_name,
-                 cost_name,
-                 bounds,
-                 channel_axis=3,
-                 preprocess=None):
-        super(PaddleModel, self).__init__(
-            bounds=bounds, channel_axis=channel_axis, preprocess=preprocess)
-
-        if preprocess is None:
-            preprocess = (0, 1)
-
-        self._program = program
-        self._place = fluid.CPUPlace()
-        self._exe = fluid.Executor(self._place)
-
-        self._input_name = input_name
-        self._logits_name = logits_name
-        self._predict_name = predict_name
-        self._cost_name = cost_name
-
-        # gradient
-        loss = self._program.block(0).var(self._cost_name)
-        param_grads = fluid.backward.append_backward(
-            loss, parameter_list=[self._input_name])
-        self._gradient = dict(param_grads)[self._input_name]
-
-    def predict(self, image_batch):
-        """
-            Predict the label of the image_batch.
-
-            Args:
-                image_batch(list): The image and label tuple list.
-            Return:
-                numpy.ndarray: predictions of the images with shape (batch_size, num_of_classes).
-        """
-        feeder = fluid.DataFeeder(
-            feed_list=[self._input_name, self._logits_name],
-            place=self._place,
-            program=self._program)
-        predict_var = self._program.block(0).var(self._predict_name)
-        predict = self._exe.run(self._program,
-                                feed=feeder.feed(image_batch),
-                                fetch_list=[predict_var])
-        return predict
-
-    def num_classes(self):
-        """
-            Calculate the number of classes of the output label. 
-
-        Return:
-            int: the number of classes
-        """
-        predict_var = self._program.block(0).var(self._predict_name)
-        assert len(predict_var.shape) == 2
-        return predict_var.shape[1]
-
-    def gradient(self, image_batch):
-        """
-        Calculate the gradient of the loss w.r.t the input.
-
-        Args:
-            image_batch(list): The image and label tuple list.
-        Return:
-            list: The list of the gradient of the image.
-        """
-        feeder = fluid.DataFeeder(
-            feed_list=[self._input_name, self._logits_name],
-            place=self._place,
-            program=self._program)
-
-        grad, = self._exe.run(self._program,
-                              feed=feeder.feed(image_batch),
-                              fetch_list=[self._gradient])
-        return grad
--- a/adversarial/fluid_mnist.py
+++ b/adversarial/fluid_mnist.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
-"""
-CNN on mnist data using fluid api of paddlepaddle
-"""
-import paddle.v2 as paddle
-import paddle.v2.fluid as fluid
-
-
-def mnist_cnn_model(img):
-    """
-    Mnist cnn model
-
-    Args:
-        img(Varaible): the input image to be recognized
-
-    Returns:
-        Variable: the label prediction
-    """
-    conv_pool_1 = fluid.nets.simple_img_conv_pool(
-        input=img,
-        num_filters=20,
-        filter_size=5,
-        pool_size=2,
-        pool_stride=2,
-        act='relu')
-
-    conv_pool_2 = fluid.nets.simple_img_conv_pool(
-        input=conv_pool_1,
-        num_filters=50,
-        filter_size=5,
-        pool_size=2,
-        pool_stride=2,
-        act='relu')
-
-    logits = fluid.layers.fc(input=conv_pool_2, size=10, act='softmax')
-    return logits
-
-
-def main():
-    """
-    Train the cnn model on mnist datasets
-    """
-    img = fluid.layers.data(name='img', shape=[1, 28, 28], dtype='float32')
-    label = fluid.layers.data(name='label', shape=[1], dtype='int64')
-    logits = mnist_cnn_model(img)
-    cost = fluid.layers.cross_entropy(input=logits, label=label)
-    avg_cost = fluid.layers.mean(x=cost)
-    optimizer = fluid.optimizer.Adam(learning_rate=0.01)
-    optimizer.minimize(avg_cost)
-
-    accuracy = fluid.evaluator.Accuracy(input=logits, label=label)
-
-    BATCH_SIZE = 50
-    PASS_NUM = 3
-    ACC_THRESHOLD = 0.98
-    LOSS_THRESHOLD = 10.0
-    train_reader = paddle.batch(
-        paddle.reader.shuffle(
-            paddle.dataset.mnist.train(), buf_size=500),
-        batch_size=BATCH_SIZE)
-
-    place = fluid.CPUPlace()
-    exe = fluid.Executor(place)
-    feeder = fluid.DataFeeder(feed_list=[img, label], place=place)
-    exe.run(fluid.default_startup_program())
-
-    for pass_id in range(PASS_NUM):
-        accuracy.reset(exe)
-        for data in train_reader():
-            loss, acc = exe.run(fluid.default_main_program(),
-                                feed=feeder.feed(data),
-                                fetch_list=[avg_cost] + accuracy.metrics)
-            pass_acc = accuracy.eval(exe)
-            print("pass_id=" + str(pass_id) + " acc=" + str(acc) + " pass_acc="
-                  + str(pass_acc))
-            if loss < LOSS_THRESHOLD and pass_acc > ACC_THRESHOLD:
-                break
-
-        pass_acc = accuracy.eval(exe)
-        print("pass_id=" + str(pass_id) + " pass_acc=" + str(pass_acc))
-    fluid.io.save_params(
-        exe, dirname='./mnist', main_program=fluid.default_main_program())
-    print('train mnist done')
-
-
-if __name__ == '__main__':
-    main()
--- a/adversarial/mnist_tutorial_fgsm.py
+++ b/adversarial/mnist_tutorial_fgsm.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
-"""
-FGSM demos on mnist using advbox tool.
-"""
-import paddle.v2 as paddle
-import paddle.v2.fluid as fluid
-import matplotlib.pyplot as plt
-import numpy as np
-
-from advbox.models.paddle import PaddleModel
-from advbox.attacks.gradientsign import GradientSignAttack
-
-
-def cnn_model(img):
-    """
-    Mnist cnn model
-    Args:
-        img(Varaible): the input image to be recognized
-    Returns:
-        Variable: the label prediction
-    """
-    #conv1 = fluid.nets.conv2d()
-    conv_pool_1 = fluid.nets.simple_img_conv_pool(
-        input=img,
-        num_filters=20,
-        filter_size=5,
-        pool_size=2,
-        pool_stride=2,
-        act='relu')
-
-    conv_pool_2 = fluid.nets.simple_img_conv_pool(
-        input=conv_pool_1,
-        num_filters=50,
-        filter_size=5,
-        pool_size=2,
-        pool_stride=2,
-        act='relu')
-
-    logits = fluid.layers.fc(input=conv_pool_2, size=10, act='softmax')
-    return logits
-
-
-def main():
-    """
-    Advbox demo which demonstrate how to use advbox.
-    """
-    IMG_NAME = 'img'
-    LABEL_NAME = 'label'
-
-    img = fluid.layers.data(name=IMG_NAME, shape=[1, 28, 28], dtype='float32')
-    # gradient should flow
-    img.stop_gradient = False
-    label = fluid.layers.data(name=LABEL_NAME, shape=[1], dtype='int64')
-    logits = cnn_model(img)
-    cost = fluid.layers.cross_entropy(input=logits, label=label)
-    avg_cost = fluid.layers.mean(x=cost)
-
-    place = fluid.CPUPlace()
-    exe = fluid.Executor(place)
-
-    BATCH_SIZE = 1
-    train_reader = paddle.batch(
-        paddle.reader.shuffle(
-            paddle.dataset.mnist.train(), buf_size=500),
-        batch_size=BATCH_SIZE)
-    feeder = fluid.DataFeeder(
-        feed_list=[IMG_NAME, LABEL_NAME],
-        place=place,
-        program=fluid.default_main_program())
-
-    fluid.io.load_params(
-        exe, "./mnist/", main_program=fluid.default_main_program())
-
-    # advbox demo
-    m = PaddleModel(fluid.default_main_program(), IMG_NAME, LABEL_NAME,
-                    logits.name, avg_cost.name, (-1, 1))
-    att = GradientSignAttack(m)
-    for data in train_reader():
-        # fgsm attack
-        adv_img = att(data)
-        plt.imshow(n[0][0], cmap='Greys_r')
-        plt.show()
-        #np.save('adv_img', adv_img)
-        break
-
-
-if __name__ == '__main__':
-    main()
--- a/benchmark/paddle/image/alexnet.py
+++ b/benchmark/paddle/image/alexnet.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
-#!/usr/bin/env python
+# 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.

 from paddle.trainer_config_helpers import *


--- a/benchmark/paddle/image/googlenet.py
+++ b/benchmark/paddle/image/googlenet.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
 #!/usr/bin/env python
 from paddle.trainer_config_helpers import *


--- a/benchmark/paddle/image/provider.py
+++ b/benchmark/paddle/image/provider.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 import io, os
 import random
 import numpy as np

--- a/benchmark/paddle/image/resnet.py
+++ b/benchmark/paddle/image/resnet.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
 #!/usr/bin/env python
 from paddle.trainer_config_helpers import *


--- a/benchmark/paddle/image/smallnet_mnist_cifar.py
+++ b/benchmark/paddle/image/smallnet_mnist_cifar.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
 #!/usr/bin/env python

 from paddle.trainer_config_helpers import *

--- a/benchmark/paddle/image/vgg.py
+++ b/benchmark/paddle/image/vgg.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
 #!/usr/bin/env python
 from paddle.trainer_config_helpers import *


--- a/benchmark/paddle/rnn/imdb.py
+++ b/benchmark/paddle/rnn/imdb.py
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 from __future__ import print_function
 import six.moves.cPickle as pickle
 import gzip

--- a/benchmark/paddle/rnn/provider.py
+++ b/benchmark/paddle/rnn/provider.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 import io, os
 import random
 import numpy as np

--- a/benchmark/paddle/rnn/rnn.py
+++ b/benchmark/paddle/rnn/rnn.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
 #!/usr/bin/env python

 from paddle.trainer_config_helpers import *

--- a/benchmark/tensorflow/image/alexnet.py
+++ b/benchmark/tensorflow/image/alexnet.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 from six.moves import xrange  # pylint: disable=redefined-builtin
 from datetime import datetime
 import math

--- a/benchmark/tensorflow/image/alexnet_multi_gpu.py
+++ b/benchmark/tensorflow/image/alexnet_multi_gpu.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 from six.moves import xrange  # pylint: disable=redefined-builtin
 from datetime import datetime
 import math

--- a/benchmark/tensorflow/image/googlenet.py
+++ b/benchmark/tensorflow/image/googlenet.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 from six.moves import xrange
 from datetime import datetime
 import math

--- a/benchmark/tensorflow/image/googlenet_multi_gpu.py
+++ b/benchmark/tensorflow/image/googlenet_multi_gpu.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 from six.moves import xrange  # pylint: disable=redefined-builtin
 from datetime import datetime
 import math

--- a/benchmark/tensorflow/image/smallnet_mnist_cifar.py
+++ b/benchmark/tensorflow/image/smallnet_mnist_cifar.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 from six.moves import xrange  # pylint: disable=redefined-builtin
 from datetime import datetime
 import math

--- a/benchmark/tensorflow/rnn/reader.py
+++ b/benchmark/tensorflow/rnn/reader.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 import os.path
 import io
 import numpy as np

--- a/benchmark/tensorflow/rnn/rnn.py
+++ b/benchmark/tensorflow/rnn/rnn.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
 #!/usr/bin/env python
 from six.moves import xrange  # pylint: disable=redefined-builtin
 import math

--- a/benchmark/tensorflow/rnn/rnn_multi_gpu.py
+++ b/benchmark/tensorflow/rnn/rnn_multi_gpu.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
 #!/usr/bin/env python
 from six.moves import xrange  # pylint: disable=redefined-builtin
 import re

--- a/cmake/external/boost.cmake
+++ b/cmake/external/boost.cmake
+# Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserve.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+include(ExternalProject)
+
+set(BOOST_PROJECT       "extern_boost")
+set(BOOST_VER           "1.66.0")
+set(BOOST_TAR           "boost_1_66_0")
+set(BOOST_URL           "https://dl.bintray.com/boostorg/release/${BOOST_VER}/source/${BOOST_TAR}.tar.gz")
+set(BOOST_SOURCES_DIR ${THIRD_PARTY_PATH}/boost)
+set(BOOST_DOWNLOAD_DIR  "${BOOST_SOURCES_DIR}/src/${BOOST_PROJECT}")
+set(BOOST_INCLUDE_DIR "${BOOST_DOWNLOAD_DIR}/${BOOST_TAR}" CACHE PATH "boost include directory." FORCE)
+
+include_directories(${BOOST_INCLUDE_DIR})
+
+ExternalProject_Add(
+    ${BOOST_PROJECT}
+    ${EXTERNAL_PROJECT_LOG_ARGS}
+    DOWNLOAD_DIR          ${BOOST_DOWNLOAD_DIR}
+    DOWNLOAD_COMMAND      wget --no-check-certificate ${BOOST_URL} -c -q -O ${BOOST_TAR}.tar.gz
+                          && tar zxf ${BOOST_TAR}.tar.gz
+    DOWNLOAD_NO_PROGRESS  1
+    PREFIX                ${BOOST_SOURCES_DIR}
+    CONFIGURE_COMMAND     ""
+    BUILD_COMMAND         ""
+    INSTALL_COMMAND       ""
+    UPDATE_COMMAND        ""
+)
+
+if (${CMAKE_VERSION} VERSION_LESS "3.3.0")
+    set(dummyfile ${CMAKE_CURRENT_BINARY_DIR}/boost_dummy.c)
+    file(WRITE ${dummyfile} "const char *dummy = \"${dummyfile}\";")
+    add_library(boost STATIC ${dummyfile})
+else()
+    add_library(boost INTERFACE)
+endif()
+
+add_dependencies(boost ${BOOST_PROJECT})
+list(APPEND external_project_dependencies boost)
+set(Boost_INCLUDE_DIR ${BOOST_INCLUDE_DIR})
--- a/cmake/external/eigen.cmake
+++ b/cmake/external/eigen.cmake
 INCLUDE(ExternalProject)

 SET(EIGEN_SOURCE_DIR ${THIRD_PARTY_PATH}/eigen3)
-
-INCLUDE_DIRECTORIES(${EIGEN_SOURCE_DIR}/src/extern_eigen3)
+SET(EIGEN_INCLUDE_DIR ${EIGEN_SOURCE_DIR}/src/extern_eigen3)
+INCLUDE_DIRECTORIES(${EIGEN_INCLUDE_DIR})

 ExternalProject_Add(
    extern_eigen3
@@ -28,3 +28,9 @@ endif()
 add_dependencies(eigen3 extern_eigen3)

 LIST(APPEND external_project_dependencies eigen3)
+
+IF(NOT WITH_C_API AND WITH_FLUID)
+    INSTALL(FILES ${EIGEN_INCLUDE_DIR}/Eigen/Core DESTINATION third_party/eigen3/Eigen)
+    INSTALL(DIRECTORY ${EIGEN_INCLUDE_DIR}/Eigen/src DESTINATION third_party/eigen3/Eigen)
+    INSTALL(DIRECTORY ${EIGEN_INCLUDE_DIR}/unsupported/Eigen DESTINATION third_party/eigen3/unsupported)
+ENDIF()
--- a/cmake/external/gflags.cmake
+++ b/cmake/external/gflags.cmake
@@ -52,7 +52,7 @@ ADD_DEPENDENCIES(gflags extern_gflags)

 LIST(APPEND external_project_dependencies gflags)

-IF(WITH_C_API)
+IF(WITH_C_API OR WITH_FLUID)
  INSTALL(DIRECTORY ${GFLAGS_INCLUDE_DIR} DESTINATION third_party/gflags)
  IF(ANDROID)
    INSTALL(FILES ${GFLAGS_LIBRARIES} DESTINATION third_party/gflags/lib/${ANDROID_ABI})

--- a/cmake/external/glog.cmake
+++ b/cmake/external/glog.cmake
@@ -68,7 +68,7 @@ LINK_LIBRARIES(glog gflags)

 LIST(APPEND external_project_dependencies glog)

-IF(WITH_C_API)
+IF(WITH_C_API OR WITH_FLUID)
  INSTALL(DIRECTORY ${GLOG_INCLUDE_DIR} DESTINATION third_party/glog)
  IF(ANDROID)
    INSTALL(FILES ${GLOG_LIBRARIES} DESTINATION third_party/glog/lib/${ANDROID_ABI})

--- a/cmake/external/openblas.cmake
+++ b/cmake/external/openblas.cmake
@@ -100,6 +100,11 @@ IF(NOT ${CBLAS_FOUND})
                \"${CBLAS_INSTALL_DIR}/lib -> ${CMAKE_INSTALL_PREFIX}/${TMP_INSTALL_DIR}\"
            )"
        )
+        INSTALL(CODE "execute_process(
+            COMMAND rm -r ${CMAKE_INSTALL_PREFIX}/${TMP_INSTALL_DIR}/cmake
+                    ${CMAKE_INSTALL_PREFIX}/${TMP_INSTALL_DIR}/pkgconfig
+            )"
+        )
    ENDIF()
 ENDIF(NOT ${CBLAS_FOUND})


--- a/cmake/external/protobuf.cmake
+++ b/cmake/external/protobuf.cmake
@@ -250,7 +250,7 @@ IF(NOT PROTOBUF_FOUND)
    SET(PROTOBUF_PROTOC_LIBRARY ${extern_protobuf_PROTOC_LIBRARY}
        CACHE FILEPATH "protoc library." FORCE)

-    IF(WITH_C_API)
+    IF(WITH_C_API OR WITH_FLUID)
        INSTALL(DIRECTORY ${PROTOBUF_INCLUDE_DIR} DESTINATION third_party/protobuf)
        IF(ANDROID)
            INSTALL(FILES ${PROTOBUF_LITE_LIBRARY} DESTINATION third_party/protobuf/lib/${ANDROID_ABI})

--- a/cmake/make_resource.py
+++ b/cmake/make_resource.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 import os
 import re
 import sys

--- a/doc/api/v1/data_provider/dataprovider_cn.rst
+++ b/doc/api/v1/data_provider/dataprovider_cn.rst
-.. _api_dataprovider:
-
-DataProvider的介绍
-==================
-
-DataProvider是PaddlePaddle负责提供数据的模块。其作用是将数据传入内存或显存，让神经网络可以进行训练或预测。用户可以通过简单使用Python接口 :ref:`api_pydataprovider2` ，来自定义传数据的过程。如果有更复杂的使用，或者需要更高的效率，用户也可以在C++端自定义一个 ``DataProvider`` 。
-
-PaddlePaddle需要用户在网络配置（trainer_config.py）中定义使用哪种DataProvider，并且在DataProvider中实现如何访问训练文件列表（train.list）或测试文件列表（test.list）。
-
- train.list和test.list存放在本地（推荐直接存放到训练目录，以相对路径引用)。一般情况下，两者均为纯文本文件，其中每一行对应一个数据文件地址：
-  
-  - 如果数据文件存于本地磁盘，这个地址则为它的绝对路径或相对路径(相对于PaddlePaddle程序运行时的路径)。
-  - 地址也可以为hdfs文件路径，或者数据库连接路径等。
-  - 由于这个地址会被DataProvider使用，因此，如何解析该地址也是用户自定义DataProvider时需要考虑的地方。
- 如果没有设置test.list，或设置为None，那么在训练过程中不会执行测试操作；否则，会根据命令行参数指定的测试方式，在训练过程中进行测试，从而防止过拟合。
--- a/doc/api/v1/data_provider/dataprovider_en.rst
+++ b/doc/api/v1/data_provider/dataprovider_en.rst
-Introduction
-==============
-DataProvider is a module that loads training or testing data into cpu or gpu
-memory for the following triaining or testing process.
-
-For simple use, users can use Python :code:`PyDataProvider` to dynamically reads
-the original data in any format or in any form, and then transfer them into a
-data format PaddlePaddle requires. The process is extremly flexible and highly
-customized, with sacrificing the efficiency only a little. This is extremly
-useful when you have to dynamically generate certain kinds of data according to,
-for example, the training performance.
-
-Besides, users also can customize a C++ :code:`DataProvider` for a more
-complex usage, or for a higher efficiency.
-
-The following parameters are required to define in the PaddlePaddle network
-configuration file (trainer_config.py): which DataProvider is chosen to used,
-and specific parameters for DataProvider, including training file list
-(train.list) and testing file list (test.list).
-
-Train.list and test.list are simply two plain text files, which defines path
-of training or testing data. It is recommended that directly placing them into
-the training directory, and reference to them by using a relative path (
-relative to the PaddePaddle program).
-
-Testing or evaluating will not be performed during training if the test.list is
-not set or set to None. Otherwise, PaddlePaddle will evaluate the trained model
-by the specified tesing data while training, every testing period (a user
-defined command line parameter in PaddlePaddle) to prevent over-fitting.
-
-Each line of train.list and test.list is an absolute or relative path (relative
-to the PaddePaddle program runtime) of data file. Fascinatingly more, each line
-can also be a HDFS file path or a SQL connection string. As long as the user
-assures how to access each file in DataProvider.
--- a/doc/api/v1/data_provider/pydataprovider2_cn.rst
+++ b/doc/api/v1/data_provider/pydataprovider2_cn.rst
-..  _api_pydataprovider2:
-
-PyDataProvider2的使用
-=====================
-
-PyDataProvider2是PaddlePaddle使用Python提供数据的推荐接口。该接口使用多线程读取数据，并提供了简单的Cache功能；同时可以使用户只关注如何从文件中读取每一条数据，而不用关心数据如何传输，如何存储等等。
-
-..  contents::
-
-MNIST的使用场景
---------------
-
-我们以MNIST手写识别为例，来说明PyDataProvider2的简单使用场景。
-
-样例数据
-++++++++
-
-MNIST是一个包含有70,000张灰度图片的数字分类数据集。样例数据 ``mnist_train.txt`` 如下：
-
-..  literalinclude:: src/mnist_train.txt
-
-其中每行数据代表一张图片，行内使用 ``;`` 分成两部分。第一部分是图片的标签，为0-9中的一个数字；第二部分是28*28的图片像素灰度值。 对应的 ``train.list`` 即为这个数据文件的名字：
-
-..  literalinclude:: src/train.list
-
-dataprovider的使用
-++++++++++++++++++
-
-..  literalinclude:: src/mnist_provider.dict.py
-
- 首先，引入PaddlePaddle的PyDataProvider2包。
- 其次，定义一个Python的 `Decorator <http://www.learnpython.org/en/Decorators>`_ `@provider`_ 。用于将下一行的数据输入函数标记成一个PyDataProvider2，同时设置它的input_types属性。
-  
-  - `input_types`_：设置这个PyDataProvider2返回什么样的数据。本例根据网络配置中 ``data_layer`` 的名字，显式指定返回的是一个28*28维的稠密浮点数向量和一个[0-9]的10维整数标签。
-
-    ..  literalinclude:: src/mnist_config.py
-         :lines: 9-10
-
-  - 注意：如果用户不显示指定返回数据的对应关系，那么PaddlePaddle会根据layer的声明顺序，来确定对应关系。但这个关系可能不正确，所以推荐使用显式指定的方式来设置input_types。
- 最后，实现数据输入函数（如本例的 ``process`` 函数）。
-
-  - 该函数的功能是：打开文本文件，读取每一行，将行中的数据转换成与input_types一致的格式，然后返回给PaddlePaddle进程。注意，
-    
-    - 返回的顺序需要和input_types中定义的顺序一致。
-    - 返回时，必须使用Python关键词 ``yield`` ，相关概念是 ``generator`` 。
-    - 一次yield调用，返回一条完整的样本。如果想为一个数据文件返回多条样本，只需要在函数中调用多次yield即可（本例中使用for循环进行多次调用）。
-  
-  - 该函数具有两个参数：
-  
-    - settings：在本例中没有使用，具体可以参考 `init_hook`_ 中的说明。
-    - filename：为 ``train.list`` 或 ``test.list`` 中的一行，即若干数据文件路径的某一个。
-
-网络配置中的调用
-++++++++++++++++
-
-在网络配置里，只需要一行代码就可以调用这个PyDataProvider2，如，
-
-..  literalinclude:: src/mnist_config.py
-     :lines: 1-7
-
-训练数据是 ``train.list`` ，没有测试数据，调用的PyDataProvider2是 ``mnist_provider`` 模块中的 ``process`` 函数。
-
-小结
-+++++
-
-至此，简单的PyDataProvider2样例就说明完毕了。对用户来说，仅需要知道如何从 **一个文件** 中读取 **一条样本** ，就可以将数据传送给PaddlePaddle了。而PaddlePaddle则会帮用户做以下工作：
-
-* 将数据组合成Batch进行训练
-* 对训练数据进行Shuffle
-* 多线程的数据读取
-* 缓存训练数据到内存(可选)
-* CPU->GPU双缓存
-
-是不是很简单呢？
-
-时序模型的使用场景
------------------
-样例数据
-++++++++
-
-时序模型是指数据的某一维度是一个序列形式，即包含时间步信息。所谓时间步信息，不一定和时间有关系，只是说明数据的顺序是重要的。例如，文本信息就是一个序列数据。
-
-本例采用英文情感分类的数据，即将一段英文文本数据，分类成正面情绪和负面情绪两类(用0和1表示)。样例数据 ``sentimental_train.txt`` 如下：
-
-..  literalinclude:: src/sentimental_train.txt
-
-dataprovider的使用
-++++++++++++++++++
-
-相对MNIST而言，这个dataprovider较复杂，主要原因是增加了初始化机制 `init_hook`_。本例的 ``on_init`` 函数就是根据该机制配置的，它会在dataprovider创建的时候执行。
-
- 其中 ``input_types`` 和在 `@provider`_ 中配置的效果一致。本例中的输入特征是词ID的序列，因此使用 ``integer_value_sequence`` 类型来设置。
- 将 ``dictionary`` 存入settings对象，在 ``process`` 函数中使用。 dictionary是从网络配置中传入的dict对象，即一个将单词字符串映射到单词ID的字典。
-
-..  literalinclude:: src/sentimental_provider.py
-
-网络配置中的调用
-++++++++++++++++
-
-调用这个PyDataProvider2的方法，基本上和MNIST样例一致，除了
-
-* 在配置中需要读取外部字典。
-* 在声明DataProvider的时候传入dictionary作为参数。
-
-..  literalinclude:: src/sentimental_config.py
-     :emphasize-lines: 12-14
-
-参考(Reference)
---------------
-
-@provider
-+++++++++
-
-``@provider`` 是一个Python的 `Decorator`_ ，可以将某一个函数标记成一个PyDataProvider2。如果不了解 `Decorator`_ 是什么也没关系，只需知道这是一个标记属性的方法就可以了。它包含的属性参数如下:
-
-*  input_types：数据输入格式。具体的格式说明，请参考 `input_types`_ 。
-*  should_shuffle：是不是要对数据做Shuffle。训练时默认shuffle，测试时默认不shuffle。
-*  min_pool_size：设置内存中最小暂存的数据条数，也是PaddlePaddle所能够保证的shuffle粒度。如果为-1，则会预先读取全部数据到内存中。
-*  pool_size： 设置内存中暂存的数据条数。如果为-1（默认），则不在乎内存暂存多少条数据。如果设置，则推荐大于训练时batch size的值，并且在内存足够的情况下越大越好。
-*  can_over_batch_size：是否允许暂存略微多余pool_size的数据。由于这样做可以避免很多死锁问题，一般推荐设置成True。
-*  calc_batch_size：可以传入一个函数，用于自定义每条数据的batch size（默认为1）。
-*  cache： 数据缓存的策略，具体请参考 `cache`_ 。
-*  init_hook：初始化时调用的函数，具体请参考 `init_hook`_ 。
-*  check：如果为true，会根据input_types检查数据的合法性。
-*  check_fail_continue：如果为true，那么当check出数据不合法时，会扔到这条数据，继续训练或预测。（对check=false的情况，没有作用）
-
-input_types
-+++++++++++
-
-PaddlePaddle的数据包括四种主要类型，和三种序列模式。
-
-四种数据类型：
-
-* dense_vector：稠密的浮点数向量。
-* sparse_binary_vector：稀疏的01向量，即大部分值为0，但有值的地方必须为1。
-* sparse_float_vector：稀疏的向量，即大部分值为0，但有值的部分可以是任何浮点数。
-* integer：整数标签。
-
-三种序列模式：
-
-* SequenceType.NO_SEQUENCE：不是一条序列
-* SequenceType.SEQUENCE：是一条时间序列
-* SequenceType.SUB_SEQUENCE： 是一条时间序列，且序列的每一个元素还是一个时间序列。
-
-不同的数据类型和序列模式返回的格式不同，列表如下：
-
-+----------------------+---------------------+-----------------------------------+------------------------------------------------+
-|                      | NO_SEQUENCE         | SEQUENCE                          |  SUB_SEQUENCE                                  |
-+======================+=====================+===================================+================================================+
-| dense_vector         | [f, f, ...]         | [[f, ...], [f, ...], ...]         | [[[f, ...], ...], [[f, ...], ...],...]         |
-+----------------------+---------------------+-----------------------------------+------------------------------------------------+
-| sparse_binary_vector | [i, i, ...]         | [[i, ...], [i, ...], ...]         | [[[i, ...], ...], [[i, ...], ...],...]         |
-+----------------------+---------------------+-----------------------------------+------------------------------------------------+
-| sparse_float_vector  | [(i,f), (i,f), ...] | [[(i,f), ...], [(i,f), ...], ...] | [[[(i,f), ...], ...], [[(i,f), ...], ...],...] |
-+----------------------+---------------------+-----------------------------------+------------------------------------------------+
-| integer_value        |  i                  | [i, i, ...]                       | [[i, ...], [i, ...], ...]                      |
-+----------------------+---------------------+-----------------------------------+------------------------------------------------+
-
-其中，f代表一个浮点数，i代表一个整数。
-
-注意：对sparse_binary_vector和sparse_float_vector，PaddlePaddle存的是有值位置的索引。例如，
-
- 对一个5维非序列的稀疏01向量 ``[0, 1, 1, 0, 0]`` ，类型是sparse_binary_vector，返回的是 ``[1, 2]`` 。
- 对一个5维非序列的稀疏浮点向量 ``[0, 0.5, 0.7, 0, 0]`` ，类型是sparse_float_vector，返回的是 ``[(1, 0.5), (2, 0.7)]`` 。
-
-init_hook
-+++++++++
-
-init_hook可以传入一个函数。该函数在初始化的时候会被调用，其参数如下:
-
-* 第一个参数是settings对象，它和数据传入函数的第一个参数（如本例中 ``process`` 函数的 ``settings`` 参数）必须一致。该对象具有以下两个属性：
-    * settings.input_types：数据输入格式，具体请参考 `input_types`_ 。
-    * settings.logger：一个logging对象。
-* 其他参数使用 ``kwargs`` （key word arguments）传入，包括以下两种：
-    * PaddlePaddle定义的参数: 1）is_train：bool型参数，表示用于训练或预测；2）file_list：所有文件列表。
-    * 用户定义的参数：使用args在网络配置中设置。
-
-注意：PaddlePaddle保留添加参数的权力，因此init_hook尽量使用 ``**kwargs`` 来接受不使用的函数以保证兼容性。
-
-cache
-+++++
-
-PyDataProvider2提供了两种简单的Cache策略：
-
-* CacheType.NO_CACHE：不缓存任何数据，每次都会从python端读取数据
-* CacheType.CACHE_PASS_IN_MEM：第一个pass会从python端读取数据，剩下的pass会直接从内存里
-  读取数据。 
-
-
-注意事项
--------
-
-可能的内存泄露问题
-++++++++++++++++++
-
-PaddlePaddle将train.list中的每一行都传递给process函数，从而生成多个generator。当训练数据非常多时，就会生成非常多的generator。
-
-虽然每个generator在没有调用的时候，是几乎不占内存的；但当调用过一次后，generator便会存下当前的上下文(Context)，而这个Context可能会非常大。并且，generator至少需要调用两次才会知道是否停止。所以，即使process函数里面只有一个yield，也需要两次随机选择到相同generator的时候，才会释放该段内存。
-
-..  code-block:: python
-
-    def func():
-        yield 0
-
-    f = func()  # 创建generator
-    tmp = next(f)  # 调用一次，返回0
-    tmp = next(f)  # 调用第二次的时候，才会Stop Iteration
-
-由于顺序调用这些generator不会出现上述问题，因此有两种解决方案：
-
-1. **最佳推荐**：将样本的地址放入另一个文本文件，train.list写入那个文本文件的地址。即不要将每一个样本都放入train.list。
-2. 在generator的上下文中尽量留下非常少的变量引用，例如
-
-..  code-block:: python
-
-    def real_process(fn):
-        # ... read from fn
-        return result   # 当函数返回的时候，python可以解除掉内部变量的引用。
-
-    def process(fn):
-        yield real_process(fn)
-
-注意：这个问题是PyDataProvider读数据时候的逻辑问题，很难整体修正。
-
-内存不够用的情况
-++++++++++++++++
-
-PyDataProvider2会尽可能多的使用内存。因此，对于内存较小的机器，推荐使用 ``pool_size`` 变量来设置内存中暂存的数据条。具体请参考 `@provider`_ 中的说明。
-
--- a/doc/api/v1/data_provider/pydataprovider2_en.rst
+++ b/doc/api/v1/data_provider/pydataprovider2_en.rst
-..  _api_pydataprovider2:
-
-PyDataProvider2
-===============
-
-We highly recommand users to use PyDataProvider2 to provide training or testing
-data to PaddlePaddle. The user only needs to focus on how to read a single
-sample from the original data file by using PyDataProvider2, leaving all of the
-trivial work, including, transfering data into cpu/gpu memory, shuffle, binary
-serialization to PyDataProvider2. PyDataProvider2 uses multithreading and a
-fanscinating but simple cache strategy to optimize the efficiency of the data
-providing process.
-
-DataProvider for the non-sequential model
-----------------------------------------
-
-Here we use the MNIST handwriting recognition data as an example to illustrate
-how to write a simple PyDataProvider.
-
-MNIST is a handwriting classification data set. It contains 70,000 digital
-grayscale images. Labels of the training sample range from 0 to 9. All the
-images have been size-normalized and centered into images with the same size
-of 28 x 28 pixels.
-
-A small part of the original data as an example is shown as below:
-
-.. literalinclude:: src/mnist_train.txt
-
-Each line of the data contains two parts, separated by :code:`;`. The first part is
-label of an image. The second part contains 28x28 pixel float values.
-
-Just write path of the above data into train.list. It looks like this:
-
-.. literalinclude:: src/train.list
-
-The corresponding dataprovider is shown as below:
-
-.. literalinclude:: src/mnist_provider.dict.py
-
-The first line imports PyDataProvider2 package.
-The main function is the process function, that has two parameters.
-The first parameter is the settings, which is not used in this example.
-The second parameter is the filename, that is exactly each line of train.list.
-This parameter is passed to the process function by PaddlePaddle.
-
-:code:`@provider` is a Python
-`Decorator <http://www.learnpython.org/en/Decorators>`_ .
-It sets some properties to DataProvider, and constructs a real PaddlePaddle
-DataProvider from a very simple user implemented python function. It does not
-matter if you are not familiar with `Decorator`_. You can keep it simple by
-just taking :code:`@provider` as a fixed mark above the provider function you
-implemented.
-
-`input_types`_ defines the data format that a DataProvider returns.
-In this example, it is set to a 28x28-dimensional dense vector and an integer
-scalar, whose value ranges from 0 to 9.
-`input_types`_ can be set to several kinds of input formats, please refer to the
-document of `input_types`_ for more details.
-
-
-The process method is the core part to construct a real DataProvider in
-PaddlePaddle. It implements how to open the text file, how to read one sample
-from the original text file, convert them into `input_types`_, and give them
-back to PaddlePaddle process at line 23.
-Note that data yielded by the process function must follow the same order that
-`input_types`_ are defined.
-
-
-With the help of PyDataProvider2, user can focus on how to generate ONE traning
-sample by using keywords :code:`yield`.
-:code:`yield` is a python keyword, and a concept related to it includes
-:code:`generator`.
-
-Only a few lines of codes need to be added into the training configuration file,
-you can take this as an example.
-
-.. literalinclude:: src/mnist_config.py
-
-Here we specify training data by :code:`train.list`, and no testing data is specified.
-The method which actually provide data is :code:`process`.
-
-User also can use another style to provide data, which defines the
-:code:`data_layer`'s name explicitly when `yield`. For example,
-the :code:`dataprovider` is shown as below.
-
-.. literalinclude:: src/mnist_provider.dict.py
-   :linenos:
-
-If user did't give the :code:`data_layer`'s name, PaddlePaddle will use
-the order of :code:`data_layer` definition roughly to determine which feature to
-which :code:`data_layer`. This order may be not correct, so TO DEFINE THE
-:code:`data_layer`'s NAMES EXPLICITLY IS THE RECOMMANDED WAY TO PROVIDER DATA.
-
-Now, this simple example of using PyDataProvider is finished.
-The only thing that the user should know is how to generte **one sample** from
-**one data file**.
-And PaddlePadle will do all of the rest things\:
-
-* Form a training batch
-* Shuffle the training data
-* Read data with multithreading
-* Cache the training data (Optional)
-* CPU-> GPU double buffering.
-
-Is this cool?
-
-..  _api_pydataprovider2_sequential_model:
-
-DataProvider for the sequential model
-------------------------------------
-A sequence model takes sequences as its input. A sequence is made up of several
-timesteps. The so-called timestep, is not necessary to have something to do
-with time. It can also be explained to that the order of data are taken into
-consideration into model design and training.
-For example, the sentence can be interpreted as a kind of sequence data in NLP
-tasks.
-
-Here is an example on data proivider for English sentiment classification data.
-The original input data are simple English text, labeled into positive or
-negative sentiment (marked by 0 and 1 respectively).
-
-A small part of the original data as an example can be found in the path below:
-
-.. literalinclude:: src/sentimental_train.txt
-
-The corresponding data provider can be found in the path below:
-
-.. literalinclude:: src/sentimental_provider.py
-
-This data provider for sequential model is a little more complex than that
-for MINST dataset.
-A new initialization method is introduced here.
-The method :code:`on_init` is configured to DataProvider by :code:`@provider`'s
-:code:`init_hook` parameter, and it will be invoked once DataProvider is
-initialized. The :code:`on_init` function has the following parameters:
-
-* The first parameter is the settings object.
-* The rest parameters are passed by key word arguments. Some of them are passed
-  by PaddlePaddle, see reference for `init_hook`_.
-  The :code:`dictionary` object is a python dict object passed from the trainer
-  configuration file, and it maps word string to word id.
-
-To pass these parameters into DataProvider, the following lines should be added
-into trainer configuration file.
-
-.. literalinclude:: src/sentimental_config.py
-
-The definition is basically same as MNIST example, except:
-* Load dictionary in this configuration
-* Pass it as a parameter to the DataProvider
-
-The `input_types` is configured in method :code:`on_init`. It has the same
-effect to configure them by :code:`@provider`'s :code:`input_types` parameter.
-However, the :code:`input_types` is set at runtime, so we can set it to
-different types according to the input data. Input of the neural network is a
-sequence of word id, so set :code:`seq_type` to :code:`integer_value_sequence`.
-
-Durning :code:`on_init`, we save :code:`dictionary` variable to
-:code:`settings`, and it will be used in :code:`process`. Note the settings
-parameter for the process function and for the on_init's function are a same
-object.
-
-The basic processing logic is the same as MNIST's :code:`process` method. Each
-sample in the data file is given back to PaddlePaddle process.
-
-Thus, the basic usage of PyDataProvider is here.
-Please refer to the following section reference for details.
-
-Reference
---------
-
-@provider
-+++++++++
-
-.. autofunction:: paddle.trainer.PyDataProvider2.provider
-
-input_types
-+++++++++++
-
-PaddlePaddle has four data types, and three sequence types.
-The four data types are:
-
-* :code:`dense_vector`: dense float vector.
-* :code:`sparse_binary_vector`: sparse binary vector, most of the value is 0, and
-  the non zero elements are fixed to 1.
-* :code:`sparse_float_vector`: sparse float vector, most of the value is 0, and some
-  non zero elements can be any float value. They are given by the user.
-* :code:`integer`: an integer scalar, that is especially used for label or word index.
-
-The three sequence types are:
-
-* :code:`SequenceType.NO_SEQUENCE` means the sample is not a sequence.
-* :code:`SequenceType.SEQUENCE` means the sample is a sequence.
-* :code:`SequenceType.SUB_SEQUENCE` means it is a nested sequence, that each timestep of
-  the input sequence is also a sequence.
-
-Different input type has a defferenct input format. Their formats are shown
-in the above table.
-
-+----------------------+---------------------+-----------------------------------+------------------------------------------------+
-|                      | NO_SEQUENCE         | SEQUENCE                          |  SUB_SEQUENCE                                  |
-+======================+=====================+===================================+================================================+
-| dense_vector         | [f, f, ...]         | [[f, ...], [f, ...], ...]         | [[[f, ...], ...], [[f, ...], ...],...]         |
-+----------------------+---------------------+-----------------------------------+------------------------------------------------+
-| sparse_binary_vector | [i, i, ...]         | [[i, ...], [i, ...], ...]         | [[[i, ...], ...], [[i, ...], ...],...]         |
-+----------------------+---------------------+-----------------------------------+------------------------------------------------+
-| sparse_float_vector  | [(i,f), (i,f), ...] | [[(i,f), ...], [(i,f), ...], ...] | [[[(i,f), ...], ...], [[(i,f), ...], ...],...] |
-+----------------------+---------------------+-----------------------------------+------------------------------------------------+
-| integer_value        |  i                  | [i, i, ...]                       | [[i, ...], [i, ...], ...]                      |
-+----------------------+---------------------+-----------------------------------+------------------------------------------------+
-
-where f represents a float value, i represents an integer value.
-
-init_hook
-+++++++++
-
-init_hook is a function that is invoked once the data provoder is initialized.
-Its parameters lists as follows:
-
-* The first parameter is a settings object, which is the same to :code:`settings`
-  in :code:`process` method. The object contains several attributes, including:
-
-  * :code:`settings.input_types`: the input types. Reference `input_types`_.
-  * :code:`settings.logger`: a logging object.
-
-* The rest parameters are the key word arguments. It is made up of PaddpePaddle
-  pre-defined parameters and user defined parameters.
-
-  * PaddlePaddle-defined parameters including:
-
-    * :code:`is_train` is a bool parameter that indicates the DataProvider is used in
-      training or testing.
-    * :code:`file_list` is the list of all files.
-
-  * User-defined parameters args can be set in training configuration.
-
-Note, PaddlePaddle reserves the right to add pre-defined parameter, so please
-use :code:`**kwargs` in init_hook to ensure compatibility by accepting the
-parameters which your init_hook does not use.
-
-cache
-+++++
-DataProvider provides two simple cache strategy. They are:
-
-* :code:`CacheType.NO_CACHE` means do not cache any data, then data is read at runtime by
-  the user implemented python module every pass.
-* :code:`CacheType.CACHE_PASS_IN_MEM` means the first pass reads data by the user
-  implemented python module, and the rest passes will directly read data from
-  memory.
--- a/doc/api/v1/data_provider/src/mnist_config.py
+++ b/doc/api/v1/data_provider/src/mnist_config.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
-from paddle.trainer_config_helpers import *
-
-define_py_data_sources2(
-    train_list='train.list',
-    test_list=None,
-    module='mnist_provider',
-    obj='process')
-
-img = data_layer(name='pixel', size=784)
-label = data_layer(name='label', size=10)
--- a/doc/api/v1/data_provider/src/mnist_provider.dict.py
+++ b/doc/api/v1/data_provider/src/mnist_provider.dict.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
-from paddle.trainer.PyDataProvider2 import *
-
-
-# Define a py data provider
-@provider(
-    input_types={'pixel': dense_vector(28 * 28),
-                 'label': integer_value(10)})
-def process(settings, filename):  # settings is not used currently.
-    f = open(filename, 'r')  # open one of training file
-
-    for line in f:  # read each line
-        label, pixel = line.split(';')
-
-        # get features and label
-        pixels_str = pixel.split(' ')
-
-        pixels_float = []
-        for each_pixel_str in pixels_str:
-            pixels_float.append(float(each_pixel_str))
-
-        # give data to paddle.
-        yield {"pixel": pixels_float, 'label': int(label)}
-
-    f.close()  # close file
--- a/doc/api/v1/data_provider/src/mnist_train.txt
+++ b/doc/api/v1/data_provider/src/mnist_train.txt
-5;0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.215686 0.533333 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.67451 0.992157 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.070588 0.886275 0.992157 0 0 0 0 0 0 0 0 0 0 0.192157 0.070588 0 0 0 0 0 0 0 0 0 0 0 0 0 0.670588 0.992157 0.992157 0 0 0 0 0 0 0 0 0 0.117647 0.933333 0.858824 0.313725 0 0 0 0 0 0 0 0 0 0 0 0.090196 0.858824 0.992157 0.831373 0 0 0 0 0 0 0 0 0 0.141176 0.992157 0.992157 0.611765 0.054902 0 0 0 0 0 0 0 0 0 0 0.258824 0.992157 0.992157 0.529412 0 0 0 0 0 0 0 0 0 0.368627 0.992157 0.992157 0.419608 0.003922 0 0 0 0 0 0 0 0 0 0.094118 0.835294 0.992157 0.992157 0.517647 0 0 0 0 0 0 0 0 0 0.603922 0.992157 0.992157 0.992157 0.603922 0.545098 0.043137 0 0 0 0 0 0 0 0.447059 0.992157 0.992157 0.956863 0.062745 0 0 0 0 0 0 0 0 0.011765 0.666667 0.992157 0.992157 0.992157 0.992157 0.992157 0.745098 0.137255 0 0 0 0 0 0.152941 0.866667 0.992157 0.992157 0.521569 0 0 0 0 0 0 0 0 0 0.070588 0.992157 0.992157 0.992157 0.803922 0.352941 0.745098 0.992157 0.945098 0.317647 0 0 0 0 0.580392 0.992157 0.992157 0.764706 0.043137 0 0 0 0 0 0 0 0 0 0.070588 0.992157 0.992157 0.776471 0.043137 0 0.007843 0.27451 0.882353 0.941176 0.176471 0 0 0.180392 0.898039 0.992157 0.992157 0.313725 0 0 0 0 0 0 0 0 0 0 0.070588 0.992157 0.992157 0.713725 0 0 0 0 0.627451 0.992157 0.729412 0.062745 0 0.509804 0.992157 0.992157 0.776471 0.035294 0 0 0 0 0 0 0 0 0 0 0.494118 0.992157 0.992157 0.968627 0.168627 0 0 0 0.423529 0.992157 0.992157 0.364706 0 0.717647 0.992157 0.992157 0.317647 0 0 0 0 0 0 0 0 0 0 0 0.533333 0.992157 0.984314 0.945098 0.603922 0 0 0 0.003922 0.466667 0.992157 0.988235 0.976471 0.992157 0.992157 0.788235 0.007843 0 0 0 0 0 0 0 0 0 0 0 0.686275 0.882353 0.364706 0 0 0 0 0 0 0.098039 0.588235 0.992157 0.992157 0.992157 0.980392 0.305882 0 0 0 0 0 0 0 0 0 0 0 0 0.101961 0.67451 0.321569 0 0 0 0 0 0 0 0.105882 0.733333 0.976471 0.811765 0.713725 0 0 0 0 0 0 0 0 0 0 0 0 0 0.65098 0.992157 0.321569 0 0 0 0 0 0 0 0 0 0.25098 0.007843 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0.94902 0.219608 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.968627 0.764706 0.152941 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.498039 0.25098 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-0;0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.298039 0.333333 0.333333 0.333333 0.337255 0.333333 0.333333 0.109804 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.027451 0.223529 0.776471 0.964706 0.988235 0.988235 0.988235 0.992157 0.988235 0.988235 0.780392 0.098039 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.14902 0.698039 0.988235 0.992157 0.988235 0.901961 0.87451 0.568627 0.882353 0.976471 0.988235 0.988235 0.501961 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.188235 0.647059 0.988235 0.988235 0.745098 0.439216 0.098039 0 0 0 0.572549 0.988235 0.988235 0.988235 0 0 0 0 0 0 0 0 0 0 0 0 0 0.2 0.933333 0.992157 0.941176 0.247059 0 0 0 0 0 0 0.188235 0.898039 0.992157 0.992157 0 0 0 0 0 0 0 0 0 0 0 0.039216 0.639216 0.933333 0.988235 0.913725 0.278431 0 0 0 0 0 0 0 0.113725 0.843137 0.988235 0.988235 0 0 0 0 0 0 0 0 0 0 0 0.235294 0.988235 0.992157 0.988235 0.815686 0.07451 0 0 0 0 0 0 0 0.333333 0.988235 0.988235 0.552941 0 0 0 0 0 0 0 0 0 0 0.211765 0.878431 0.988235 0.992157 0.701961 0.329412 0.109804 0 0 0 0 0 0 0 0.698039 0.988235 0.913725 0.145098 0 0 0 0 0 0 0 0 0 0.188235 0.890196 0.988235 0.988235 0.745098 0.047059 0 0 0 0 0 0 0 0 0 0.882353 0.988235 0.568627 0 0 0 0 0 0 0 0 0 0.2 0.933333 0.992157 0.992157 0.992157 0.447059 0.294118 0 0 0 0 0 0 0 0 0.447059 0.992157 0.768627 0 0 0 0 0 0 0 0 0 0 0.623529 0.988235 0.988235 0.988235 0.988235 0.992157 0.47451 0 0 0 0 0 0 0 0.188235 0.933333 0.87451 0.509804 0 0 0 0 0 0 0 0 0 0 0.992157 0.988235 0.937255 0.792157 0.988235 0.894118 0.082353 0 0 0 0 0 0 0.027451 0.647059 0.992157 0.654902 0 0 0 0 0 0 0 0 0 0 0 0.623529 0.988235 0.913725 0.329412 0.376471 0.184314 0 0 0 0 0 0 0.027451 0.513725 0.988235 0.635294 0.219608 0 0 0 0 0 0 0 0 0 0 0 0.196078 0.929412 0.988235 0.988235 0.741176 0.309804 0 0 0 0 0 0 0.529412 0.988235 0.678431 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.223529 0.992157 0.992157 1 0.992157 0.992157 0.992157 0.992157 1 0.992157 0.992157 0.882353 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.023529 0.478431 0.654902 0.658824 0.952941 0.988235 0.988235 0.988235 0.992157 0.988235 0.729412 0.278431 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.196078 0.647059 0.764706 0.764706 0.768627 0.580392 0.047059 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
-4;0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.180392 0.470588 0.623529 0.623529 0.623529 0.588235 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.243137 0.494118 0.862745 0.870588 0.960784 0.996078 0.996078 0.996078 0.996078 0.992157 0.466667 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.317647 0.639216 0.639216 0.639216 0.639216 0.639216 0.470588 0.262745 0.333333 0.929412 0.694118 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.811765 0.694118 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.811765 0.694118 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.811765 0.694118 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.184314 0.992157 0.694118 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.192157 0.996078 0.384314 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.454902 0.980392 0.219608 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.564706 0.941176 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.588235 0.776471 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.945098 0.560784 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.054902 0.952941 0.356863 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.337255 0.917647 0.109804 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.698039 0.701961 0.019608 0.4 0.662745 0.662745 0.662745 0.662745 0.662745 0.662745 0.662745 0.376471 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.090196 0.639216 0.972549 0.945098 0.913725 0.996078 0.996078 0.996078 0.996078 1 0.996078 0.996078 1 0.996078 0 0 0 0 0 0 0 0 0 0 0.007843 0.105882 0.717647 0.776471 0.905882 0.996078 0.996078 0.988235 0.980392 0.862745 0.537255 0.223529 0.223529 0.368627 0.376471 0.6 0.6 0.6 0 0 0 0 0 0 0 0 0.262745 0.470588 0.6 0.996078 0.996078 0.996078 0.996078 0.847059 0.356863 0.156863 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.909804 0.705882 0.823529 0.635294 0.490196 0.219608 0.113725 0.062745 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0.152941 0.152941 0.156863 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0;
--- a/doc/api/v1/data_provider/src/sentimental_config.py
+++ b/doc/api/v1/data_provider/src/sentimental_config.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
-from paddle.trainer_config_helpers import *
-
-dictionary = dict()
-...  #  read dictionary from outside
-
-define_py_data_sources2(
-    train_list='train.list',
-    test_list=None,
-    module='sentimental_provider',
-    obj='process',
-    # above codes same as mnist sample.
-    args={  # pass to provider.
-        'dictionary': dictionary
-    })
--- a/doc/api/v1/data_provider/src/sentimental_provider.py
+++ b/doc/api/v1/data_provider/src/sentimental_provider.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
-from paddle.trainer.PyDataProvider2 import *
-
-
-def on_init(settings, dictionary, **kwargs):
-    # on_init will invoke when data provider is initialized. The dictionary
-    # is passed from trainer_config, and is a dict object with type
-    # (word string => word id).
-
-    # set input types in runtime. It will do the same thing as
-    # @provider(input_types) will do, but it is set dynamically during runtime.
-    settings.input_types = {
-        # The text is a sequence of integer values, and each value is a word id.
-        # The whole sequence is the sentences that we want to predict its
-        # sentimental.
-        'data': integer_value_sequence(len(dictionary)),  # text input
-        'label': integer_value(2)  # label positive/negative
-    }
-
-    # save dictionary as settings.dictionary. 
-    # It will be used in process method.
-    settings.dictionary = dictionary
-
-
-@provider(init_hook=on_init)
-def process(settings, filename):
-    f = open(filename, 'r')
-
-    for line in f:  # read each line of file
-        label, sentence = line.split('\t')  # get label and sentence
-        words = sentence.split(' ')  # get words
-
-        # convert word string to word id
-        # the word not in dictionary will be ignored.
-        word_ids = []
-
-        for each_word in words:
-            if each_word in settings.dictionary:
-                word_ids.append(settings.dictionary[each_word])
-
-        # give data to paddle.
-        yield word_ids, int(label)
-
-    f.close()
--- a/doc/api/v1/data_provider/src/sentimental_train.txt
+++ b/doc/api/v1/data_provider/src/sentimental_train.txt
-0       I saw this movie at the AFI Dallas festival . It all takes place at a lake house and it looks wonderful .
-1       This documentary makes you travel all around the globe . It contains rare and stunning sequels from the wilderness .
-...
--- a/doc/api/v1/data_provider/src/train.list
+++ b/doc/api/v1/data_provider/src/train.list
-mnist_train.txt
--- a/doc/api/v1/index_cn.rst
+++ b/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
--- a/doc/api/v1/index_en.rst
+++ b/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
--- a/doc/api/v1/predict/src/predict_sample.py
+++ b/doc/api/v1/predict/src/predict_sample.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.
-
-from py_paddle import swig_paddle, DataProviderConverter
-from paddle.trainer.PyDataProvider2 import dense_vector
-from paddle.trainer.config_parser import parse_config
-
-TEST_DATA = [[[
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.215686, 0.533333, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.67451, 0.992157, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0.070588, 0.886275, 0.992157, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.192157,
-    0.070588, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.670588, 0.992157,
-    0.992157, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.117647, 0.933333, 0.858824, 0.313725,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.090196, 0.858824, 0.992157, 0.831373, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0.141176, 0.992157, 0.992157, 0.611765, 0.054902, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0.258824, 0.992157, 0.992157, 0.529412, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0.368627, 0.992157, 0.992157, 0.419608, 0.003922, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0.094118, 0.835294, 0.992157, 0.992157, 0.517647, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0.603922, 0.992157, 0.992157, 0.992157, 0.603922,
-    0.545098, 0.043137, 0, 0, 0, 0, 0, 0, 0, 0.447059, 0.992157, 0.992157,
-    0.956863, 0.062745, 0, 0, 0, 0, 0, 0, 0, 0, 0.011765, 0.666667, 0.992157,
-    0.992157, 0.992157, 0.992157, 0.992157, 0.745098, 0.137255, 0, 0, 0, 0, 0,
-    0.152941, 0.866667, 0.992157, 0.992157, 0.521569, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0.070588, 0.992157, 0.992157, 0.992157, 0.803922, 0.352941, 0.745098,
-    0.992157, 0.945098, 0.317647, 0, 0, 0, 0, 0.580392, 0.992157, 0.992157,
-    0.764706, 0.043137, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.070588, 0.992157, 0.992157,
-    0.776471, 0.043137, 0, 0.007843, 0.27451, 0.882353, 0.941176, 0.176471, 0,
-    0, 0.180392, 0.898039, 0.992157, 0.992157, 0.313725, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0.070588, 0.992157, 0.992157, 0.713725, 0, 0, 0, 0, 0.627451,
-    0.992157, 0.729412, 0.062745, 0, 0.509804, 0.992157, 0.992157, 0.776471,
-    0.035294, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.494118, 0.992157, 0.992157,
-    0.968627, 0.168627, 0, 0, 0, 0.423529, 0.992157, 0.992157, 0.364706, 0,
-    0.717647, 0.992157, 0.992157, 0.317647, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0.533333, 0.992157, 0.984314, 0.945098, 0.603922, 0, 0, 0, 0.003922,
-    0.466667, 0.992157, 0.988235, 0.976471, 0.992157, 0.992157, 0.788235,
-    0.007843, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.686275, 0.882353, 0.364706, 0,
-    0, 0, 0, 0, 0, 0.098039, 0.588235, 0.992157, 0.992157, 0.992157, 0.980392,
-    0.305882, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.101961, 0.67451, 0.321569,
-    0, 0, 0, 0, 0, 0, 0, 0.105882, 0.733333, 0.976471, 0.811765, 0.713725, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.65098, 0.992157, 0.321569, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0.25098, 0.007843, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1,
-    0.94902, 0.219608, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0.968627, 0.764706, 0.152941, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.498039, 0.25098, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0
-]], [[
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0.298039, 0.333333, 0.333333, 0.333333, 0.337255,
-    0.333333, 0.333333, 0.109804, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0.027451, 0.223529, 0.776471, 0.964706, 0.988235, 0.988235, 0.988235,
-    0.992157, 0.988235, 0.988235, 0.780392, 0.098039, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0.14902, 0.698039, 0.988235, 0.992157, 0.988235, 0.901961,
-    0.87451, 0.568627, 0.882353, 0.976471, 0.988235, 0.988235, 0.501961, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.188235, 0.647059, 0.988235, 0.988235,
-    0.745098, 0.439216, 0.098039, 0, 0, 0, 0.572549, 0.988235, 0.988235,
-    0.988235, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.2, 0.933333, 0.992157,
-    0.941176, 0.247059, 0, 0, 0, 0, 0, 0, 0.188235, 0.898039, 0.992157,
-    0.992157, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.039216, 0.639216, 0.933333,
-    0.988235, 0.913725, 0.278431, 0, 0, 0, 0, 0, 0, 0, 0.113725, 0.843137,
-    0.988235, 0.988235, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.235294, 0.988235,
-    0.992157, 0.988235, 0.815686, 0.07451, 0, 0, 0, 0, 0, 0, 0, 0.333333,
-    0.988235, 0.988235, 0.552941, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.211765,
-    0.878431, 0.988235, 0.992157, 0.701961, 0.329412, 0.109804, 0, 0, 0, 0, 0,
-    0, 0, 0.698039, 0.988235, 0.913725, 0.145098, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0.188235, 0.890196, 0.988235, 0.988235, 0.745098, 0.047059, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0.882353, 0.988235, 0.568627, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.2,
-    0.933333, 0.992157, 0.992157, 0.992157, 0.447059, 0.294118, 0, 0, 0, 0, 0,
-    0, 0, 0, 0.447059, 0.992157, 0.768627, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0.623529, 0.988235, 0.988235, 0.988235, 0.988235, 0.992157, 0.47451, 0, 0,
-    0, 0, 0, 0, 0, 0.188235, 0.933333, 0.87451, 0.509804, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0.992157, 0.988235, 0.937255, 0.792157, 0.988235, 0.894118,
-    0.082353, 0, 0, 0, 0, 0, 0, 0.027451, 0.647059, 0.992157, 0.654902, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0.623529, 0.988235, 0.913725, 0.329412, 0.376471,
-    0.184314, 0, 0, 0, 0, 0, 0, 0.027451, 0.513725, 0.988235, 0.635294,
-    0.219608, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.196078, 0.929412, 0.988235,
-    0.988235, 0.741176, 0.309804, 0, 0, 0, 0, 0, 0, 0.529412, 0.988235,
-    0.678431, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.223529, 0.992157,
-    0.992157, 1, 0.992157, 0.992157, 0.992157, 0.992157, 1, 0.992157, 0.992157,
-    0.882353, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0.023529,
-    0.478431, 0.654902, 0.658824, 0.952941, 0.988235, 0.988235, 0.988235,
-    0.992157, 0.988235, 0.729412, 0.278431, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0.196078, 0.647059, 0.764706, 0.764706, 0.768627,
-    0.580392, 0.047059, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
-    0, 0, 0, 0, 0, 0, 0
-]]]
-
-
-def main():
-    conf = parse_config("./mnist_model/trainer_config.py", "")
-    print conf.data_config.load_data_args
-    network = swig_paddle.GradientMachine.createFromConfigProto(
-        conf.model_config)
-    assert isinstance(network, swig_paddle.GradientMachine)  # For code hint.
-    network.loadParameters("./mnist_model/")
-    converter = DataProviderConverter([dense_vector(784)])
-    inArg = converter(TEST_DATA)
-    print network.forwardTest(inArg)
-
-
-if __name__ == '__main__':
-    swig_paddle.initPaddle("--use_gpu=0")
-    main()
--- a/doc/api/v1/predict/swig_py_paddle_cn.rst
+++ b/doc/api/v1/predict/swig_py_paddle_cn.rst
-.. _api_swig_py_paddle:
-
-基于Python的预测
-================
-
-预测流程
--------
-
-PaddlePaddle使用swig对常用的预测接口进行了封装，通过编译会生成py_paddle软件包，安装该软件包就可以在python环境下实现模型预测。可以使用python的 ``help()`` 函数查询软件包相关API说明。
-
-基于Python的模型预测，主要包括以下五个步骤。
-
-1. 初始化PaddlePaddle环境
-
-   在程序开始阶段，通过调用 ``swig_paddle.initPaddle()`` 并传入相应的命令行参数初始化PaddlePaddle。
-
-2. 解析模型配置文件
-   
-   初始化之后，可以通过调用 ``parse_config()`` 解析训练模型时用的配置文件。注意预测数据通常不包含label, 同时预测网络通常直接输出最后一层的结果而不是像训练网络一样再接一层cost layer，所以一般需要对训练用的模型配置文件稍作相应修改才能在预测时使用。
-
-3. 构造paddle.GradientMachine
-  
-   通过调用 ``swig_paddle.GradientMachine.createFromConfigproto()`` 传入上一步解析出来的模型配置就可以创建一个 ``GradientMachine``。
-
-4. 准备预测数据
-  
-   swig_paddle中的预测接口的参数是自定义的C++数据类型，py_paddle里面提供了一个工具类 ``DataProviderConverter`` 可以用于接收和PyDataProvider2一样的输入数据并转换成预测接口所需的数据类型。
-
-5. 模型预测
-  
-   通过调用 ``forwardTest()`` 传入预测数据，直接返回计算结果。
-
-
-预测Demo
--------
-
-如下是一段使用mnist model来实现手写识别的预测代码。完整的代码见 ``src_root/doc/ui/predict/predict_sample.py`` 。mnist model可以通过 ``src_root\demo\mnist`` 目录下的demo训练出来。
-
-..  literalinclude:: src/predict_sample.py
-    :language: python
-    :lines: 15-18,121-136
-
-
-Demo预测输出如下，其中value即为softmax层的输出。由于TEST_DATA包含两条预测数据，所以输出的value包含两个向量 。
-
-..  code-block:: text
-
-    [{'id': None, 'value': array(
-      [[  5.53018653e-09,   1.12194102e-05,   1.96644767e-09,
-          1.43630644e-02,   1.51111044e-13,   9.85625684e-01,
-          2.08823112e-10,   2.32777140e-08,   2.00186201e-09,
-          1.15501715e-08],
-       [  9.99982715e-01,   1.27787406e-10,   1.72296313e-05,
-          1.49316648e-09,   1.36540484e-11,   6.93137714e-10,
-          2.70634608e-08,   3.48565123e-08,   5.25639710e-09,
-          4.48684503e-08]], dtype=float32)}]
-
-
--- a/doc/api/v1/predict/swig_py_paddle_en.rst
+++ b/doc/api/v1/predict/swig_py_paddle_en.rst
-Python Prediction
-==================
-
-PaddlePaddle offers a set of clean prediction interfaces for python with the help of
-SWIG. The main steps of predict values in python are:
-
-* Parse training configurations
-* Construct GradientMachine
-* Prepare data
-* Predict
-
-Here is a sample python script that shows the typical prediction process for the
-MNIST classification problem. A complete sample code could be found at
-:code:`src_root/doc/ui/predict/predict_sample.py`.
-
-..  literalinclude:: src/predict_sample.py
-    :language: python
-    :lines: 15-18,90-100,101-104
-
-The module that does the most of the job is py_paddle.swig_paddle, it's
-generated by SWIG and has complete documents, for more details you can use
-python's :code:`help()` function. Let's walk through the above python script:
-
-* At the beginning, use :code:`swig_paddle.initPaddle()` to initialize
-  PaddlePaddle with command line arguments, for more about command line arguments
-  see :ref:`cmd_detail_introduction` .
-* Parse the configuration file that is used in training with :code:`parse_config()`.
-  Because data to predict with always have no label, and output of prediction work
-  normally is the output layer rather than the cost layer, so you should modify
-  the configuration file accordingly before using it in the prediction work.
-* Create a neural network with
-  :code:`swig_paddle.GradientMachine.createFromConfigproto()`, which takes the
-  parsed configuration :code:`conf.model_config` as argument. Then load the
-  trained parameters from the model with :code:`network.loadParameters()`.
-* Create a data converter object of utility class :code:`DataProviderConverter`.
-    - Note: As swig_paddle can only accept C++ matrices, we offer a utility
-      class DataProviderConverter that can accept the same input data with
-      PyDataProvider2, for more information please refer to document
-      of :ref:`api_pydataprovider2` .
-* Do the prediction with :code:`forwardTest()`, which takes the converted
-  input data and outputs the activations of the output layer.
-
-Here is a typical output:
-
-..  code-block:: text
-
-    [{'id': None, 'value': array([[  5.53018653e-09,   1.12194102e-05,   1.96644767e-09,
-          1.43630644e-02,   1.51111044e-13,   9.85625684e-01,
-          2.08823112e-10,   2.32777140e-08,   2.00186201e-09,
-          1.15501715e-08],
-       [  9.99982715e-01,   1.27787406e-10,   1.72296313e-05,
-          1.49316648e-09,   1.36540484e-11,   6.93137714e-10,
-          2.70634608e-08,   3.48565123e-08,   5.25639710e-09,
-          4.48684503e-08]], dtype=float32)}]
-
-:code:`value` is the output of the output layer, each row represents result of
-the corresponding row in the input data, each element represents activation of
-the corresponding neuron in the output layer.
-
--- a/doc/api/v2/fluid/layers.rst
+++ b/doc/api/v2/fluid/layers.rst
@@ -18,6 +18,11 @@ dynamic_lstm
 ..  autofunction:: paddle.v2.fluid.layers.dynamic_lstm
    :noindex:

+dynamic_gru
+-----------
+..  autofunction:: paddle.v2.fluid.layers.dynamic_gru
+    :noindex:
+
 data
 ----
 ..  autofunction:: paddle.v2.fluid.layers.data
@@ -499,3 +504,33 @@ swish
 ------
 ..  autofunction:: paddle.v2.fluid.layers.swish
    :noindex:
+
+im2sequence
+------
+..  autofunction:: paddle.v2.fluid.layers.im2sequence
+    :noindex:
+
+edit_distance
+---------------
+..  autofunction:: paddle.v2.fluid.layers.edit_distance_error
+    :noindex:
+
+ctc_greedy_decoder
+---------------
+..  autofunction:: paddle.v2.fluid.layers.ctc_greedy_decoder
+    :noindex:
+
+l2_normalize
+------------
+..  autofunction:: paddle.v2.fluid.layers.l2_normalize
+    :noindex:
+
+sequence_reshape
+----------------
+..  autofunction:: paddle.v2.fluid.layers.sequence_reshape
+    :noindex:
+
+row_conv
+--------
+..  autofunction:: paddle.v2.fluid.layers.row_conv
+    :noindex:
--- a/doc/design/csp.md
+++ b/doc/design/csp.md
+# Design Doc: CSP in PaddlePaddle Fluid
+
+## Motivation
+
+Concurrent programming is important for deep learning.  Few example applications are:
+
+1.  The main thread keeps reading the next mini-batch while another thread uses the GPU for computing.
+2.  The main thread performs the computation while another thread uploads the local gradients from each trainer to the parameter server.
+
+Most DL systems, including TensorFlow, Caffe2, and MxNet, can asynchronously execute operators in a graph. However, Fluid doesn't have the concept of a graph at all, as the design goal of Fluid is that of a programming language.
+
+## Concurrent Programming Models
+
+There were many concurrent programming models, implemented in various forms:
+
+| concurrent programming model | implementation |
+|-----|-----|
+| mutex | types and functions in standard libraries |
+| semaphore | types and functions in standard libraries |
+| communicating sequential processes (CSP) | Go programming language |
+| actor model | Erlang programming language |
+| message passing | MPI |
+| bulk synchronous parallel (BSP) | Pregel distributed programming framework |
+
+Since Fluid was designed to be a programming language, we would like to implement CSP in Fluid.
+
+### CSP v.s. Actor Model
+
+A well-known implementation of Actor Model is the Erlang programming language.  In Actor Model, *processes* could send messages to another process and receive messages from another process given the process IDs.  We can find the three ingredients, process with ID, send, and recv, in MPI too.  Indeed, we can rewrite Erlang programs in Python + MPI with possibly fewer lines of code.  Our concern with Actor Model is that it doesn't seem reasonable to implement process management in a programming language's runtime library; instead, it should be the operating systems' responsibility to manage processes and libraries like MPI for send/recv.
+
+## CSP in Fluid
+
+Fluid has two fundamental control-flows: *if-else* and *while*.  If we are to implement CSP, we need the following:
+
+1. a new data type: *channel* and operators *send* and *recv*,
+1. *goroutine* or thread, and
+1. a new control-flow: select.
+
+We also need Python wrappers for the above components.
+
+The type *channel* is conceptually the blocking queue.  In Go, its implemented is a [blocking circular queue](https://github.com/golang/go/blob/68ce117cf17b8debf5754bfd476345779b5b6616/src/runtime/chan.go#L31-L50), which supports send and recv.
+
+The `select` operation has been in OS kernels long before Go language.  All Unix kernels implement system calls *poll* and *select*.  They monitor multiple file descriptors to see if I/O is possible on any of them.  This takes O(N) time.  Since Linux 2.6, a new system call, *epoll*, can do the same in O(1) time.  In BSD systems, there is a similar system call *kqueue*.  Go's Linux implementation uses epoll.
+
+It might be a good idea to implement Fluid's select using epoll too.  In this design doc, we start from the O(N) way, so we could focus on Python binding and the syntax.
+
+### Type Channel
+
+Fluid supports many data types:
+
+1. Tensor,
+1. Row-sparse Tensor
+1. LoD Tensor,
+1. Tensor array, etc
+
+Each data type is registered in the [`framework.proto`](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/framework/framework.proto#L117-L127) as an enum value.  To add a new type channel, we need to add a new type enum.
+
+To expose a C++ type to Python, we need to edit the [`pybind.cc`](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/pybind/pybind.cc) file.  [Here](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/pybind/pybind.cc#L120-L164) is an example how we expose C++ class LoDTensor.
+
+## Syntax Design
+
+### Create Channel
+
+In Go, we create a channel by specifying the element type and buffer size:
+
+```go
+ch  := make(chan int)       // a channel without buffer
+ch1 := make(chan int, 100)  // a channel that can buffer 100 ints.
+```
+
+In Fluid, we should be able to do the same:
+
+```python
+ch  = fluid.make_chan(dtype=INT)
+ch1 = fluid.make_chan(dtype=INT, 100)
+```
+
+In addition to that, we want channels that can hold more complex element types, e.g., Tensors of float16:
+
+```python
+ch = fluid.make_chan(dtype=Tensor, etype=float16)
+```
+
+or Tensors of Tensors of float16 etc.
+
+The point here is that we need a consistent way to compose types, like in C++ we can have `Tensor<Tensor<...<float16>...> >`.
+
+### Send and Recv
+
+### Select
+
+## Example Programs
+
+### 1. RPC between Trainers and Parameter Servers
+
+### 2. Concurrent Minibatch Loading
--- a/doc/design/dist_refactor/parameter_server.md
+++ b/doc/design/dist_refactor/parameter_server.md
@@ -9,16 +9,16 @@ different purposes.

 ## Background

-The previous implementations of the parameter server does not run a
+The previous implementations of the parameter server do not run a
 fluid sub-program. Parameter initialization, optimizer computation, network
 communication and checkpointing are implemented twice on both the
-trainer and the parameter server.
+trainer as well as the parameter server.

-It would be great if we can write code once and use them on both the
-trainer and the parameter server: reduces code duplication and
-improves extensibility. Given that after the current refactor, we are
-representing everything as a computing graph on the
-trainer. Representing everything as a computing graph on the parameter
+It would be great if we can write code once and use them on both: the
+trainer and the parameter server, since this reduces code duplication and
+improves extensibility. Given that after the current refactoring, we are
+representing everything as a computation graph on the
+trainer. Representing everything as a computation graph on the parameter
 server becomes a natural extension.

 ## Design
@@ -30,9 +30,9 @@ into sub-programs to be scheduled on different nodes with the following
 steps:

 1. OP placement: the OPs will be placed on different nodes according
-   to heuristic that minimizes estimated total computation
+   to a heuristic that minimizes the estimated total computation
   time. Currently we will use a simple heuristic that puts parameter
-   varable on parameter server workers and everything else on trainer
+   variable on parameter server workers and everything else on trainer
   workers.
 1. Add communication OPs to enable the communication between nodes.

@@ -47,22 +47,22 @@ After converting:

 <img src="src/dist-graph.png" width="700"/>

-1. The parameter variable W and it's optimizer program are placed on the parameter server.
+1. The parameter variable W and its optimizer program are placed on the parameter server.
 1. Operators are added to the program.
   - *Send* sends data to the connected *Recv* operator.  The
 	 scheduler on the receive node will only schedule *Recv* operator
 	 to run when the *Send* operator has ran (the *Send* OP will mark
 	 the *Recv* OP runnable automatically).
-   - *Enueue* enqueues the input variable, it can block until space
+   - *Enqueue* enqueues the input variable, it can block until space
     become available in the queue.
   - *Dequeue* outputs configurable numbers of tensors from the
-     queue. It will block until the queue have the required number of
+     queue. It will block until the queue has the required number of
     tensors.


 ### Benefits

- Model parallelism become easier to implement: it's an extension to
+- Model parallelism becomes easier to implement: it is an extension to
  the trainer - parameter server approach. We can have several "Transpilers"
  to achieve different goals.
 - User-defined optimizer is easier to add - user can now express it as
@@ -72,22 +72,22 @@ After converting:

 ### Challenges

- It's important to balance the parameter shards of on multiple
-  parameter server. If a single parameter is very big (some
+- It is important to balance the parameter shards on multiple
+  parameter servers. If a single parameter is very big (for example: some
  word-embedding, fully connected, softmax layer), we need to
  automatically partition the single parameter onto different
  parameter servers when possible (only element-wise optimizer depends
  on the parameter variable).
- In the "Aync SGD" figure, the "W" variable on the parameter server
-  could be read and wrote concurrently. See
+- In the "Async SGD" figure, the "W" variable on the parameter server
+  could be read and written concurrently. See
  [here](https://github.com/PaddlePaddle/Paddle/pull/6394) for more
-  details about concurrent program in fluid.
+  details about concurrent program in Fluid.

 ### Discussion

 - Can the Enqueue OP be implemented under our current tensor design
-  (puts the input tensor into the queue tensor)?
- *Dequeue* OP will have variable numbers of output (depends on the
+  (put the input tensor into the queue tensor)?
+- *Dequeue* OP will have variable numbers of output (depending on the
  `min_count` attribute), does our current design support it? (similar
  question for the *Add* OP)


--- a/doc/design/error_clip.md
+++ b/doc/design/error_clip.md
@@ -46,12 +46,12 @@ class ErrorClipByValue(BaseErrorClipAttr):
        self.min = min

    def append_clip_op(self, block, grad_name):
-        block.append_op(
-            type="clip",
-            inputs={"X": grad_name},
-            outputs={"Out": grad_name},
-            attrs={"min": self.min,
-                   "max": self.max})
+        clip_op_desc = block.desc.append_op()
+        clip_op_desc.set_type("clip")
+        clip_op_desc.set_input("X", [grad_name])
+        clip_op_desc.set_output("Out", [grad_name])
+        clip_op_desc.set_attr("min", self.min)
+        clip_op_desc.set_attr("max", self.max)
 ```

 The `BaseErrorClipAttr` have one main member functions: `append_clip_op(self, block, grad_name)`.
@@ -80,6 +80,11 @@ def error_clip_callback(block, context):
                         op_desc.output_arg_names()):
        fwd_var = block.var_recursive(grad_to_var[grad_n])
        error_clip = getattr(fwd_var, "error_clip", None)
+        if not (error_clip is None or isinstance(error_clip,
+                                                 BaseErrorClipAttr)):
+            raise TypeError(
+                "Variable's error_clip should be an instance of BaseErrorClipAttr or None."
+            )
        if error_clip is not None:
            error_clip.append_clip_op(block, grad_n)
 ```

--- a/doc/design/fluid.md
+++ b/doc/design/fluid.md
@@ -105,18 +105,10 @@ There are two ways to execute a Fluid program.  When a program is executed, it c

 There is a C++ class [`Executor`](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/framework/executor.h), which runs a `ProgramDesc`, similar to how an interpreter runs a Python program.

-Fluid is moving towards the direction of a compiler, which is explain in more detail later in this article.
+Fluid is moving towards the direction of a compiler, which is explain in [fluid_compiler.md](fluid_compiler.md).

 ## Backward Compatibility of Fluid

 Given all the advantages from the removal of the concept of a *model*, hardware manufacturers might still prefer the existence of the concept of a model, so it would be easier for them to support multiple frameworks all at once and could run a trained model during inference.  For example, Nervana, a startup company acquired by Intel, has been working on an XPU that reads the models in the format known as [n-graph](https://github.com/NervanaSystems/ngraph).  Similarly, [Movidius](https://www.movidius.com/) is producing a mobile deep learning chip that reads and runs graphs of operators.  The well-known [ONNX](https://github.com/onnx/onnx) is also a file format of graphs of operators.

 For Fluid, we can write a converter that extracts the parts in the `ProgramDesc` protobuf message, converts them into a graph of operators, and exports the graph into the ONNX or n-graph format.
-
-## Towards a Deep Learning Language and the Compiler
-
-We can change the `if-then-else` and loop structure a little bit in the above Fluid example programs, to make it into a new programming language, different than Python.
-
-Even if we do not invent a new language, as long as we get the `ProgramDesc` message filled in, we can write a transpiler, which translates each invocation to an operator, into a C++ call to a kernel function of that operator. For example, a transpiler that weaves the CUDA kernels outputs an NVIDIA-friendly C++ program, which can be built using `nvcc`.  Another transpiler could generate MKL-friendly code that should be built using `icc` from Intel.  More interestingly, we can translate a Fluid program into its distributed version of two `ProgramDesc` messages, one for running on the trainer process, and the other one for the parameter server.  For more details of the last example, the [concurrent programming design](concurrent_programming.md) document would be a good pointer.  The following figure explains the proposed two-stage process:
-
-![](fluid-compiler.png)
--- a/doc/design/fluid_compiler.md
+++ b/doc/design/fluid_compiler.md
+# PaddlePaddle Fluid: Towards a Compiled Programming Language
+
+As described in [fluid.md](fluid.md), when a Fluid application program
+runs, it generates a `ProgramDesc` protobuf message as an intermediate
+representation of itself.  The C++ class `Executor` can run this
+protobuf message as an interpreter.  This article describes the Fluid
+compiler.
+
+![](fluid-compiler.png)
+
+## ProgramDesc
+
+Before we go deeper into the idea of compiled language, let us take a
+look at a simple example Fluid application.
+
+```python
+import "fluid"
+
+func paddlepaddle() {
+  X = fluid.read(...)
+  W = fluid.Tensor(...)
+  Y = fluid.mult(X, W)
+}
+```
+
+This program consists of a [block](block.md) of three operators --
+`read`, `assign`, and `mult`.  Its `ProgramDesc` message looks like
+the following
+
+```protobuf
+message ProgramDesc {
+  block[0] = Block {
+    vars = [X, W, Y],
+    ops = [
+      read(output = X)
+      assign(input = ..., output = W)
+      mult(input = {X, W}, output = Y)
+    ],
+  }
+}
+```
+ 
+## Transpilers
+
+We can write a transpiler program that takes a `ProgramDesc`, e.g.,
+the above one, and outputs another `ProgramDesc`.  Let us take some
+examples:
+
+1. *Memory optimization transpiler*: We can write a transpiler that
+   inserts some `FreeMemoryOp`s in the above example `ProgramDesc` so
+   to free memory early, before the end of an iteration, so to keep a
+   small memory footprint.
+
+1. *Distributed training transpiler*: We can write a transpiler that
+   converts a`ProgramDesc` into its distributed version of two
+   `ProgramDesc`s -- one for running by the trainer processes and the
+   other for the parameter server.
+
+In the rest of this article, we talk about a special kind of
+transpiler, *Native code generator*, which takes a `ProgramDesc` and
+generates a `.cu` (or `.cc`) file, which could be built by C++
+compilers (gcc, nvcc, icc) into binaries.
+
+## Native Code Generator
+
+For the above example, the native code generator transpiler, say, the
+CUDA code generator, should generate a `main` function:
+
+```c++
+void main() {
+  auto X = fluid_cuda_read(...);
+  auto W = fluid_cuda_create_tensor(...);
+  auto Y = fluid_cuda_mult(X, W);
+}
+```
+
+and the definitions of functions `fluid_cuda_read`,
+`fluid_cuda_create_tensor`, and `fluid_cuda_mult`.  Please be aware
+that each function could just define a C++ instance of an operator and
+run it.  For example
+
+```c++
+paddle::Tensor fluid_cuda_read(...) {
+  paddle::Tensor t;
+  paddle::operator::Read r(&t, ...);
+  r.Run();
+  return t;
+}
+```
+
+For computational operators that have multiple *kernels*, each for a
+specific hardware platform, for example, the `mult` operator, the
+generated code should call its CUDA kernel:
+
+```c++
+paddle::Tensor fluid_cuda_mult(const paddle::Tensor& a, 
+                               const paddle::Tensor& b) {
+  paddle::Tensor t;
+  paddle::operator::Mult m(a, b, ...);
+  Mult.Run(cuda_context);
+}
+```
+
+where `cuda_context` could be a global variable of type
+`paddle::CUDADeviceContext`.
+
+## Multi-Block Code Generation
+
+Most Fluid application programs may have more than one blocks.  To
+execute them, we need to trace [scopes](scope.md).
--- a/doc/design/ops/sequence_decoder.md
+++ b/doc/design/ops/sequence_decoder.md
@@ -22,7 +22,7 @@ The current `LoDTensor` is designed to store levels of variable-length sequences
 The integers in each level represent the begin and end (not inclusive) offset of a sequence **in the underlying tensor**,
 let's call this format the **absolute-offset LoD** for clarity.

-The relative-offset LoD can retrieve any sequence very quickly but fails to represent empty sequences, for example, a two-level LoD is as follows
+The absolute-offset LoD can retrieve any sequence very quickly but fails to represent empty sequences, for example, a two-level LoD is as follows
 ```python
 [[0, 3, 9]
 [0, 2, 3, 3, 3, 9]]
@@ -119,7 +119,7 @@ def generate():
        encoder_ctx_expanded = pd.lod_expand(encoder_ctx, target_word)
        decoder_input = pd.fc(
            act=pd.activation.Linear(),
-            input=[target_word, encoder_ctx],
+            input=[target_word, encoder_ctx_expanded],
            size=3 * decoder_dim)
        gru_out, cur_mem = pd.gru_step(
            decoder_input, mem=decoder_mem, size=decoder_dim)

--- a/doc/design/switch_kernel.md
+++ b/doc/design/switch_kernel.md
 ## Background
-Every operator has many kernels because there are multiple data types, places, data layout that Fluid supports. We use the `KernelType` to describe kernel types that operators can hold. 
+Every operator has many kernels because there are multiple data types, places, data layout, library type that Fluid supports. We use the `OpKernelType ` to describe kernel types that operators can hold.

-The `KernelType` is as follows.
+The `OpKernelType ` is as follows:

-```
-struct KernelType {
+```cpp
+struct OpKernelType {
  Place place_;
  DataType data_type_;
-  LayoutType layout_;
+  DataLayout data_layout_;
+  LibraryType library_type_;
 };
 ```

-The `place_` is a descriptor of the device and the computational library, e.g., `MKLDNNPlace`, `CUDAPlace`.
+- The `place_` is a descriptor of the device, e.g., CPUPlace, CUDAPlace.

-The `data_type_` is the data type that this kernel performs on, e.g., `FP32`, `INT64`. Note that one kernel may have inputs with different data types. However, it will be a major `data_type`. For example, the `cross_entropy` takes `int64` as it label, and `double`/`float` as its input logit and output cost. The major `data_type` of `cross_entropy` is `float`/`double`.
+- The `data_type_` is the data type that this kernel performs on, e.g., `FP32`, `INT64`. Note that one kernel may have inputs with different data types. However, it will be a major `data_type`. For example, the `cross_entropy` takes `int64` as it label, and `double`/`float` as its input logit and output cost. The major `data_type` of `cross_entropy` is `float` or `double`.

-The `layout` is useful for some computational library. One example is that MKLDNN uses many kinds of layout, such as `nChw8c`. Each kind of layout will invoke the different kernel.
+- The `data_layout_ ` is useful for some computational library. One example is that MKLDNN uses many kinds of layout, such as `nChw8c`. Each kind of layout will invoke the different kernel.
+
+- The `library_type_` describes the computational library, e.g., `MKLDNN`, `CUDNN`.

 ## Problem

@@ -25,42 +28,72 @@ We register a kernel for every operator and every kernel type ideally. However,
 2. Some operators will take too many memory. It is better to force them into CPU. However, the rest of operators in this neural network will be performed on GPU, i.e., model parallel problem.
 3. Some layout and place are particular. One example is that MKLDNN uses `nChw8` and there is no other library uses `nChw8c`.

-Problems under these situations are similar. We can formalise this problem as follow.
+Take one situation to give a detailed explanation, if we have two Operators: OP1 and OP2, OP1 has one output `op1_to_op2`, and `op1_to_op2` is the input of OP2.
+
+If OP1 and OP2 run on the same place(for example CPUPlace), then `op1_2_op2` can be used directly by OP2.
+
+```
+OP1(CPUPlace)
+     |
+ op1_2_op2
+     |
+OP2(CPUPlace)
+```
+
+If OP1 and OP2 run one different place, then OP2 cannot `use op1_2_op2` directly.
+
+Problems under these situations are similar. We can formalize this problem as follow.

 We register kernels with types $KT = \{kt_1, kt_2, kt_3, ...\}$ for one operator. The inputs of this operator should be run on kernel type $kt_{?}$, which the $kt_{?} \notin KT$. How to cast the input of this operator from $kt_{?}$ to any of kernel type in $KT$.

-## Solution
+## Solution: data transform

-It is clearly that transforming inputs of an operator toadapt another kernel type is not related to the particular operator. So we should register these transformation methods as global methods.
+It is clear that transforming inputs of an operator to adapt another kernel type is not related to the particular operator. So we should register these transformation methods as global methods.

-We can infer a kernel type from the inputs of an operators. We let this kernel type as `actual kernel type`, which means this kernel type is the actually kernel type that operator should be performed.
+We can infer kernel type for each input of an operator. We let this kernel type as `actual kernel type for var`, which means this kernel type is the kernel type that can process this input variable.

 We can get a kernel type by 1) The configuration of operator description. (Users may want to force use `MKL` for `conv` operator). 2) The place of the current executor. (Executor is running on GPU). This kernel type is what we expect the operator will be performed on. We let this kernel type as `expect kernel type`.

-We transform the input data from `actual` to `expect` if the expect kernel type is not as same as actual kernel type.
+We transform the input data from `actual` to `expect` if the actual kernel type is not as same as expect kernel type.

-The algorithm is described as follow
+The algorithm is described as following

 ```cpp
-using DataTransformationFN = std::function<void(const Tensor& in, Tensor* out)>;
-using KernelTypePair = std::pair<KernelType, KernelType>;
-
-map<KernelTypePair, DataTransformationFN> g_data_transformation_;
-
-void OpWithKernel::Run() {
-  vec<Tensor> inputs = ...
-  auto actual_kernel_type = GetActualKernelType(inputs);
-  
-  // The expected kernel type is related to actual kernel type.
-  // For the most operators, the expected kernel type is as same as
-  // actual kernel type.
-  //
-  // So we pass `actual_kernel_type` as a parameter of 
-  // GetExpectedKernelType
-  auto expect_kernel_type = GetExpectedKernelType(actual_kernel_type);
-  
-  auto trans = g_data_transformation_[{actual_kernel_type, expect_kernel_type}];
-  
-  kernel.run(trans(inputs));
+void OperatorWithKernel::Run(
+        const Scope& scope,
+        const platform::Place& place) const {
+  ExecutionContext ctx(...);
+  auto expected_kernel_key = this->GetExpectedKernelType(ctx);
+
+  Scope& new_scope = scope.NewScope();
+
+  for (auto& var_name : this->Inputs()) {
+    auto* tensor_in = GetTensor(var_name);
+    auto kernel_type_for_var = this->GetKernelTypeForVar(...);
+    if (kernel_type_for_var.place_ != expected_kernel_key.place_) {
+      auto* trans_var = new_scope.Var(var_name);
+      auto* out = DataTransform(expected_kernel_key,
+                                kernel_type_for_var,
+                                *tensor_in);
+      CopyVariableWithTensor(...);
+    }
+  }
+
+  auto kernel = kernels.find(expected_kernel_key);
+  kernel->Compute(ExecutionContext(...));
 }
 ```
+
+then the actual process for the multi-device above will be:
+
+```
+OP1(CPUPlace)
+     |
+op1_2_op2(on CPU)
+     |
+[transform](from CPU to GPU)
+     |
+op1_2_op2(on GPU)
+     |
+OP2(CUDAPlace)
+```
--- a/doc/faq/local/index_cn.rst
+++ b/doc/faq/local/index_cn.rst
@@ -211,3 +211,49 @@ decoder_inputs = paddle.layer.fc(
 * list 中元素的个数等于网络中输出层的个数；
 * list 中每个元素是一个layer的输出结果矩阵，类型是numpy的ndarray；
 * 每一个layer输出矩阵的高度，在非序列输入时：等于样本数；序列输入时等于：输入序列中元素的总数；宽度等于配置中layer的size；
+
+6.  如何在训练过程中获得某一个layer的output
+-----------------------------------------------
+
+可以在event_handler中，通过 :code:`event.gm.getLayerOutputs("layer_name")` 获得在模型配置中某一层的name :code:`layer_name` 在当前
+mini-batch forward的output的值。获得的值类型均为 :code:`numpy.ndarray` ，可以通过这个输出来完成自定义的评估指标计算等功能。例如下面代码：
+
+..      code-block:: python
+
+        def score_diff(right_score, left_score):
+            return np.average(np.abs(right_score - left_score))
+
+        def event_handler(event):
+            if isinstance(event, paddle.event.EndIteration):
+                if event.batch_id % 25 == 0:
+                    diff = score_diff(
+                        event.gm.getLayerOutputs("right_score")["right_score"][
+                            "value"],
+                        event.gm.getLayerOutputs("left_score")["left_score"][
+                            "value"])
+                    logger.info(("Pass %d Batch %d : Cost %.6f, "
+                                "average absolute diff scores: %.6f") %
+                                (event.pass_id, event.batch_id, event.cost, diff))
+
+注意：此方法不能获取 :code:`paddle.layer.recurrent_group` 里step的内容，但可以获取 :code:`paddle.layer.recurrent_group` 的输出。
+
+7.  如何在训练过程中获得参数的权重和梯度
+-----------------------------------------------
+
+在某些情况下，获得当前mini-batch的权重（或称作weights, parameters）有助于在训练时观察具体数值，方便排查以及快速定位问题。
+可以通过在 :code:`event_handler` 中打印其值（注意，需要使用 :code:`paddle.event.EndForwardBackward` 保证使用GPU训练时也可以获得），
+示例代码如下：
+
+..      code-block:: python
+
+        ...
+        parameters = paddle.parameters.create(cost)
+        ...
+        def event_handler(event):
+            if isinstance(event, paddle.event.EndForwardBackward):
+                if event.batch_id % 25 == 0:
+                    for p in parameters.keys():
+                        logger.info("Param %s, Grad %s",
+                            parameters.get(p), parameters.get_grad(p))
+
+注意：“在训练过程中获得某一个layer的output”和“在训练过程中获得参数的权重和梯度”都会造成训练中的数据从C++拷贝到numpy，会对训练性能造成影响。不要在注重性能的训练场景下使用。
\ No newline at end of file
--- a/doc/faq/local/src/reduce_min_pool_size.py
+++ b/doc/faq/local/src/reduce_min_pool_size.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
 @provider(min_pool_size=0, ...)
 def process(settings, filename):
    os.system('shuf %s > %s.shuf' % (filename, filename))  # shuffle before.

--- a/doc/faq/local/src/word2vec_config.py
+++ b/doc/faq/local/src/word2vec_config.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 ...  # the settings and define data provider is omitted.
 DICT_DIM = 3000  # dictionary dimension.
 word_ids = data_layer('word_ids', size=DICT_DIM)

--- a/doc/faq/local/src/word2vec_dataprovider.py
+++ b/doc/faq/local/src/word2vec_dataprovider.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 DICT_DIM = 3000



--- a/doc/faq/model/index_cn.rst
+++ b/doc/faq/model/index_cn.rst
@@ -67,3 +67,14 @@

  * 不同于上述介绍的recurrent layer , :code:`paddle.networks.lstmemory_unit` 定义了LSTM单元在一个时间步内的计算过程，它并不是一个完整的recurrent layer，也不能接收序列数据作为输入；
  * :code:`paddle.networks.lstmemory_unit` 只能在recurrent_group中作为step function使用；
+
+5. PaddlePaddle的softmax能否指定计算的维度
+-----------------------------------------
+
+PaddlePaddle的softmax不能指定计算维度，只能按行计算。
+在图像任务中，对于NCHW，如果需要在C维度计算softmax，可以先使用 :code:`paddle.layer.switch_order` 改变维度顺序，即将NCHW转换成NHWC，再做一定的reshape，最后计算softmax。
+
+6. PaddlePaddle是否支持维数可变的数据输入
+------------------------------------------
+
+PaddlePaddle提供的 :code:`paddle.data_type.dense_array` 支持维数可变的数据输入。在使用时，将对应数据层的维数设置成一个大于输入数据维数的值用于占位即可。
--- a/doc/getstarted/build_and_install/docker_install_cn.rst
+++ b/doc/getstarted/build_and_install/docker_install_cn.rst
@@ -25,14 +25,14 @@

  .. code-block:: bash

-     docker pull docker.paddlepaddle.org/paddle
+     docker pull docker.paddlepaddlehub.com/paddle

 下载GPU版本（cuda8.0_cudnn5_avx_mkl）的Docker镜像：

  .. code-block:: bash

     docker pull paddlepaddle/paddle:latest-gpu
-     docker pull docker.paddlepaddle.org/paddle:latest-gpu
+     docker pull docker.paddlepaddlehub.com/paddle:latest-gpu

 选择下载使用不同的BLAS库的Docker镜像：

@@ -49,7 +49,7 @@

     docker pull paddlepaddle/paddle:[tag]
     # 比如：
-     docker pull docker.paddlepaddle.org/paddle:0.10.0-gpu
+     docker pull docker.paddlepaddlehub.com/paddle:0.11.0-gpu

 .. _docker_run:


--- a/doc/getstarted/build_and_install/docker_install_en.rst
+++ b/doc/getstarted/build_and_install/docker_install_en.rst
@@ -26,14 +26,14 @@ For users in China, we provide a faster mirror:

  .. code-block:: bash

-     docker pull docker.paddlepaddle.org/paddle
+     docker pull docker.paddlepaddlehub.com/paddle

 Download GPU version (cuda8.0_cudnn5_avx_mkl) images:

  .. code-block:: bash

     docker pull paddlepaddle/paddle:latest-gpu
-     docker pull docker.paddlepaddle.org/paddle:latest-gpu
+     docker pull docker.paddlepaddlehub.com/paddle:latest-gpu

 Choose between different BLAS version:

@@ -53,7 +53,7 @@ and run:

     docker pull paddlepaddle/paddle:[tag]
     # i.e.
-     docker pull docker.paddlepaddle.org/paddle:0.10.0-gpu
+     docker pull docker.paddlepaddlehub.com/paddle:0.11.0-gpu

 .. _docker_run:


--- a/doc/getstarted/concepts/src/infer.py
+++ b/doc/getstarted/concepts/src/infer.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 import paddle.v2 as paddle
 import numpy as np


--- a/doc/getstarted/concepts/src/train.py
+++ b/doc/getstarted/concepts/src/train.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 import paddle.v2 as paddle
 import numpy as np


--- a/doc/howto/dev/new_op_en.md
+++ b/doc/howto/dev/new_op_en.md
@@ -4,7 +4,8 @@
 - [Implementing C++ Types](#implementing-c-types)
   - [Defining ProtoMaker](#defining-protomaker)
   - [Defining Operator](#defining-operator)
-   - [Registering Operator](#registering-operator)
+   - [Defining OpKernel](#defining-opkernel)
+   - [Registering Operator and OpKernel](#registering-operator-and-opkernel)
   - [Compilation](#compilation)
 - [Python Binding](#python-binding)
 - [Unit Tests](#unit-tests)
@@ -16,12 +17,13 @@

 Here are the base types needed. For details, please refer to the design docs.

- `framework::OperatorBase`: Operator (Op)base class.
- `framework::OpKernel`: Base class for Op computation.
- `framework::OperatorWithKernel`: Inherited from OperatorBase, describing an operator with computation.
 - `class OpProtoAndCheckerMaker`: Describes an Operator's input, output, attributes and description, mainly used to interface with Python API.
+- `framework::OperatorBase`: Operator (Op)base class.
+- `framework::OpKernel`: Base class for Op computation kernel.
+- `framework::OperatorWithKernel`: Inherited from OperatorBase, describing an operator with computation kernels.
+

-An operator can be differentiated by whether in has kernel methods. An operator with kernel inherits from `OperatorWithKernel` while the ones without inherit from `OperatorBase`. This tutorial focuses on implementing operators with kernels. In short, an operator includes the following information:
+Operators can be categorized into two groups: operator with kernel(s) and operator without kernel(s). An operator with kernel(s) inherits from `OperatorWithKernel` while the one without kernel(s) inherits from `OperatorBase`. This tutorial focuses on implementing operators with kernels. In short, an operator includes the following information:


 Information           | Where is it defined
@@ -32,7 +34,7 @@ Kernel implementation       | The kernel methods shared between CPU and CUDA are
 Registering the Op           | Ops are registered in `.cc` files; For Kernel registration, `.cc` files contain the CPU implementation, while `.cu` files contain the CUDA implementation.


-New Operator implementations are added to the list [paddle/operators](https://github.com/PaddlePaddle/Paddle/tree/develop/paddle/operators), with file names in the format `*_op.h` (if applicable), `*_op.cc`, `*_op.cu` (if applicable).** The system will use the naming scheme to automatically build operators and their corresponding Python extensions. **
+New Operator implementations are added to the list [paddle/operators](https://github.com/PaddlePaddle/Paddle/tree/develop/paddle/operators), with file names in the format `*_op.h` (if applicable), `*_op.cc`, `*_op.cu` (if applicable).** The system will use the naming scheme to automatically build operators and their corresponding Python extensions.**


 Let's take matrix multiplication operator, [MulOp](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/operators/mul_op.cc), as an example to introduce the writing of an Operator with Kernel.
@@ -156,7 +158,8 @@ Usually `OpProtoMaker` and `Op`'s type definitions are written in `.cc` files, w
 - `typename T` denotes data type, such as `float` or `double`.

 `MulKernel` types need to rewrite the interface for `Compute`.
- `Compute` takes one input variable `const framework::ExecutionContext& context`.
+
+- `Compute` takes one input parameter: `const framework::ExecutionContext& context`.
 - Compared with `InferShapeContext`, `ExecutionContext` includes device types, and can similarly extract input, output, and attribute variables.
 - `Compute` implements the computation logics of an `OpKernel`.

@@ -177,7 +180,7 @@ Usually `OpProtoMaker` and `Op`'s type definitions are written in `.cc` files, w
  };
  ```

-Note that **different devices (CPU, CUDA)share an Op definition; whether or not they share the same `OpKernel` depends on whether `Compute` calls functions that support both devices.**
+Note that **different devices (CPU, CUDA)share one Op definition; whether or not they share the same `OpKernel` depends on whether `Compute` calls functions can support both devices.**

 `MulOp`'s CPU and CUDA share the same `Kernel`. A non-sharing  `OpKernel` example can be seen in [`OnehotCrossEntropyOpKernel`](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/operators/cross_entropy_op.h#L43).

@@ -188,13 +191,14 @@ This concludes the forward implementation of an operator. Next its operation and

 The definition of its corresponding backward operator, if applicable, is similar to that of an forward operator. **Note that a backward operator does not include a `ProtoMaker`**.

-### Registering Operator
+### Registering Operator and OpKernel

 - In `.cc` files, register forward and backward operator classes and the CPU kernel.

    ```cpp
    namespace ops = paddle::operators;
    REGISTER_OP(mul, ops::MulOp, ops::MulOpMaker, mul_grad, ops::MulOpGrad);
+
    REGISTER_OP_CPU_KERNEL(mul, ops::MulKernel<paddle::platform::CPUDeviceContext, float>);
    REGISTER_OP_CPU_KERNEL(mul_grad,
                  ops::MulGradKernel<paddle::platform::CPUDeviceContext, float>);
@@ -204,6 +208,7 @@ The definition of its corresponding backward operator, if applicable, is similar

    - `REGISTER_OP` registers the `ops::MulOp` class, type named `mul`, its type `ProtoMaker` is `ops::MulOpMaker`, registering `ops::MulOpGrad` as `mul_grad`.
    - `REGISTER_OP_WITHOUT_GRADIENT` registers an operator without gradient.
+
    - `REGISTER_OP_CPU_KERNEL` registers `ops::MulKernel` class and specialized template types `paddle::platform::CPUPlace` and `float`, which also registers `ops::MulGradKernel`.


@@ -225,6 +230,7 @@ The definition of its corresponding backward operator, if applicable, is similar
 Run the following commands to compile.

 ```
+# maybe you need to rerun cmake
 make mul_op
 ```


--- a/doc/howto/dev/new_op_kernel_en.md
+++ b/doc/howto/dev/new_op_kernel_en.md
+## Add Kernels for a New Device
+
+### Background
+
+PaddlePaddle Fluid have hundreds of operators.  Each operator could have one or more kernels.  A kernel is an implementation of the operator for a certain device, which could be a hardware device, e.g., the CUDA GPU, or a library that utilizes a device, e.g., Intel MKL that makes full use of the Xeon CPU.
+
+[This document](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/howto/dev/new_op_en.md) explains how to add an operator, and its kernels.  The kernels of an operator are indexed by a C++ type [`OpKernelType`](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/operator_kernel_type.md).  An operator chooses the right kernel at runtime.  This choosing mechanism is described [here](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/switch_kernel.md).
+
+### Write Kernels for A New Device 
+
+#### Add A New Device
+
+  For some historical reaons, we misuse the word *library* for *device*.  For example, we call the deivce type by *library type*.  An example is the header file [`library_type.h`](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/framework/library_type.h#L24).  We will correct this ASAP.
+
+To register a new device, we need to add an enum value to `LibraryType`:
+
+```
+enum class LibraryType {
+  kPlain = 0,
+  kMKLDNN = 1,
+  kCUDNN = 2,
+};
+```
+
+
+#### Add A New [Place](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/place.h#L53)
+
+If you have a new kind of Device, firstly you need to add a new kind of [`Place`](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/place.h#L53). For example `CUDAPlace`:
+
+```cpp
+struct CUDAPlace {
+  CUDAPlace() : CUDAPlace(0) {}
+  explicit CUDAPlace(int d) : device(d) {}
+
+  inline int GetDeviceId() const { return device; }
+  // needed for variant equality comparison
+  inline bool operator==(const CUDAPlace &o) const {
+    return device == o.device;
+  }
+  inline bool operator!=(const CUDAPlace &o) const { return !(*this == o); }
+
+  int device;
+};
+
+typedef boost::variant<CUDAPlace, CPUPlace> Place;
+```
+
+#### Add [device context]((https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/device_context.h#L37))
+After a new kind of Device is added, you should add a corresponding [DeviceContext](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/platform/device_context.h#L37) for it.
+
+```cpp
+class DeviceContext {
+ public:
+  virtual ~DeviceContext() {}
+  virtual Place GetPlace() const = 0;
+
+  virtual void Wait() const {}
+};
+```
+
+#### Implement new [OpKernel](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/framework/operator.h#L351) for your Device.
+
+A detailed documentation can be found in [`new_op_and_kernel`](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/howto/dev/new_op_en.md)
+
+```cpp
+class OpKernelBase {
+ public:
+  /**
+   * ExecutionContext is the only parameter of Kernel Run function.
+   * Run will get input/output variables, state such as momentum and
+   * device resource such as CUDA stream, cublas handle, etc. from
+   * ExecutionContext. User should construct it before run the Operator.
+   */
+
+  virtual void Compute(const ExecutionContext& context) const = 0;
+
+  virtual ~OpKernelBase() = default;
+};
+
+template <typename T>
+class OpKernel : public OpKernelBase {
+ public:
+  using ELEMENT_TYPE = T;
+};
+```
+
+
+#### Register the OpKernel to framework
+
+After writing the components described above, we should register the kernel to the framework.
+
+We use `REGISTER_OP_KERNEL` to do the registration.
+
+```cpp
+REGISTER_OP_KERNEL(
+	op_type,
+	library_type,
+	place_type,
+	kernel0, kernel1, ...)
+```
+
+kernel0, kernel1 are kernels that have the same `op_type`, `library_type`, `place_type` but different `data_types`.
+
+take [`conv2d`]((https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/operators/conv_cudnn_op.cu.cc#L318)) as an example:
+
+	```cpp
+	REGISTER_OP_KERNEL(conv2d, CPU, paddle::platform::CPUPlace,
+    		paddle::operators::GemmConvKernel<paddle::platform::CPUDeviceContext, float>,
+    		paddle::operators::GemmConvKernel<paddle::platform::CPUDeviceContext, double>);
+    
+	REGISTER_OP_KERNEL(conv2d, CUDNN, ::paddle::platform::CUDAPlace,
+	       paddle::operators::CUDNNConvOpKernel<float>,
+	       paddle::operators::CUDNNConvOpKernel<double>);
+	```
+
+In the code above:
+
+ - `conv2d` is the type/name of the operator
+ - `CUDNN/CPU` is `library`
+ - `paddle::platform::CUDAPlace/CPUPlace` is `place`
+ - template parameter `float/double` on `CUDNNConvOpKernel<T>` is `data_type`.
--- a/doc/howto/optimization/cpu_profiling.md
+++ b/doc/howto/optimization/cpu_profiling.md
@@ -60,8 +60,7 @@ each column is as follows:
 | column | meaning |
 | --- | --- |
 | ncalls | the number of calls into a function |
-| tottime | the total execution time of the function, not including the
- execution time of other functions called by the function |
+| tottime | the total execution time of the function, not including the execution time of other functions called by the function |
 | percall | tottime divided by ncalls |
 | cumtime | the total execution time of the function, including the execution time of other functions being called |
 | percall | cumtime divided by ncalls |

--- a/doc/howto/usage/cluster/fluid_cluster_train_en.md
+++ b/doc/howto/usage/cluster/fluid_cluster_train_en.md
@@ -2,27 +2,27 @@

 ## Introduction

-In this article, we'll explain how to config and run distributed training jobs with PaddlePaddle Fluid in a bare metal cluster.
+In this article, we'll explain how to configure and run distributed training jobs with PaddlePaddle Fluid in a bare metal cluster.

 ## Preparations

-### Get your cluster ready
+### Getting the cluster ready

-Prepare your computer nodes in the cluster. Nodes in this cluster can be of any specification that runs PaddlePaddle, and with a unique IP address assigned to it. Make sure they can communicate with each other.
+Prepare the compute nodes in the cluster. Nodes in this cluster can be of any specification that runs PaddlePaddle, and with a unique IP address assigned to it. Make sure they can communicate to each other.

 ### Have PaddlePaddle installed

 PaddlePaddle must be installed on all nodes. If you have GPU cards on your nodes, be sure to properly install drivers and CUDA libraries.

-PaddlePaddle build and installation guide can be found from [here](http://www.paddlepaddle.org/docs/develop/documentation/en/getstarted/build_and_install/index_en.html).
+PaddlePaddle build and installation guide can be found  [here](http://www.paddlepaddle.org/docs/develop/documentation/en/getstarted/build_and_install/index_en.html).

-### Update training script
+### Update the training script

 #### Non-cluster training script

 Let's take [Deep Learning 101](http://www.paddlepaddle.org/docs/develop/book/01.fit_a_line/index.html)'s first chapter: "fit a line" as an example.

-This demo's non-cluster version with fluid API is as follows:
+The non-cluster version of this demo with fluid API is as follows:

 ``` python
 import paddle.v2 as paddle
@@ -65,25 +65,25 @@ for pass_id in range(PASS_NUM):
 exit(1)
 ```

-We created a simple fully connected neural networks training program and handed it to the fluid executor to run for 100 passes.
+We created a simple fully-connected neural network training program and handed it to the fluid executor to run for 100 passes.

-Now let's try to convert it to a distributed version to run in a cluster.
+Now let's try to convert it to a distributed version to run on a cluster.

 #### Introducing parameter server

-As you see from the non-cluster version of training script, there is only one role in it: the trainer, who does the computing as well as holding parameters. In cluster training, since multi-trainers are working on the same task, they need one centralized place to hold and distribute parameters. This centralized place is called the Parameter Server in PaddlePaddle.
+As we can see from the non-cluster version of training script, there is only one role in the script: the trainer, that performs the computing as well as holds the parameters. In cluster training, since multi-trainers are working on the same task, they need one centralized place to hold and distribute parameters. This centralized place is called the Parameter Server in PaddlePaddle.

-![parameter server architect](src/trainer.png)
+![parameter server architecture](src/trainer.png)

-Parameter Server in fluid does not only hold parameters but is also assigned with a part of the program. Trainers communicate with parameter servers via send/receive OPs. For more tech detail, please refer to this [document](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/dist_refactor/distributed_architecture.md).
+Parameter Server in fluid not only holds the parameters but is also assigned with a part of the program. Trainers communicate with parameter servers via send/receive OPs. For more technical details, please refer to  [this document](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/design/dist_refactor/distributed_architecture.md).

-Now we need to create program for both trainers and parameter servers, the question is how?
+Now we need to create programs for both: trainers and parameter servers, the question is how?

 #### Slice the program

-Fluid provides a tool called "Distribute Transpiler" to automatically convert the non-cluster program into cluster program.
+Fluid provides a tool called "Distributed Transpiler" that automatically converts the non-cluster program into cluster program.

-The idea behind this tool is to find optimize OPs and gradient parameters, slice the program into 2 pieces and connect them with send/receive OP.
+The idea behind this tool is to find the optimize OPs and gradient parameters, slice the program into 2 pieces and connect them with send/receive OP.

 Optimize OPs and gradient parameters can be found from the return values of optimizer's minimize function.

@@ -94,16 +94,18 @@ To put them together:

 optimize_ops, params_grads = sgd_optimizer.minimize(avg_cost) #get optimize OPs and gradient parameters

-t = fluid.DistributeTranspiler() # create transpiler instance
+t = fluid.DistributeTranspiler() # create the transpiler instance
 # slice the program into 2 pieces with optimizer_ops and gradient parameters list, as well as pserver_endpoints, which is a comma separated list of [IP:PORT] and number of trainers
-t.transpile(optimize_ops, params_grads, pservers=pserver_endpoints, trainers=2) 
+t.transpile(optimize_ops, params_grads, pservers=pserver_endpoints, trainers=2)

 ... #create executor

 # in pserver, run this
-exe.run(fluid.default_startup_program())
 #current_endpoint here means current pserver IP:PORT you wish to run on
-exe.run(t.get_pserver_program(current_endpoint, optimize_ops)) 
+pserver_prog = t.get_pserver_program(current_endpoint)
+pserver_startup = t.get_startup_program(current_endpoint, pserver_prog)
+exe.run(pserver_startup)
+exe.run(pserver_prog)

 # in trainer, run this
 ... # define data reader
@@ -117,7 +119,7 @@ for pass_id in range(100):

 ### E2E demo

-Please find the complete demo from [here](https://github.com/PaddlePaddle/Paddle/blob/develop/python/paddle/v2/fluid/tests/book_distribute/notest_dist_fit_a_line.py). In parameter server node run this in the command line:
+Please find the complete demo from [here](https://github.com/PaddlePaddle/Paddle/blob/develop/python/paddle/v2/fluid/tests/book_distribute/notest_dist_fit_a_line.py). In parameter server node run the following in the command line:

 ``` bash
 PSERVERS=192.168.1.2:6174 SERVER_ENDPOINT=192.168.1.2:6174 TRAINING_ROLE=PSERVER python notest_dist_fit_a_line.py
@@ -127,12 +129,12 @@ PSERVERS=192.168.1.2:6174 SERVER_ENDPOINT=192.168.1.2:6174 TRAINING_ROLE=PSERVER

 Wait until the prompt `Server listening on 192.168.1.2:6174`

-Then in 2 of your trainer node run this:
+Then in 2 of your trainer nodes run this:

 ``` bash
 PSERVERS=192.168.1.2:6174 SERVER_ENDPOINT=192.168.1.2:6174 TRAINING_ROLE=TRAINER python notest_dist_fit_a_line.py
 ```

-*the reason you need to run this command twice in 2 nodes is: in the script we set the trainer count to be 2. You can change this setting on line 50*
+*the reason you need to run this command twice in 2 nodes is because: in the script we set the trainer count to be 2. You can change this setting on line 50*

 Now you have 2 trainers and 1 parameter server up and running.
--- a/doc/howto/usage/cluster/src/k8s_train/start_paddle.py
+++ b/doc/howto/usage/cluster/src/k8s_train/start_paddle.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
-#
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
-#
-#    http://www.apache.org/licenses/LICENSE-2.0
-#
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
 #!/usr/bin/python
 # Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved
 #

--- a/doc/howto/usage/cluster/src/word2vec/api_train_v2.py
+++ b/doc/howto/usage/cluster/src/word2vec/api_train_v2.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 import gzip
 import math


--- a/doc/howto/usage/cluster/src/word2vec/api_train_v2_cluster.py
+++ b/doc/howto/usage/cluster/src/word2vec/api_train_v2_cluster.py
-#  Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
 #
-#Licensed under the Apache License, Version 2.0 (the "License");
-#you may not use this file except in compliance with the License.
-#You may obtain a copy of the License at
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
 #
-#    http://www.apache.org/licenses/LICENSE-2.0
+#     http://www.apache.org/licenses/LICENSE-2.0
 #
-#Unless required by applicable law or agreed to in writing, software
-#distributed under the License is distributed on an "AS IS" BASIS,
-#WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-#See the License for the specific language governing permissions and
-#limitations under the License.
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 import math
 import os
 import paddle.v2 as paddle

--- a/doc/howto/usage/cluster/src/word2vec/prepare.py
+++ b/doc/howto/usage/cluster/src/word2vec/prepare.py
+#   Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserve.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 import paddle.v2 as paddle
 import tarfile
 import os

--- a/doc/v1_api_tutorials/README.md
+++ b/doc/v1_api_tutorials/README.md
-The tutorials in v1_api_tutorials are using v1_api currently, and will be upgraded to v2_api later.
-Thus, v1_api_tutorials is a temporary directory. We decide not to maintain it and will delete it in future.
-
-Please go to [PaddlePaddle/book](https://github.com/PaddlePaddle/book) and 
-[PaddlePaddle/models](https://github.com/PaddlePaddle/models) to learn PaddlePaddle.
--- a/doc/v1_api_tutorials/embedding_model/index_cn.md
+++ b/doc/v1_api_tutorials/embedding_model/index_cn.md
-# 中文词向量模型的使用 #
----------
-本文档介绍如何在PaddlePaddle平台上,使用预训练的标准格式词向量模型。
-
-在此感谢 @lipeng 提出的代码需求，并给出的相关模型格式的定义。
-
-## 介绍 ###
-### 中文字典 ###
-我们的字典使用内部的分词工具对百度知道和百度百科的语料进行分词后产生。分词风格如下： "《红楼梦》"将被分为 "《"，"红楼梦"，"》"，和 "《红楼梦》"。字典采用UTF8编码，输出有2列：词本身和词频。字典共包含 3206326个词和4个特殊标记：
-  - `<s>`: 分词序列的开始
-  - `<e>`: 分词序列的结束
-  - `PALCEHOLDER_JUST_IGNORE_THE_EMBEDDING`: 占位符，没有实际意义
-  - `<unk>`: 未知词
-
-### 中文词向量的预训练模型 ###
-遵循文章 [A Neural Probabilistic Language Model](http://www.jmlr.org/papers/volume3/bengio03a/bengio03a.pdf)中介绍的方法，模型采用 n-gram 语言模型，结构如下图：6元上下文作为输入层->全连接层->softmax层 。对应于字典，我们预训练得到4种不同维度的词向量，分别为：32维、64维、128维和256维。
-<center>![](./neural-n-gram-model.png)</center>
-<center>Figure 1. neural-n-gram-model</center>
-
-### 下载和数据抽取 ###
-运行以下的命令下载和获取我们的字典和预训练模型：
-
-    cd $PADDLE_ROOT/demo/model_zoo/embedding
-    ./pre_DictAndModel.sh
-
-## 中文短语改写的例子 ##
-以下示范如何使用预训练的中文字典和词向量进行短语改写。
-
-### 数据的准备和预处理 ###
-首先，运行以下的命令下载数据集。该数据集（utf8编码）包含20个训练样例，5个测试样例和2个生成式样例。
-
-    cd $PADDLE_ROOT/demo/seqToseq/data
-    ./paraphrase_data.sh
-
-第二步，将数据处理成规范格式，在训练数集上训练生成词向量字典（数据将保存在 `$PADDLE_SOURCE_ROOT/demo/seqToseq/data/pre-paraphrase`）:
-
-    cd $PADDLE_ROOT/demo/seqToseq/
-    python preprocess.py -i data/paraphrase [--mergeDict]
-
- 其中，如果使用`--mergeDict`选项，源语言短语和目标语言短语的字典将被合并（源语言和目标语言共享相同的编码字典）。本实例中，源语言和目标语言都是相同的语言，因此可以使用该选项。
-
-
-### 使用用户指定的词向量字典 ###
-使用如下命令，从预训练模型中，根据用户指定的字典，抽取对应的词向量构成新的词表:
-    cd $PADDLE_ROOT/demo/model_zoo/embedding
-    python extract_para.py --preModel PREMODEL --preDict PREDICT --usrModel USRMODEL--usrDict USRDICT -d DIM
-
- `--preModel PREMODEL`: 预训练词向量字典模型的路径
- `--preDict PREDICT`:  预训练模型使用的字典的路径
- `--usrModel USRMODEL`: 抽取出的新词表的保存路径
- `--usrDict USRDICT`: 用户指定新的字典的路径，用于构成新的词表
- `-d DIM`: 参数（词向量）的维度
-
-此处，你也可以简单的运行以下的命令：
-
-    cd $PADDLE_ROOT/demo/seqToseq/data/
-    ./paraphrase_model.sh
-
-运行成功以后，你将会看到以下的模型结构：
-
-    paraphrase_model
-    |--- _source_language_embedding
-    |--- _target_language_embedding
-
-### 在PaddlePaddle平台训练模型 ###
-首先，配置模型文件，配置如下（可以参考保存在 `demo/seqToseq/paraphrase/train.conf`的配置）:
-
-    from seqToseq_net import *
-    is_generating = False
-
-    ################## Data Definition #####################
-    train_conf = seq_to_seq_data(data_dir = "./data/pre-paraphrase",
-                                 job_mode = job_mode)
-
-    ############## Algorithm Configuration ##################
-    settings(
-          learning_method = AdamOptimizer(),
-          batch_size = 50,
-          learning_rate = 5e-4)
-
-    ################# Network configure #####################
-    gru_encoder_decoder(train_conf, is_generating, word_vector_dim = 32)
-
-这个配置与`demo/seqToseq/translation/train.conf` 基本相同
-
-然后，使用以下命令进行模型训练:
-
-    cd $PADDLE_SOURCE_ROOT/demo/seqToseq/paraphrase
-    ./train.sh
-
-其中，`train.sh` 与`demo/seqToseq/translation/train.sh` 基本相同，只有2个配置不一样:
-
- `--init_model_path`: 初始化模型的路径配置为`data/paraphrase_modeldata/paraphrase_model`
- `--load_missing_parameter_strategy`：如果参数模型文件缺失，除词向量模型外的参数将使用正态分布随机初始化
-
-如果用户想要了解详细的数据集的格式、模型的结构和训练过程，请查看 [Text generation Tutorial](../text_generation/index_cn.md).
-
-## 可选功能 ##
-###  观测词向量
-PaddlePaddle 平台为想观测词向量的用户提供了将二进制词向量模型转换为文本模型的功能:
-
-    cd $PADDLE_ROOT/demo/model_zoo/embedding
-    python paraconvert.py --b2t -i INPUT -o OUTPUT -d DIM
-
- `-i INPUT`: 输入的（二进制）词向量模型名称
- `-o OUTPUT`: 输出的文本模型名称
- `-d DIM`: （词向量）参数维度
-
-运行完以上命令，用户可以在输出的文本模型中看到:
-
-    0,4,32156096
-    -0.7845433,1.1937413,-0.1704215,0.4154715,0.9566584,-0.5558153,-0.2503305, ......
-    0.0000909,0.0009465,-0.0008813,-0.0008428,0.0007879,0.0000183,0.0001984, ......
-    ......
-
- 其中，第一行是`PaddlePaddle` 输出文件的格式说明，包含3个属性：:
-  - `PaddlePaddle`的版本号，本例中为0
-  - 浮点数占用的字节数，本例中为4
-  - 总计的参数个数，本例中为32,156,096
- 其余行是（词向量）参数行（假设词向量维度为32）
-  - 每行打印32个参数以','分隔
-  - 共有32,156,096/32 = 1,004,877行，也就是说，模型共包含1,004,877个被向量化的词
-
-### 词向量模型的修正
-`PaddlePaddle` 为想修正词向量模型的用户提供了将文本词向量模型转换为二进制模型的命令:
-
-    cd $PADDLE_ROOT/demo/model_zoo/embedding
-    python paraconvert.py --t2b -i INPUT -o OUTPUT
-
- `-i INPUT`: 输入的文本词向量模型名称
- `-o OUTPUT`: 输出的二进制词向量模型名称
-
-请注意，输入的文本格式如下:
-
-    -0.7845433,1.1937413,-0.1704215,0.4154715,0.9566584,-0.5558153,-0.2503305, ......
-    0.0000909,0.0009465,-0.0008813,-0.0008428,0.0007879,0.0000183,0.0001984, ......
-    ......
- 输入文本中没有头部（格式说明）行
- （输入文本）每行存储一个词，以逗号','分隔
--- a/doc/v1_api_tutorials/embedding_model/index_en.md
+++ b/doc/v1_api_tutorials/embedding_model/index_en.md
-# Chinese Word Embedding Model Tutorial #
----------
-This tutorial is to guide you through the process of using a Pretrained Chinese Word Embedding Model in the PaddlePaddle standard format.
-
-We thank @lipeng for the pull request that defined the model schemas and pretrained the models.
-
-## Introduction ###
-### Chinese Word Dictionary ###
-Our Chinese-word dictionary is created on Baidu ZhiDao and Baidu Baike by using in-house word segmentor. For example, the participle of "《红楼梦》" is "《"，"红楼梦"，"》"，and "《红楼梦》". Our dictionary (using UTF-8 format) has has two columns: word and its frequency. The total word count is 3206326, including 4 special token:
-  - `<s>`: the start of a sequence
-  - `<e>`: the end of a sequence
-  - `PALCEHOLDER_JUST_IGNORE_THE_EMBEDDING`: a placeholder, just ignore it and its embedding
-  - `<unk>`: a word not included in dictionary
-
-### Pretrained Chinese Word Embedding Model ###
-Inspired by paper [A Neural Probabilistic Language Model](http://www.jmlr.org/papers/volume3/bengio03a/bengio03a.pdf), our model architecture (**Embedding joint of six words->FullyConnect->SoftMax**) is as following graph. And for our dictionary, we pretrain four models with different word vector dimenstions, i.e 32, 64, 128, 256.
-<center>![](./neural-n-gram-model.png)</center>
-<center>Figure 1. neural-n-gram-model</center>
-
-### Download and Extract ###
-To download and extract our dictionary and pretrained model, run the following commands.
-
-    cd $PADDLE_ROOT/demo/model_zoo/embedding
-    ./pre_DictAndModel.sh
-
-## Chinese Paraphrasing Example ##
-We provide a paraphrasing task to show the usage of pretrained Chinese Word Dictionary and Embedding Model.
-
-### Data Preparation and Preprocess ###
-
-First, run the following commands to download and extract the in-house dataset. The dataset (using UTF-8 format) has 20 training samples, 5 testing samples and 2 generating samples.
-
-    cd $PADDLE_ROOT/demo/seqToseq/data
-    ./paraphrase_data.sh
-
-Second, preprocess data and build dictionary on train data by running the following commands, and the preprocessed dataset is stored in `$PADDLE_SOURCE_ROOT/demo/seqToseq/data/pre-paraphrase`:
-
-    cd $PADDLE_ROOT/demo/seqToseq/
-    python preprocess.py -i data/paraphrase [--mergeDict]
-
- `--mergeDict`: if using this option, the source and target dictionary are merged, i.e, two dictionaries have the same context. Here, as source and target data are all chinese words, this option can be used.
-
-### User Specified Embedding Model ###
-The general command of extracting desired parameters from the pretrained embedding model based on user dictionary is:
-
-    cd $PADDLE_ROOT/demo/model_zoo/embedding
-    python extract_para.py --preModel PREMODEL --preDict PREDICT --usrModel USRMODEL--usrDict USRDICT -d DIM
-
- `--preModel PREMODEL`: the name of pretrained embedding model
- `--preDict PREDICT`: the name of pretrained dictionary
- `--usrModel USRMODEL`: the name of extracted embedding model
- `--usrDict USRDICT`: the name of user specified dictionary
- `-d DIM`: dimension of parameter
-
-Here, you can simply run the command:
-
-    cd $PADDLE_ROOT/demo/seqToseq/data/
-    ./paraphrase_model.sh
-
-And you will see following embedding model structure:
-
-    paraphrase_model
-    |--- _source_language_embedding
-    |--- _target_language_embedding
-
-### Training Model in PaddlePaddle ###
-First, create a model config file, see example `demo/seqToseq/paraphrase/train.conf`:
-
-    from seqToseq_net import *
-    is_generating = False
-
-    ################## Data Definition #####################
-    train_conf = seq_to_seq_data(data_dir = "./data/pre-paraphrase",
-                                 job_mode = job_mode)
-
-    ############## Algorithm Configuration ##################
-    settings(
-          learning_method = AdamOptimizer(),
-          batch_size = 50,
-          learning_rate = 5e-4)
-
-    ################# Network configure #####################
-    gru_encoder_decoder(train_conf, is_generating, word_vector_dim = 32)
-
-This config is almost the same as `demo/seqToseq/translation/train.conf`.
-
-Then, train the model by running the command:
-
-    cd $PADDLE_SOURCE_ROOT/demo/seqToseq/paraphrase
-    ./train.sh
-
-where `train.sh` is almost the same as `demo/seqToseq/translation/train.sh`, the only difference is following two command arguments:
-
- `--init_model_path`: path of the initialization model, here is `data/paraphrase_model`
- `--load_missing_parameter_strategy`: operations when model file is missing, here use a normal distibution to initialize the other parameters except for the embedding layer
-
-For users who want to understand the dataset format, model architecture and training procedure in detail, please refer to [Text generation Tutorial](../text_generation/index_en.md).
-
-## Optional Function ##
-###  Embedding Parameters Observation
-For users who want to observe the embedding parameters, this function can convert a PaddlePaddle binary embedding model to a text model by running the command:
-
-    cd $PADDLE_ROOT/demo/model_zoo/embedding
-    python paraconvert.py --b2t -i INPUT -o OUTPUT -d DIM
-
- `-i INPUT`: the name of input binary embedding model
- `-o OUTPUT`: the name of output text embedding model
- `-d DIM`: the dimension of parameter
-
-You will see parameters like this in output text model:
-
-    0,4,32156096
-    -0.7845433,1.1937413,-0.1704215,0.4154715,0.9566584,-0.5558153,-0.2503305, ......
-    0.0000909,0.0009465,-0.0008813,-0.0008428,0.0007879,0.0000183,0.0001984, ......
-    ......
-
- 1st line is **PaddlePaddle format file head**, it has 3 attributes:
-  - version of PaddlePaddle, here is 0
-  - sizeof(float), here is 4
-  - total number of parameter, here is 32156096
- Other lines print the paramters (assume `<dim>` = 32)
-  - each line print 32 paramters splitted by ','
-  - there is 32156096/32 = 1004877 lines, meaning there is 1004877 embedding words
-
-### Embedding Parameters Revision
-For users who want to revise the embedding parameters, this function can convert a revised text embedding model to a PaddlePaddle binary model by running the command:
-
-    cd $PADDLE_ROOT/demo/model_zoo/embedding
-    python paraconvert.py --t2b -i INPUT -o OUTPUT
-
- `-i INPUT`: the name of input text embedding model.
- `-o OUTPUT`: the name of output binary embedding model
-
-Note that the format of input text model is as follows:
-
-    -0.7845433,1.1937413,-0.1704215,0.4154715,0.9566584,-0.5558153,-0.2503305, ......
-    0.0000909,0.0009465,-0.0008813,-0.0008428,0.0007879,0.0000183,0.0001984, ......
-    ......
- there is no file header in 1st line
- each line stores parameters for one word, the separator is commas ','
--- a/doc/v1_api_tutorials/embedding_model/neural-n-gram-model.png
+++ b/doc/v1_api_tutorials/embedding_model/neural-n-gram-model.png
--- a/doc/v1_api_tutorials/gan/gan.png
+++ b/doc/v1_api_tutorials/gan/gan.png
--- a/doc/v1_api_tutorials/gan/index_en.md
+++ b/doc/v1_api_tutorials/gan/index_en.md
-# Generative Adversarial Networks (GAN) 
-
-This demo implements GAN training described in the original [GAN paper](https://arxiv.org/abs/1406.2661) and deep convolutional generative adversarial networks [DCGAN paper](https://arxiv.org/abs/1511.06434).
-
-The high-level structure of GAN is shown in Figure. 1 below. It is composed of two major parts: a generator and a discriminator, both of which are based on neural networks. The generator takes in some kind of noise with a known distribution and transforms it into an image. The discriminator takes in an image and determines whether it is artificially generated by the generator or a real image. So the generator and the discriminator are in a competitive game in which generator is trying to generate image to look as real as possible to fool the discriminator, while the discriminator is trying to distinguish between real and fake images. 
-
-<center>![](./gan.png)</center>
-<p align="center">
-    Figure 1. GAN-Model-Structure
-    <a href="https://ishmaelbelghazi.github.io/ALI/">figure credit</a>
-</p>
-
-The generator and discriminator take turn to be trained using SGD. The objective function of the generator is for its generated images being classified as real by the discriminator, and the objective function of the discriminator is to correctly classify real and fake images. When the GAN model is trained to converge to the equilibrium state, the generator will transform the given noise distribution to the distribution of real images, and the discriminator will not be able to distinguish between real and fake images at all. 
-
-## Implementation of GAN Model Structure
-Since GAN model involves multiple neural networks, it requires to use paddle python API. So the code walk-through below can also partially serve as an introduction to the usage of Paddle Python API.
-
-There are three networks defined in gan_conf.py, namely **generator_training**, **discriminator_training** and **generator**. The relationship to the model structure we defined above is that **discriminator_training** is the discriminator, **generator** is the generator, and the **generator_training** combined the generator and discriminator since training generator would require the discriminator to provide loss function. This relationship is described in the following code:
-```python
-if is_generator_training:
-    noise = data_layer(name="noise", size=noise_dim)
-    sample = generator(noise)
-
-if is_discriminator_training:
-    sample = data_layer(name="sample", size=sample_dim)
-
-if is_generator_training or is_discriminator_training:
-    label = data_layer(name="label", size=1)
-    prob = discriminator(sample)
-    cost = cross_entropy(input=prob, label=label)
-    classification_error_evaluator(
-        input=prob, label=label, name=mode + '_error')
-    outputs(cost)
-
-if is_generator:
-    noise = data_layer(name="noise", size=noise_dim)
-    outputs(generator(noise))
-```
-
-In order to train the networks defined in gan_conf.py, one first needs to initialize a Paddle environment, parse the config, create GradientMachine from the config and create trainer from GradientMachine as done in the code chunk below:
-```python
-import py_paddle.swig_paddle as api
-# init paddle environment
-api.initPaddle('--use_gpu=' + use_gpu, '--dot_period=10',
-               '--log_period=100', '--gpu_id=' + args.gpu_id,
-               '--save_dir=' + "./%s_params/" % data_source)
-
-# Parse config
-gen_conf = parse_config(conf, "mode=generator_training,data=" + data_source)
-dis_conf = parse_config(conf, "mode=discriminator_training,data=" + data_source)
-generator_conf = parse_config(conf, "mode=generator,data=" + data_source)
-
-# Create GradientMachine
-dis_training_machine = api.GradientMachine.createFromConfigProto(
-dis_conf.model_config)
-gen_training_machine = api.GradientMachine.createFromConfigProto(
-gen_conf.model_config)
-generator_machine = api.GradientMachine.createFromConfigProto(
-generator_conf.model_config)
-
-# Create trainer
-dis_trainer = api.Trainer.create(dis_conf, dis_training_machine)
-gen_trainer = api.Trainer.create(gen_conf, gen_training_machine)
-```
-
-In order to balance the strength between generator and discriminator, we schedule to train whichever one is performing worse by comparing their loss function value. The loss function value can be calculated by a forward pass through the GradientMachine.
-```python
-def get_training_loss(training_machine, inputs):
-    outputs = api.Arguments.createArguments(0)
-    training_machine.forward(inputs, outputs, api.PASS_TEST)
-    loss = outputs.getSlotValue(0).copyToNumpyMat()
-    return numpy.mean(loss)
-```
-
-After training one network, one needs to sync the new parameters to the other networks. The code below demonstrates one example of such use case:
-```python
-# Train the gen_training
-gen_trainer.trainOneDataBatch(batch_size, data_batch_gen)
-
-# Copy the parameters from gen_training to dis_training and generator
-copy_shared_parameters(gen_training_machine,
-dis_training_machine)
-copy_shared_parameters(gen_training_machine, generator_machine)
-```
-
-
-## A Toy Example 
-With the infrastructure explained above, we can now walk you through a toy example of generating two dimensional uniform distribution using 10 dimensional Gaussian noise. 
-
-The Gaussian noises are generated using the code below:
-```python
-def get_noise(batch_size, noise_dim):
-    return numpy.random.normal(size=(batch_size, noise_dim)).astype('float32')
-```
-
-The real samples (2-D uniform) are generated using the code below:
-```python
-# synthesize 2-D uniform data in gan_trainer.py:114
-def load_uniform_data():
-    data = numpy.random.rand(1000000, 2).astype('float32')
-    return data
-```
-
-The generator and discriminator network are built using fully-connected layer and batch_norm layer, and are defined in gan_conf.py. 
-
-To train the GAN model, one can use the command below. The flag -d specifies the training data (cifar, mnist or uniform) and flag --useGpu specifies whether to use gpu for training (0 is cpu, 1 is gpu).  
-```bash
-$python gan_trainer.py -d uniform --useGpu 1
-```
-The generated samples can be found in ./uniform_samples/ and one example is shown below as Figure 2. One can see that it roughly recovers the 2D uniform distribution. 
-
-<center>![](./uniform_sample.png)</center>
-<p align="center">
-    Figure 2. Uniform Sample
-</p>
-
-## MNIST Example
-### Data preparation
-To download the MNIST data, one can use the following commands:
-```bash
-$cd data/
-$./get_mnist_data.sh
-```
-
-### Model description
-Following the DC-Gan paper (https://arxiv.org/abs/1511.06434), we use convolution/convolution-transpose layer in the discriminator/generator network to better deal with images. The details of the network structures are defined in gan_conf_image.py. 
-
-### Training the model
-To train the GAN model on mnist data, one can use the following command:
-```bash
-$python gan_trainer.py -d mnist --useGpu 1
-```
-The generated sample images can be found at ./mnist_samples/ and one example is shown below as Figure 3. 
-<center>![](./mnist_sample.png)</center>
-<p align="center">
-    Figure 3. MNIST Sample
-</p>
--- a/doc/v1_api_tutorials/gan/mnist_sample.png
+++ b/doc/v1_api_tutorials/gan/mnist_sample.png
--- a/doc/v1_api_tutorials/gan/uniform_sample.png
+++ b/doc/v1_api_tutorials/gan/uniform_sample.png
--- a/doc/v1_api_tutorials/imagenet_model/resnet_block.jpg
+++ b/doc/v1_api_tutorials/imagenet_model/resnet_block.jpg
--- a/doc/v1_api_tutorials/imagenet_model/resnet_model_cn.md
+++ b/doc/v1_api_tutorials/imagenet_model/resnet_model_cn.md
-# Model Zoo - ImageNet #
-
-[ImageNet](http://www.image-net.org/) 是通用物体分类领域一个众所周知的数据库。本教程提供了一个用于ImageNet上的卷积分类网络模型。
-
-## ResNet 介绍
-
-论文 [Deep Residual Learning for Image Recognition](http://arxiv.org/abs/1512.03385) 中提出的ResNet网络结构在2015年ImageNet大规模视觉识别竞赛(ILSVRC 2015)的分类任务中赢得了第一名。他们提出残差学习的框架来简化网络的训练，所构建网络结构的的深度比之前使用的网络有大幅度的提高。下图展示的是基于残差的连接方式。左图构造网络模块的方式被用于34层的网络中，而右图的瓶颈连接模块用于50层，101层和152层的网络结构中。
-
-<center>![resnet_block](./resnet_block.jpg)</center>
-<center>图 1. ResNet 网络模块</center>
-
-本教程中我们给出了三个ResNet模型，这些模型都是由原作者提供的模型<https://github.com/KaimingHe/deep-residual-networks>转换过来的。我们使用PaddlePaddle在ILSVRC的验证集共50,000幅图像上测试了模型的分类错误率，其中输入图像的颜色通道顺序为**BGR**，保持宽高比缩放到短边为256，只截取中心方形的图像区域。分类错误率和模型大小由下表给出。
-<center>
-<table border="2" cellspacing="0" cellpadding="6" rules="all" frame="border">
-<colgroup>
-<col  class="left" />
-<col  class="left" />
-<col  class="left" />
-</colgroup>
-<thead>
-<tr>
-<th scope="col" class="left">ResNet</th>
-<th scope="col" class="left">Top-1</th>
-<th scope="col" class="left">Model Size</th>
-</tr>
-</thead>
-
-<tbody>
-<tr>
-<td class="left">ResNet-50</td>
-<td class="left">24.9%</td>
-<td class="left">99M</td>
-</tr>
-<tr>
-<td class="left">ResNet-101</td>
-<td class="left">23.7%</td>
-<td class="left">173M</td>
-</tr>
-<tr>
-<td class="left">ResNet-152</td>
-<td class="left">23.2%</td>
-<td class="left">234M</td>
-</tr>
-</tbody>
-
-</table></center>
-<br>
-
-## ResNet 模型
-
-50层，101层和152层的网络配置文件可参照```demo/model_zoo/resnet/resnet.py```。你也可以通过在命令行参数中增加一个参数如```--config_args=layer_num=50```来指定网络层的数目。
-
-### 网络可视化
-
-你可以通过执行下面的命令来得到ResNet网络的结构可视化图。该脚本会生成一个dot文件，然后可以转换为图片。需要安装graphviz来转换dot文件为图片。
-
-```
-cd demo/model_zoo/resnet
-./net_diagram.sh
-```
-
-### 模型下载
-
-```
-cd demo/model_zoo/resnet
-./get_model.sh
-```
-你可以执行上述命令来下载所有的模型和均值文件，如果下载成功，这些文件将会被保存在```demo/model_zoo/resnet/model```路径下。
-
-```
-mean_meta_224  resnet_101  resnet_152  resnet_50
-```
-   * resnet_50: 50层网络模型。
-   * resnet_101: 101层网络模型。
-   * resnet_152: 152层网络模型。
-   * mean\_meta\_224: 均值图像文件，图像大小为3 x 224 x 224，颜色通道顺序为**BGR**。你也可以使用这三个值: 103.939, 116.779, 123.68。
-
-### 参数信息
-
-* **卷积层权重**
-
-  由于每个卷积层后面连接的是batch normalization层，因此该层中没有偏置(bias)参数，并且只有一个权重。
-  形状: `(Co, ky, kx, Ci)`
-   * Co: 输出特征图的通道数目
-   * ky: 滤波器核在垂直方向上的尺寸
-   * kx: 滤波器核在水平方向上的尺寸
-   * Ci: 输入特征图的通道数目
-
-  二维矩阵: (Co * ky * kx, Ci), 行优先次序存储。
-
-* **全连接层权重**
-
-  二维矩阵: (输入层尺寸, 本层尺寸), 行优先次序存储。
-
-* **[Batch Normalization](<http://arxiv.org/abs/1502.03167>) 层权重**
-
-本层有四个参数，实际上只有.w0和.wbias是需要学习的参数，另外两个分别是滑动均值和方差。在测试阶段它们将会被加载到模型中。下表展示了batch normalization层的参数。
-<center>
-<table border="2" cellspacing="0" cellpadding="6" rules="all" frame="border">
-<colgroup>
-<col  class="left" />
-<col  class="left" />
-<col  class="left" />
-</colgroup>
-<thead>
-<tr>
-<th scope="col" class="left">参数名</th>
-<th scope="col" class="left">尺寸</th>
-<th scope="col" class="left">含义</th>
-</tr>
-</thead>
-
-<tbody>
-<tr>
-<td class="left">_res2_1_branch1_bn.w0</td>
-<td class="left">256</td>
-<td class="left">gamma, 缩放参数</td>
-</tr>
-<tr>
-<td class="left">_res2_1_branch1_bn.w1</td>
-<td class="left">256</td>
-<td class="left">特征图均值</td>
-</tr>
-<tr>
-<td class="left">_res2_1_branch1_bn.w2</td>
-<td class="left">256</td>
-<td class="left">特征图方差</td>
-</tr>
-<tr>
-<td class="left">_res2_1_branch1_bn.wbias</td>
-<td class="left">256</td>
-<td class="left">beta, 偏置参数</td>
-</tr>
-</tbody>
-
-</table></center>
-<br>
-
-### 参数读取
-
-使用者可以使用下面的Python脚本来读取参数值:
-
-```
-import sys
-import numpy as np
-
-def load(file_name):
-    with open(file_name, 'rb') as f:
-        f.read(16) # skip header for float type.
-        return np.fromfile(f, dtype=np.float32)
-
-if __name__=='__main__':
-    weight = load(sys.argv[1])
-```
-
-或者直接使用下面的shell命令:
-
-```
-od -j 16 -f _res2_1_branch1_bn.w0
-```
-
-## 特征提取
-
-我们提供了C++和Python接口来提取特征。下面的例子使用了`demo/model_zoo/resnet/example`中的数据，详细地展示了整个特征提取的过程。
-
-### C++接口
-
-首先，在配置文件中的`define_py_data_sources2`里指定图像数据列表，具体请参照示例`demo/model_zoo/resnet/resnet.py`。
-
-```
-    train_list = 'train.list' if not is_test else None
-    # mean.meta is mean file of ImageNet dataset.
-    # mean.meta size : 3 x 224 x 224.
-    # If you use three mean value, set like:
-    # "mean_value:103.939,116.779,123.68;"
-    args={
-        'mean_meta': "model/mean_meta_224/mean.meta",
-        'image_size': 224, 'crop_size': 224,
-        'color': True,'swap_channel:': [2, 1, 0]}
-    define_py_data_sources2(train_list,
-                           'example/test.list',
-                           module="example.image_list_provider",
-                           obj="processData",
-                           args=args)
-```
-
-第二步，在`resnet.py`文件中指定要提取特征的网络层的名字。例如，
-
-```
-Outputs("res5_3_branch2c_conv", "res5_3_branch2c_bn")
-```
-
-第三步，在`extract_fea_c++.sh`文件中指定模型路径和输出的目录，然后执行下面的命令。
-
-```
-cd demo/model_zoo/resnet
-./extract_fea_c++.sh
-```
-
-如果执行成功，特征将会存到`fea_output/rank-00000`文件中，如下所示。同时你可以使用`load_feature.py`文件中的`load_feature_c`接口来加载该文件。
-
-```
-0.115318 -0.108358 ... -0.087884;-1.27664 ... -1.11516 -2.59123;
-0.126383 -0.116248 ... -0.00534909;-1.42593 ... -1.04501 -1.40769;
-```
-
-* 每行存储的是一个样本的特征。其中，第一行存的是图像`example/dog.jpg`的特征，第二行存的是图像`example/cat.jpg`的特征。
-* 不同层的特征由分号`;`隔开，并且它们的顺序与`Outputs()`中指定的层顺序一致。这里，左边是`res5_3_branch2c_conv`层的特征，右边是`res5_3_branch2c_bn`层特征。
-
-### Python接口
-
-示例`demo/model_zoo/resnet/classify.py`中展示了如何使用Python来提取特征。下面的例子同样使用了`./example/test.list`中的数据。执行的命令如下：
-
-```
-cd demo/model_zoo/resnet
-./extract_fea_py.sh
-```
-
-extract_fea_py.sh:
-
-```
-python classify.py \
-     --job=extract \
-     --conf=resnet.py\
-     --use_gpu=1 \
-     --mean=model/mean_meta_224/mean.meta \
-     --model=model/resnet_50 \
-     --data=./example/test.list \
-     --output_layer="res5_3_branch2c_conv,res5_3_branch2c_bn" \
-     --output_dir=features
-
-```
-* \--job=extract:              指定工作模式来提取特征。
-* \--conf=resnet.py:           网络配置文件。
-* \--use_gpu=1:                指定是否使用GPU。
-* \--model=model/resnet_50:    模型路径。
-* \--data=./example/test.list: 数据列表。
-* \--output_layer="xxx,xxx":   指定提取特征的层。
-* \--output_dir=features:      输出目录。
-
-如果运行成功，你将会看到特征存储在`features/batch_0`文件中，该文件是由cPickle产生的。你可以使用`load_feature.py`中的`load_feature_py`接口来打开该文件，它将返回如下的字典：
-
-```
-{
-'cat.jpg': {'res5_3_branch2c_conv': array([[-0.12638293, -0.116248  , -0.11883899, ..., -0.00895038, 0.01994277, -0.00534909]], dtype=float32), 'res5_3_branch2c_bn': array([[-1.42593431, -1.28918779, -1.32414699, ..., -1.45933616, -1.04501402, -1.40769434]], dtype=float32)},
-'dog.jpg': {'res5_3_branch2c_conv': array([[-0.11531784, -0.10835785, -0.08809858, ...,0.0055237, 0.01505112, -0.08788397]], dtype=float32), 'res5_3_branch2c_bn': array([[-1.27663755, -1.18272924, -0.90937918, ..., -1.25178063, -1.11515927, -2.59122872]], dtype=float32)}
-}
-```
-
-仔细观察，这些特征值与上述使用C++接口提取的结果是一致的。
-
-## 预测
-
-`classify.py`文件也可以用于对样本进行预测。我们提供了一个示例脚本`predict.sh`，它使用50层的ResNet模型来对`example/test.list`中的数据进行预测。
-
-```
-cd demo/model_zoo/resnet
-./predict.sh
-```
-
-predict.sh调用了`classify.py`:
-
-```
-python classify.py \
-     --job=predict \
-     --conf=resnet.py\
-     --multi_crop \
-     --model=model/resnet_50 \
-     --use_gpu=1 \
-     --data=./example/test.list
-```
-* \--job=extract:              指定工作模型进行预测。
-* \--conf=resnet.py:           网络配置文件。network configure.
-* \--multi_crop:               使用10个裁剪图像块，预测概率取平均。
-* \--use_gpu=1:                指定是否使用GPU。
-* \--model=model/resnet_50:    模型路径。
-* \--data=./example/test.list: 数据列表。
-
-如果运行成功，你将会看到如下结果，其中156和285是这些图像的分类标签。
-
-```
-Label of example/dog.jpg is: 156
-Label of example/cat.jpg is: 282
-```
--- a/doc/v1_api_tutorials/imagenet_model/resnet_model_en.md
+++ b/doc/v1_api_tutorials/imagenet_model/resnet_model_en.md
-# Model Zoo - ImageNet #
-
-[ImageNet](http://www.image-net.org/) is a popular dataset for generic object classification. This tutorial provides convolutional neural network(CNN) models for ImageNet.
-
-## ResNet Introduction
-
-ResNets from paper [Deep Residual Learning for Image Recognition](http://arxiv.org/abs/1512.03385) won the 1st place on the ILSVRC 2015 classification task. They present residual learning framework to ease the training of networks that are substantially deeper than those used previously. The residual connections are shown in following figure. The left building block is used in network of 34 layers and the right bottleneck building block is used in network of 50, 101, 152 layers .
-
-<center>![resnet_block](./resnet_block.jpg)</center>
-<center>Figure 1. ResNet Block</center>
-
-We present three ResNet models, which are converted from the models provided by the authors <https://github.com/KaimingHe/deep-residual-networks>.  The classfication errors tested in PaddlePaddle on 50,000 ILSVRC validation set with input images channel order of **BGR** by single scale with the shorter side of 256 and single crop as following table.
-<center>
-<table border="2" cellspacing="0" cellpadding="6" rules="all" frame="border">
-<colgroup>
-<col  class="left" />
-<col  class="left" />
-<col  class="left" />
-</colgroup>
-<thead>
-<tr>
-<th scope="col" class="left">ResNet</th>
-<th scope="col" class="left">Top-1</th>
-<th scope="col" class="left">Model Size</th>
-</tr>
-</thead>
-
-<tbody>
-<tr>
-<td class="left">ResNet-50</td>
-<td class="left">24.9%</td>
-<td class="left">99M</td>
-</tr>
-<tr>
-<td class="left">ResNet-101</td>
-<td class="left">23.7%</td>
-<td class="left">173M</td>
-</tr>
-<tr>
-<td class="left">ResNet-152</td>
-<td class="left">23.2%</td>
-<td class="left">234M</td>
-</tr>
-</tbody>
-
-</table></center>
-<br>
-
-## ResNet Model
-
-See ```demo/model_zoo/resnet/resnet.py```. This config contains network of 50, 101 and 152 layers. You can specify layer number by adding argument like ```--config_args=layer_num=50``` in command line arguments.
-
-### Network Visualization
-
-You can get a diagram of ResNet network by running the following commands. The script generates dot file and then converts dot file to PNG file, which needs to install graphviz to convert.
-
-```
-cd demo/model_zoo/resnet
-./net_diagram.sh
-```
-
-### Model Download
-
-```
-cd demo/model_zoo/resnet
-./get_model.sh
-```
-You can run above command to download all models and mean file and save them in ```demo/model_zoo/resnet/model``` if downloading successfully.
-
-```
-mean_meta_224  resnet_101  resnet_152  resnet_50
-```
-   * resnet_50: model of 50 layers.
-   * resnet_101: model of 101 layers.
-   * resnet_152: model of 152 layers.
-   * mean\_meta\_224: mean file with 3 x 224 x 224 size in **BGR** order. You also can use three mean values: 103.939, 116.779, 123.68.
-
-### Parameter Info
-
-* **Convolution Layer Weight**
-
-  As batch normalization layer is connected after each convolution layer, there is no parameter of bias and only one weight in this layer.
-  shape: `(Co, ky, kx, Ci)`
-   * Co: channle number of output feature map.
-   * ky: filter size in vertical direction.
-   * kx: filter size in horizontal direction.
-   * Ci: channle number of input feature map.
-
-  2-Dim matrix: (Co * ky * kx, Ci), saved in row-major order.
-
-* **Fully connected Layer Weight**
-
-  2-Dim matrix: (input layer size, this layer size), saved in row-major order.
-
-* **[Batch Normalization](<http://arxiv.org/abs/1502.03167>) Layer Weight**
-
-There are four parameters in this layer. In fact, only .w0 and .wbias are the learned parameters. The other two are therunning mean and variance respectively. They will be loaded in testing. Following table shows parameters of a batch normzalization layer.
-<center>
-<table border="2" cellspacing="0" cellpadding="6" rules="all" frame="border">
-<colgroup>
-<col  class="left" />
-<col  class="left" />
-<col  class="left" />
-</colgroup>
-<thead>
-<tr>
-<th scope="col" class="left">Parameter Name</th>
-<th scope="col" class="left">Number</th>
-<th scope="col" class="left">Meaning</th>
-</tr>
-</thead>
-
-<tbody>
-<tr>
-<td class="left">_res2_1_branch1_bn.w0</td>
-<td class="left">256</td>
-<td class="left">gamma, scale parameter</td>
-</tr>
-<tr>
-<td class="left">_res2_1_branch1_bn.w1</td>
-<td class="left">256</td>
-<td class="left">mean value of feature map</td>
-</tr>
-<tr>
-<td class="left">_res2_1_branch1_bn.w2</td>
-<td class="left">256</td>
-<td class="left">variance of feature map</td>
-</tr>
-<tr>
-<td class="left">_res2_1_branch1_bn.wbias</td>
-<td class="left">256</td>
-<td class="left">beta, shift parameter</td>
-</tr>
-</tbody>
-
-</table></center>
-<br>
-
-### Parameter Observation
-
-Users who want to observe the parameters can use Python to read:
-
-```
-import sys
-import numpy as np
-
-def load(file_name):
-    with open(file_name, 'rb') as f:
-        f.read(16) # skip header for float type.
-        return np.fromfile(f, dtype=np.float32)
-
-if __name__=='__main__':
-    weight = load(sys.argv[1])
-```
-
-or simply use following shell command:
-
-```
-od -j 16 -f _res2_1_branch1_bn.w0
-```
-
-## Feature Extraction
-
-We provide both C++ and Python interfaces to extract features. The following examples use data in `demo/model_zoo/resnet/example` to show the extracting process in detail.
-
-### C++ Interface
-
-First, specify image data list in `define_py_data_sources2` in the config, see example `demo/model_zoo/resnet/resnet.py`.
-
-```
-    train_list = 'train.list' if not is_test else None
-    # mean.meta is mean file of ImageNet dataset.
-    # mean.meta size : 3 x 224 x 224.
-    # If you use three mean value, set like:
-    # "mean_value:103.939,116.779,123.68;"
-    args={
-        'mean_meta': "model/mean_meta_224/mean.meta",
-        'image_size': 224, 'crop_size': 224,
-        'color': True,'swap_channel:': [2, 1, 0]}
-    define_py_data_sources2(train_list,
-                           'example/test.list',
-                           module="example.image_list_provider",
-                           obj="processData",
-                           args=args)
-```
-
-Second, specify layers to extract features in `Outputs()` of `resnet.py`. For example,
-
-```
-Outputs("res5_3_branch2c_conv", "res5_3_branch2c_bn")
-```
-
-Third, specify model path and output directory in `extract_fea_c++.sh`, and then run the following commands.
-
-```
-cd demo/model_zoo/resnet
-./extract_fea_c++.sh
-```
-
-If successful, features are saved in `fea_output/rank-00000` as follows. And you can use `load_feature_c` interface in `load_feature.py ` to load such a file.
-
-```
-0.115318 -0.108358 ... -0.087884;-1.27664 ... -1.11516 -2.59123;
-0.126383 -0.116248 ... -0.00534909;-1.42593 ... -1.04501 -1.40769;
-```
-
-* Each line stores features of a sample. Here, the first line stores features of `example/dog.jpg` and second line stores features of `example/cat.jpg`.
-* Features of different layers are splitted by `;`, and their order is consistent with the layer order in `Outputs()`. Here, the left features are `res5_3_branch2c_conv` layer and right features are `res5_3_branch2c_bn` layer.
-
-### Python Interface
-
-`demo/model_zoo/resnet/classify.py` is an example to show how to use Python to extract features. Following example still uses data of `./example/test.list`. Command is as follows:
-
-```
-cd demo/model_zoo/resnet
-./extract_fea_py.sh
-```
-
-extract_fea_py.sh:
-
-```
-python classify.py \
-     --job=extract \
-     --conf=resnet.py\
-     --use_gpu=1 \
-     --mean=model/mean_meta_224/mean.meta \
-     --model=model/resnet_50 \
-     --data=./example/test.list \
-     --output_layer="res5_3_branch2c_conv,res5_3_branch2c_bn" \
-     --output_dir=features
-
-```
-* \--job=extract:              specify job mode to extract feature.
-* \--conf=resnet.py:           network configure.
-* \--use_gpu=1:             speficy GPU mode.
-* \--model=model/resnet_5:     model path.
-* \--data=./example/test.list: data list.
-* \--output_layer="xxx,xxx":   specify layers to extract features.
-* \--output_dir=features:      output diretcoty.
-
-If run successfully, you will see features saved in `features/batch_0`, this file is produced with cPickle. You can use `load_feature_py` interface in `load_feature.py` to open the file, and it returns a dictionary as follows:
-
-```
-{
-'cat.jpg': {'res5_3_branch2c_conv': array([[-0.12638293, -0.116248  , -0.11883899, ..., -0.00895038, 0.01994277, -0.00534909]], dtype=float32), 'res5_3_branch2c_bn': array([[-1.42593431, -1.28918779, -1.32414699, ..., -1.45933616, -1.04501402, -1.40769434]], dtype=float32)},
-'dog.jpg': {'res5_3_branch2c_conv': array([[-0.11531784, -0.10835785, -0.08809858, ...,0.0055237, 0.01505112, -0.08788397]], dtype=float32), 'res5_3_branch2c_bn': array([[-1.27663755, -1.18272924, -0.90937918, ..., -1.25178063, -1.11515927, -2.59122872]], dtype=float32)}
-}
-```
-
-Observed carefully, these feature values are consistent with the above results extracted by C++ interface.
-
-## Prediction
-
-`classify.py` also can be used to predict. We provide an example script `predict.sh` to predict data in `example/test.list` using a ResNet model with 50 layers.
-
-```
-cd demo/model_zoo/resnet
-./predict.sh
-```
-
-predict.sh calls the `classify.py`:
-
-```
-python classify.py \
-     --job=predict \
-     --conf=resnet.py\
-     --multi_crop \
-     --model=model/resnet_50 \
-     --use_gpu=1 \
-     --data=./example/test.list
-```
-* \--job=extract:              speficy job mode to predict.
-* \--conf=resnet.py:           network configure.
-* \--multi_crop:               use 10 crops and average predicting probability.
-* \--use_gpu=1:             speficy GPU mode.
-* \--model=model/resnet_50:    model path.
-* \--data=./example/test.list: data list.
-
-If run successfully, you will see following results, where 156 and 285 are labels of the images.
-
-```
-Label of example/dog.jpg is: 156
-Label of example/cat.jpg is: 282
-```
--- a/doc/v1_api_tutorials/quick_start/index_cn.rst
+++ b/doc/v1_api_tutorials/quick_start/index_cn.rst
-=============
-快速入门教程
-=============
-
-我们将以 `文本分类问题 <https://en.wikipedia.org/wiki/Document_classification>`_ 为例,
-介绍PaddlePaddle的基本使用方法。
-
-安装
-====
-
-请参考 :ref:`install_steps` 安装PaddlePaddle。
-
-使用概述
-========
-
-**文本分类问题**：对于给定的一条文本，我们从提前给定的类别集合中选择其所属类别。
-
-比如, 在购物网站上，通过查看买家对某个产品的评价反馈, 评估该产品的质量。
-
- 这个显示器很棒！ （好评）
- 用了两个月之后这个显示器屏幕碎了。（差评）
-
-使用PaddlePaddle, 每一个任务流程都可以被划分为如下五个步骤。
-
-    ..  image:: src/Pipeline_cn.jpg
-        :align: center
-        :scale: 80%
-
-1. 数据格式准备
-    - 本例每行保存一条样本，类别Id和文本信息用 ``Tab`` 间隔，文本中的单词用空格分隔（如果不切词，则字与字之间用空格分隔），例如：``类别Id '\t' 这 个 显 示 器 很 棒 ！``
-2. 向系统传送数据
-    - PaddlePaddle可以执行用户的python脚本程序来读取各种格式的数据文件。
-    - 本例的所有字符都将转换为连续整数表示的Id传给模型。
-3. 描述网络结构和优化算法
-    - 本例由易到难展示4种不同的文本分类网络配置：逻辑回归模型，词向量模型，卷积模型，时序模型。
-    - 常用优化算法包括Momentum, RMSProp，AdaDelta，AdaGrad，Adam，Adamax等，本例采用Adam优化方法，加了L2正则和梯度截断。
-4. 训练模型
-5. 应用模型
-
-数据格式准备
------------
-
-接下来我们将展示如何用PaddlePaddle训练一个文本分类模型，将 `Amazon电子产品评论数据 <http://jmcauley.ucsd.edu/data/amazon/>`_ 分为好评(正样本)和差评(负样本)两种类别。
-`源代码 <https://github.com/PaddlePaddle/Paddle>`_ 的 ``demo/quick_start`` 目录里提供了该数据的下载脚本和预处理脚本，你只需要在命令行输入以下命令，就能够很方便的完成数据下载和相应的预处理工作。
-
-.. code-block:: bash
-
-    cd demo/quick_start
-    ./data/get_data.sh
-    ./preprocess.sh
-
-数据预处理完成之后，通过配置类似于 ``dataprovider_*.py`` 的数据读取脚本和类似于 ``trainer_config.*.py`` 的训练模型脚本，PaddlePaddle将以设置参数的方式来设置
-相应的数据读取脚本和训练模型脚本。接下来，我们将对这两个步骤给出了详细的解释，你也可以先跳过本文的解释环节，直接进入训练模型章节, 使用 ``sh train.sh`` 开始训练模型，
-查看`train.sh`内容，通过 **自底向上法** (bottom-up approach)来帮助你理解PaddlePaddle的内部运行机制。
-
-
-向系统传送数据
-==============
-
-Python脚本读取数据
------------------
-
-`DataProvider` 是PaddlePaddle负责提供数据的模块，主要职责在于将训练数据传入内存或者显存，让模型能够得到训练更新，其包括两个函数：
-
-* initializer：PaddlePaddle会在调用读取数据的Python脚本之前，先调用initializer函数。在下面例子里，我们在initialzier函数里初始化词表，并且在随后的读取数据过程中填充词表。
-* process：PaddlePaddle调用process函数来读取数据。每次读取一条数据后，process函数会用yield语句输出这条数据，从而能够被PaddlePaddle 捕获 (harvest)。
-
-``dataprovider_bow.py`` 文件给出了完整例子：
-
-..  literalinclude:: ../../../demo/quick_start/dataprovider_bow.py
-     :language: python
-     :lines: 21-70
-     :linenos:
-     :emphasize-lines: 8,33
-
-详细内容请参见 :ref:`api_dataprovider` 。
-
-配置中的数据加载定义
--------------------
-
-在模型配置中通过 ``define_py_data_sources2`` 接口来加载数据：
-
-..  literalinclude:: ../../../demo/quick_start/trainer_config.emb.py
-     :language: python
-     :lines: 19-35
-     :linenos:
-     :emphasize-lines: 12
-
-
-以下是对上述数据加载的解释：
-
- data/train.list,data/test.list: 指定训练数据和测试数据
- module="dataprovider_bow": 处理数据的Python脚本文件
- obj="process": 指定生成数据的函数
- args={"dictionary": word_dict}: 额外的参数，这里指定词典
-
-更详细数据格式和用例请参考 :ref:`api_pydataprovider2` 。
-
-模型网络结构
-============
-
-本小节我们将介绍模型网络结构。
-
-    ..  image:: src/PipelineNetwork_cn.jpg
-        :align: center
-        :scale: 80%
-
-
-我们将以最基本的逻辑回归网络作为起点，并逐渐展示更加深入的功能。更详细的网络配置连接请参考 :ref:`api_trainer_config_helpers_layers` 。
-所有配置都能在 `源代码 <https://github.com/PaddlePaddle/Paddle>`_ 的 ``demo/quick_start`` 目录下找到。
-
-逻辑回归模型
------------
-
-具体流程如下:
-
-    ..  image:: src/NetLR_cn.jpg
-        :align: center
-        :scale: 80%
-
- 获取利用 `one-hot vector <https://en.wikipedia.org/wiki/One-hot>`_ 表示的每个单词，维度是词典大小
-
-    .. code-block:: python
-
-        word = data_layer(name="word",  size=word_dim)
-
- 获取该条样本类别Id，维度是类别个数。
-
-    .. code-block:: python
-
-        label = data_layer(name="label", size=label_dim)
-
- 利用逻辑回归模型对该向量进行分类，同时会计算分类准确率
-
-    .. code-block:: python
-
-        # Define a fully connected layer with logistic activation (also called softmax activation).
-        output = fc_layer(input=word,
-                        size=label_dim,
-                        act_type=SoftmaxActivation())
-        # Define cross-entropy classification loss and error.
-        classification_cost(input=output, label=label)
-
-
- - input: 除去data层，每个层都有一个或多个input,多个input以list方式输入
- - size: 该层神经元个数
- - act_type: 激活函数类型
-
-**效果总结**：我们将在后面介绍训练和预测流程的脚本。在此为方便对比不同网络结构，我们总结了各个网络的复杂度和效果。
-
-    =====================  ===============================  =================
-    网络名称                        参数数量                    错误率
-    =====================  ===============================  =================
-    逻辑回归                      252 KB                       8.652 %
-    =====================  ===============================  =================
-
-词向量模型
----------
-
-embedding模型需要稍微改变提供数据的Python脚本，即 ``dataprovider_emb.py``，词向量模型、
-卷积模型、时序模型均使用该脚本。其中文本输入类型定义为整数时序类型integer_value_sequence。
-
-.. code-block:: python
-
-    def initializer(settings, dictionary, **kwargs):
-        settings.word_dict = dictionary
-        settings.input_types = [
-            # Define the type of the first input as sequence of integer.
-            # The value of the integers range from 0 to len(dictrionary)-1
-            integer_value_sequence(len(dictionary)),
-            # Define the second input for label id
-            integer_value(2)]
-
-    @provider(init_hook=initializer)
-    def process(settings, file_name):
-        ...
-        # omitted, it is same as the data provider for LR model
-
-该模型依然使用逻辑回归分类网络的框架， 只是将句子用连续向量表示替换为用稀疏向量表示， 即对第三步进行替换。句子表示的计算更新为两步：
-
-..  image:: src/NetContinuous_cn.jpg
-    :align: center
-    :scale: 80%
-
- 利用单词Id查找该单词对应的连续向量(维度为word_dim)， 输入N个单词，输出为N个word_dim维度向量
-
-    .. code-block:: python
-
-        emb = embedding_layer(input=word, size=word_dim)
-
- 将该句话包含的所有单词向量求平均, 得到句子的表示
-
-    .. code-block:: python
-
-        avg = pooling_layer(input=emb, pooling_type=AvgPooling())
-
-其它部分和逻辑回归网络结构一致。
-
-**效果总结：**
-
-    =====================  ===============================  ==================
-    网络名称                        参数数量                    错误率
-    =====================  ===============================  ==================
-    词向量模型                      15 MB                       8.484 %
-    =====================  ===============================  ==================
-
-卷积模型
-----------
-
-卷积网络是一种特殊的从词向量表示到句子表示的方法， 也就是将词向量模型进一步演化为三个新步骤。
-
-..  image:: src/NetConv_cn.jpg
-    :align: center
-    :scale: 80%
-
-文本卷积分可为三个步骤:
-
-1. 首先，从每个单词左右两端分别获取k个相邻的单词, 拼接成一个新的向量；
-
-2. 其次，对该向量进行非线性变换(例如Sigmoid变换), 使其转变为维度为hidden_dim的新向量；
-
-3. 最后，对整个新向量集合的每一个维度取最大值来表示最后的句子。
-
-这三个步骤可配置为:
-
-.. code-block:: python
-
-    text_conv = sequence_conv_pool(input=emb,
-                                context_start=k,
-                                context_len=2 * k + 1)
-
-**效果总结：**
-
-    =====================  ===============================  ========================
-    网络名称                        参数数量                    错误率
-    =====================  ===============================  ========================
-    卷积模型                      16 MB                       5.628 %
-    =====================  ===============================  ========================
-
-时序模型
----------
-
-..  image:: src/NetRNN_cn.jpg
-    :align: center
-    :scale: 80%
-
-时序模型，也称为RNN模型, 包括简单的 `RNN模型 <https://en.wikipedia.org/wiki/Recurrent_neural_network>`_, `GRU模型 <https://en.wikipedia.org/wiki/Gated_recurrent_unit>`_ 和 `LSTM模型 <https://en.wikipedia.org/wiki/Long_short-term_memory>`_ 等等。
-
- GRU模型配置：
-
-    .. code-block:: python
-
-        gru = simple_gru(input=emb, size=gru_size)
-
-
- LSTM模型配置：
-
-    .. code-block:: python
-
-        lstm = simple_lstm(input=emb, size=lstm_size)
-
-本次试验，我们采用单层LSTM模型，并使用了Dropout，**效果总结：**
-
-    =====================  ===============================  =========================
-    网络名称                        参数数量                    错误率
-    =====================  ===============================  =========================
-    时序模型                      16 MB                       4.812 %
-    =====================  ===============================  =========================
-
-优化算法
-=========
-
-`优化算法 <http://www.paddlepaddle.org/doc/ui/api/trainer_config_helpers/optimizers_index.html>`_ 包括
-Momentum, RMSProp，AdaDelta，AdaGrad，ADAM，Adamax等，这里采用Adam优化方法，同时使用了L2正则(L2 Regularization)和梯度截断(Gradient Clipping)。
-
-.. code-block:: python
-
-    settings(batch_size=128,
-            learning_rate=2e-3,
-            learning_method=AdamOptimizer(),
-            regularization=L2Regularization(8e-4),
-            gradient_clipping_threshold=25)
-
-训练模型
-=========
-
-在数据加载和网络配置完成之后， 我们就可以训练模型了。
-
-..  image:: src/PipelineTrain_cn.jpg
-    :align: center
-    :scale: 80%
-
-训练模型，我们只需要运行 ``train.sh`` 训练脚本：
-
-    .. code-block:: bash
-
-        ./train.sh
-
-``train.sh`` 中包含了训练模型的基本命令。训练时所需设置的主要参数如下：
-
-    .. code-block:: bash
-
-        paddle train \
-        --config=trainer_config.py \
-        --log_period=20 \
-        --save_dir=./output \
-        --num_passes=15 \
-        --use_gpu=false
-
-这里只简单介绍了单机训练，如何进行分布式训练，请参考 :ref:`cluster_train` 。
-
-预测
-=====
-
-当模型训练好了之后，我们就可以进行预测了。
-
-..  image:: src/PipelineTest_cn.jpg
-    :align: center
-    :scale: 80%
-
-之前配置文件中 ``test.list`` 指定的数据将会被测试，这里直接通过预测脚本 ``predict.sh`` 进行预测,
-更详细的说明，请参考 :ref:`api_swig_py_paddle` 。
-
-    .. code-block:: bash
-
-        model="output/pass-00003"
-        paddle train \
-            --config=trainer_config.lstm.py \
-            --use_gpu=false \
-            --job=test \
-            --init_model_path=$model \
-            --config_args=is_predict=1 \
-            --predict_output_dir=. \
-
-        mv rank-00000 result.txt
-
-这里以 ``output/pass-00003`` 为例进行预测，用户可以根据训练日志，选择测试结果最好的模型来预测。
-
-预测结果以文本的形式保存在 ``result.txt`` 中，一行为一个样本，格式如下：
-
-    .. code-block:: bash
-
-        预测ID;ID为0的概率 ID为1的概率
-        预测ID;ID为0的概率 ID为1的概率
-
-总体效果总结
-==============
-
-在 ``/demo/quick_start`` 目录下，能够找到这里使用的所有数据, 网络配置, 训练脚本等等。
-对于Amazon-Elec测试集(25k), 如下表格，展示了上述网络模型的训练效果:
-
-    =====================  ===============================  =============  ==================================
-    网络名称                       参数数量                    错误率          配置文件
-    =====================  ===============================  =============  ==================================
-    逻辑回归模型                      252 KB                     8.652%          trainer_config.lr.py
-    词向量模型                         15 MB                      8.484%         trainer_config.emb.py
-    卷积模型                        16 MB                     5.628%          trainer_config.cnn.py
-    时序模型                         16 MB                     4.812%          trainer_config.lstm.py
-    =====================  ===============================  =============  ==================================
-
-
-附录
-=====
-
-命令行参数
----------
-
-* \--config：网络配置
-* \--save_dir：模型存储路径
-* \--log_period：每隔多少batch打印一次日志
-* \--num_passes：训练轮次，一个pass表示过一遍所有训练样本
-* \--config_args：命令指定的参数会传入网络配置中。
-* \--init_model_path：指定初始化模型路径，可用在测试或训练时指定初始化模型。
-
-默认一个pass保存一次模型，也可以通过saving_period_by_batches设置每隔多少batch保存一次模型。
-可以通过show_parameter_stats_period设置打印参数信息等。
-其他参数请参考 命令行参数文档（链接待补充）。
-
-输出日志
---------
-
-.. code-block:: bash
-
-    TrainerInternal.cpp:160]  Batch=20 samples=2560 AvgCost=0.628761 CurrentCost=0.628761 Eval: classification_error_evaluator=0.304297  CurrentEval: classification_error_evaluator=0.304297
-
-模型训练会看到类似上面这样的日志信息，详细的参数解释，请参考如下表格：
-
-    ===========================================  ==============================================================
-    名称                                             解释
-    ===========================================  ==============================================================
-    Batch=20                                      表示过了20个batch
-    samples=2560                                  表示过了2560个样本
-    AvgCost                                          每个pass的第0个batch到当前batch所有样本的平均cost
-    CurrentCost                                      当前log_period个batch所有样本的平均cost
-    Eval: classification_error_evaluator          每个pass的第0个batch到当前batch所有样本的平均分类错误率
-    CurrentEval: classification_error_evaluator      当前log_period个batch所有样本的平均分类错误率
-    ===========================================  ==============================================================
--- a/doc/v1_api_tutorials/quick_start/index_en.md
+++ b/doc/v1_api_tutorials/quick_start/index_en.md
--- a/doc/v1_api_tutorials/quick_start/src/NetContinuous_cn.jpg
+++ b/doc/v1_api_tutorials/quick_start/src/NetContinuous_cn.jpg
--- a/doc/v1_api_tutorials/quick_start/src/NetContinuous_en.png
+++ b/doc/v1_api_tutorials/quick_start/src/NetContinuous_en.png
--- a/doc/v1_api_tutorials/quick_start/src/NetConv_cn.jpg
+++ b/doc/v1_api_tutorials/quick_start/src/NetConv_cn.jpg
--- a/doc/v1_api_tutorials/quick_start/src/NetConv_en.png
+++ b/doc/v1_api_tutorials/quick_start/src/NetConv_en.png
--- a/doc/v1_api_tutorials/quick_start/src/NetLR_cn.jpg
+++ b/doc/v1_api_tutorials/quick_start/src/NetLR_cn.jpg
--- a/doc/v1_api_tutorials/quick_start/src/NetLR_en.png
+++ b/doc/v1_api_tutorials/quick_start/src/NetLR_en.png
--- a/doc/v1_api_tutorials/quick_start/src/NetRNN_cn.jpg
+++ b/doc/v1_api_tutorials/quick_start/src/NetRNN_cn.jpg
--- a/doc/v1_api_tutorials/quick_start/src/NetRNN_en.png
+++ b/doc/v1_api_tutorials/quick_start/src/NetRNN_en.png
--- a/doc/v1_api_tutorials/quick_start/src/PipelineNetwork_cn.jpg
+++ b/doc/v1_api_tutorials/quick_start/src/PipelineNetwork_cn.jpg
--- a/doc/v1_api_tutorials/quick_start/src/PipelineNetwork_en.jpg
+++ b/doc/v1_api_tutorials/quick_start/src/PipelineNetwork_en.jpg
--- a/doc/v1_api_tutorials/quick_start/src/PipelineTest_cn.jpg
+++ b/doc/v1_api_tutorials/quick_start/src/PipelineTest_cn.jpg
--- a/doc/v1_api_tutorials/quick_start/src/PipelineTest_en.png
+++ b/doc/v1_api_tutorials/quick_start/src/PipelineTest_en.png
--- a/doc/v1_api_tutorials/quick_start/src/PipelineTrain_cn.jpg
+++ b/doc/v1_api_tutorials/quick_start/src/PipelineTrain_cn.jpg
--- a/doc/v1_api_tutorials/quick_start/src/PipelineTrain_en.png
+++ b/doc/v1_api_tutorials/quick_start/src/PipelineTrain_en.png
--- a/doc/v1_api_tutorials/quick_start/src/Pipeline_cn.jpg
+++ b/doc/v1_api_tutorials/quick_start/src/Pipeline_cn.jpg
--- a/doc/v1_api_tutorials/quick_start/src/Pipeline_en.jpg
+++ b/doc/v1_api_tutorials/quick_start/src/Pipeline_en.jpg
--- a/go/pserver/client/c/test/test_mnist.py
+++ b/go/pserver/client/c/test/test_mnist.py
--- a/go/pserver/client/c/test/test_train.py
+++ b/go/pserver/client/c/test/test_train.py
--- a/paddle/CMakeLists.txt
+++ b/paddle/CMakeLists.txt
--- a/paddle/api/test/testTrainConfig.py
+++ b/paddle/api/test/testTrainConfig.py
--- a/paddle/capi/examples/model_inference/dense/main.c
+++ b/paddle/capi/examples/model_inference/dense/main.c
--- a/paddle/capi/examples/model_inference/dense/merge_v2_model.py
+++ b/paddle/capi/examples/model_inference/dense/merge_v2_model.py
--- a/paddle/capi/examples/model_inference/dense/mnist_v2.py
+++ b/paddle/capi/examples/model_inference/dense/mnist_v2.py
--- a/paddle/capi/examples/model_inference/dense/trainer_config.py
+++ b/paddle/capi/examples/model_inference/dense/trainer_config.py
--- a/paddle/capi/examples/model_inference/multi_thread/main.c
+++ b/paddle/capi/examples/model_inference/multi_thread/main.c
--- a/paddle/capi/examples/model_inference/multi_thread/main_gpu.c
+++ b/paddle/capi/examples/model_inference/multi_thread/main_gpu.c
--- a/paddle/capi/examples/model_inference/sequence/main.c
+++ b/paddle/capi/examples/model_inference/sequence/main.c
--- a/paddle/capi/examples/model_inference/sequence/trainer_config.py
+++ b/paddle/capi/examples/model_inference/sequence/trainer_config.py
--- a/paddle/capi/examples/model_inference/sparse_binary/main.c
+++ b/paddle/capi/examples/model_inference/sparse_binary/main.c
--- a/paddle/capi/tests/test_predict_network.py
+++ b/paddle/capi/tests/test_predict_network.py
--- a/paddle/framework/CMakeLists.txt
+++ b/paddle/framework/CMakeLists.txt
--- a/paddle/framework/attribute.cc
+++ b/paddle/framework/attribute.cc
--- a/paddle/framework/attribute.h
+++ b/paddle/framework/attribute.h
--- a/paddle/framework/block_desc.cc
+++ b/paddle/framework/block_desc.cc
--- a/paddle/framework/data_device_transform.cc
+++ b/paddle/framework/data_device_transform.cc
--- a/paddle/framework/data_device_transform.h
+++ b/paddle/framework/data_device_transform.h
--- a/paddle/framework/data_device_transform_test.cu
+++ b/paddle/framework/data_device_transform_test.cu
--- a/paddle/framework/data_layout_transform.cc
+++ b/paddle/framework/data_layout_transform.cc
--- a/paddle/framework/data_layout_transform.h
+++ b/paddle/framework/data_layout_transform.h
--- a/paddle/framework/data_layout_transform_test.cc
+++ b/paddle/framework/data_layout_transform_test.cc
--- a/paddle/framework/data_transform.cc
+++ b/paddle/framework/data_transform.cc
--- a/paddle/framework/data_type_transform.cc
+++ b/paddle/framework/data_type_transform.cc
--- a/paddle/framework/data_type_transform.h
+++ b/paddle/framework/data_type_transform.h
--- a/paddle/framework/data_type_transform_test.cc
+++ b/paddle/framework/data_type_transform_test.cc
--- a/paddle/framework/eigen_test.cc
+++ b/paddle/framework/eigen_test.cc
--- a/paddle/framework/executor.cc
+++ b/paddle/framework/executor.cc
--- a/paddle/framework/framework.proto
+++ b/paddle/framework/framework.proto
--- a/paddle/framework/lod_tensor.cc
+++ b/paddle/framework/lod_tensor.cc
--- a/paddle/framework/lod_tensor_test.cc
+++ b/paddle/framework/lod_tensor_test.cc
--- a/paddle/framework/lod_tensor_test.cu
+++ b/paddle/framework/lod_tensor_test.cu
--- a/paddle/framework/op_desc.cc
+++ b/paddle/framework/op_desc.cc
--- a/paddle/framework/op_desc.h
+++ b/paddle/framework/op_desc.h
--- a/paddle/framework/op_kernel_type.h
+++ b/paddle/framework/op_kernel_type.h
--- a/paddle/framework/operator.cc
+++ b/paddle/framework/operator.cc
--- a/paddle/framework/scope.cc
+++ b/paddle/framework/scope.cc
--- a/paddle/framework/tensor_test.cc
+++ b/paddle/framework/tensor_test.cc
--- a/paddle/framework/tensor_util_test.cc
+++ b/paddle/framework/tensor_util_test.cc
--- a/paddle/framework/type_defs.h
+++ b/paddle/framework/type_defs.h
--- a/paddle/framework/var_desc.h
+++ b/paddle/framework/var_desc.h
--- a/paddle/framework/variable.h
+++ b/paddle/framework/variable.h
--- a/paddle/framework/variable_test.cc
+++ b/paddle/framework/variable_test.cc
--- a/paddle/gserver/layers/PriorBox.cpp
+++ b/paddle/gserver/layers/PriorBox.cpp
--- a/paddle/inference/CMakeLists.txt
+++ b/paddle/inference/CMakeLists.txt
--- a/paddle/inference/example.cc
+++ b/paddle/inference/example.cc
--- a/paddle/inference/inference.cc
+++ b/paddle/inference/inference.cc
--- a/paddle/inference/inference.h
+++ b/paddle/inference/inference.h
--- a/paddle/memory/CMakeLists.txt
+++ b/paddle/memory/CMakeLists.txt
--- a/paddle/operators/CMakeLists.txt
+++ b/paddle/operators/CMakeLists.txt
--- a/paddle/operators/assign_value_op.cc
+++ b/paddle/operators/assign_value_op.cc
--- a/paddle/operators/assign_value_op.h
+++ b/paddle/operators/assign_value_op.h
--- a/paddle/operators/beam_search_op.cc
+++ b/paddle/operators/beam_search_op.cc
--- a/paddle/operators/beam_search_op.h
+++ b/paddle/operators/beam_search_op.h
--- a/paddle/operators/beam_search_op_test.cc
+++ b/paddle/operators/beam_search_op_test.cc
--- a/paddle/operators/bipartite_match_op.cc
+++ b/paddle/operators/bipartite_match_op.cc
--- a/paddle/operators/clip_op.cc
+++ b/paddle/operators/clip_op.cc
--- a/paddle/operators/compare_op.cc
+++ b/paddle/operators/compare_op.cc
--- a/paddle/operators/compare_op.cu
+++ b/paddle/operators/compare_op.cu
--- a/paddle/operators/compare_op.h
+++ b/paddle/operators/compare_op.h
--- a/paddle/operators/conv_op.cc
+++ b/paddle/operators/conv_op.cc
--- a/paddle/operators/conv_transpose_op.cc
+++ b/paddle/operators/conv_transpose_op.cc
--- a/paddle/operators/conv_transpose_op.h
+++ b/paddle/operators/conv_transpose_op.h
--- a/paddle/operators/ctc_align_op.cc
+++ b/paddle/operators/ctc_align_op.cc
--- a/paddle/operators/ctc_align_op.cu
+++ b/paddle/operators/ctc_align_op.cu
--- a/paddle/operators/ctc_align_op.h
+++ b/paddle/operators/ctc_align_op.h
--- a/paddle/operators/detail/grpc_client.cc
+++ b/paddle/operators/detail/grpc_client.cc
--- a/paddle/operators/detail/grpc_server.cc
+++ b/paddle/operators/detail/grpc_server.cc
--- a/paddle/operators/detail/grpc_server.h
+++ b/paddle/operators/detail/grpc_server.h
--- a/paddle/operators/edit_distance_op.cc
+++ b/paddle/operators/edit_distance_op.cc
--- a/paddle/operators/edit_distance_op.cu
+++ b/paddle/operators/edit_distance_op.cu
--- a/paddle/operators/edit_distance_op.h
+++ b/paddle/operators/edit_distance_op.h
--- a/paddle/operators/elementwise_div_op.h
+++ b/paddle/operators/elementwise_div_op.h
--- a/paddle/operators/elementwise_mul_op.h
+++ b/paddle/operators/elementwise_mul_op.h
--- a/paddle/operators/elementwise_op.h
+++ b/paddle/operators/elementwise_op.h
--- a/paddle/operators/elementwise_op_function.h
+++ b/paddle/operators/elementwise_op_function.h
--- a/paddle/operators/elementwise_sub_op.h
+++ b/paddle/operators/elementwise_sub_op.h
--- a/paddle/operators/expand_op.cc
+++ b/paddle/operators/expand_op.cc
--- a/paddle/operators/im2sequence_op.cc
+++ b/paddle/operators/im2sequence_op.cc
--- a/paddle/operators/im2sequence_op.cu
+++ b/paddle/operators/im2sequence_op.cu
--- a/paddle/operators/im2sequence_op.h
+++ b/paddle/operators/im2sequence_op.h
--- a/paddle/operators/iou_similarity_op.cc
+++ b/paddle/operators/iou_similarity_op.cc
--- a/paddle/operators/iou_similarity_op.cu
+++ b/paddle/operators/iou_similarity_op.cu
--- a/paddle/operators/iou_similarity_op.h
+++ b/paddle/operators/iou_similarity_op.h
--- a/paddle/operators/linear_chain_crf_op.cc
+++ b/paddle/operators/linear_chain_crf_op.cc
--- a/paddle/operators/linear_chain_crf_op.h
+++ b/paddle/operators/linear_chain_crf_op.h
--- a/paddle/operators/lookup_table_op.cc
+++ b/paddle/operators/lookup_table_op.cc
--- a/paddle/operators/lookup_table_op.cu
+++ b/paddle/operators/lookup_table_op.cu
--- a/paddle/operators/lookup_table_op.h
+++ b/paddle/operators/lookup_table_op.h
--- a/paddle/operators/lstm_op.cc
+++ b/paddle/operators/lstm_op.cc
--- a/paddle/operators/math/matmul.h
+++ b/paddle/operators/math/matmul.h
--- a/paddle/operators/math/sampler.cc
+++ b/paddle/operators/math/sampler.cc
--- a/paddle/operators/math/sampler.h
+++ b/paddle/operators/math/sampler.h
--- a/paddle/operators/math/sequence_padding.cc
+++ b/paddle/operators/math/sequence_padding.cc
--- a/paddle/operators/math/sequence_padding.cu
+++ b/paddle/operators/math/sequence_padding.cu
--- a/paddle/operators/math/sequence_padding_test.cc
+++ b/paddle/operators/math/sequence_padding_test.cc
--- a/paddle/operators/matmul_op.cc
+++ b/paddle/operators/matmul_op.cc
--- a/paddle/operators/matmul_op.h
+++ b/paddle/operators/matmul_op.h
--- a/paddle/operators/nce_op.cc
+++ b/paddle/operators/nce_op.cc
--- a/paddle/operators/nce_op.h
+++ b/paddle/operators/nce_op.h
--- a/paddle/operators/net_op.cc
+++ b/paddle/operators/net_op.cc
--- a/paddle/operators/parallel_do_op.cc
+++ b/paddle/operators/parallel_do_op.cc
--- a/paddle/operators/pool_op.cc
+++ b/paddle/operators/pool_op.cc
--- a/paddle/operators/prior_box_op.cc
+++ b/paddle/operators/prior_box_op.cc
--- a/paddle/operators/prior_box_op.h
+++ b/paddle/operators/prior_box_op.h
--- a/paddle/operators/recv_op.cc
+++ b/paddle/operators/recv_op.cc
--- a/paddle/operators/reduce_op.cc
+++ b/paddle/operators/reduce_op.cc
--- a/paddle/operators/reduce_op.cu
+++ b/paddle/operators/reduce_op.cu
--- a/paddle/operators/scale_op.cc
+++ b/paddle/operators/scale_op.cc
--- a/paddle/operators/send_op.cc
+++ b/paddle/operators/send_op.cc
--- a/paddle/operators/send_recv_op_test.cc
+++ b/paddle/operators/send_recv_op_test.cc
--- a/paddle/operators/sequence_erase_op.cc
+++ b/paddle/operators/sequence_erase_op.cc
--- a/paddle/operators/sequence_erase_op.cu
+++ b/paddle/operators/sequence_erase_op.cu
--- a/paddle/operators/sequence_expand_op.cc
+++ b/paddle/operators/sequence_expand_op.cc
--- a/paddle/operators/sequence_expand_op.h
+++ b/paddle/operators/sequence_expand_op.h
--- a/paddle/operators/sequence_reshape_op.cc
+++ b/paddle/operators/sequence_reshape_op.cc
--- a/paddle/operators/sequence_reshape_op.cu
+++ b/paddle/operators/sequence_reshape_op.cu
--- a/paddle/operators/sequence_reshape_op.h
+++ b/paddle/operators/sequence_reshape_op.h
--- a/paddle/operators/split_selected_rows_op.cc
+++ b/paddle/operators/split_selected_rows_op.cc
--- a/paddle/operators/split_selected_rows_op.cu
+++ b/paddle/operators/split_selected_rows_op.cu
--- a/paddle/operators/split_selected_rows_op.h
+++ b/paddle/operators/split_selected_rows_op.h
--- a/paddle/operators/top_k_op.h
+++ b/paddle/operators/top_k_op.h
--- a/paddle/operators/transpose_op.cc
+++ b/paddle/operators/transpose_op.cc
--- a/paddle/optimizer/parameter_optimizer_test.cc
+++ b/paddle/optimizer/parameter_optimizer_test.cc
--- a/paddle/optimizer/serialization.h
+++ b/paddle/optimizer/serialization.h
--- a/paddle/optimizer/serialization_test.cc
+++ b/paddle/optimizer/serialization_test.cc
--- a/paddle/platform/CMakeLists.txt
+++ b/paddle/platform/CMakeLists.txt
--- a/paddle/platform/assert.h
+++ b/paddle/platform/assert.h
--- a/paddle/platform/profiler.cc
+++ b/paddle/platform/profiler.cc
--- a/paddle/platform/profiler.h
+++ b/paddle/platform/profiler.h
--- a/paddle/platform/profiler_test.cc
+++ b/paddle/platform/profiler_test.cc
--- a/paddle/pybind/CMakeLists.txt
+++ b/paddle/pybind/CMakeLists.txt
--- a/paddle/pybind/print_operators_doc.cc
+++ b/paddle/pybind/print_operators_doc.cc
--- a/paddle/pybind/protobuf.cc
+++ b/paddle/pybind/protobuf.cc
--- a/paddle/pybind/protobuf.h
+++ b/paddle/pybind/protobuf.h
--- a/paddle/pybind/pybind.cc
+++ b/paddle/pybind/pybind.cc
--- a/paddle/scripts/cluster_train/paddle.py
+++ b/paddle/scripts/cluster_train/paddle.py
--- a/paddle/scripts/cpplint.py
+++ b/paddle/scripts/cpplint.py
--- a/paddle/string/CMakeLists.txt
+++ b/paddle/string/CMakeLists.txt
--- a/paddle/string/piece.cc
+++ b/paddle/string/piece.cc
--- a/paddle/string/piece.h
+++ b/paddle/string/piece.h
--- a/paddle/string/piece_test.cc
+++ b/paddle/string/piece_test.cc
--- a/paddle/string/printf.h
+++ b/paddle/string/printf.h
--- a/paddle/string/tinyformat/tinyformat.h
+++ b/paddle/string/tinyformat/tinyformat.h
--- a/paddle/trainer/tests/picojson.h
+++ b/paddle/trainer/tests/picojson.h
--- a/paddle/trainer/tests/simple_sparse_neural_network.py
+++ b/paddle/trainer/tests/simple_sparse_neural_network.py
--- a/paddle/trainer/tests/simple_sparse_neural_network_dp.py
+++ b/paddle/trainer/tests/simple_sparse_neural_network_dp.py
--- a/paddle/utils/enable_virtualenv.py
+++ b/paddle/utils/enable_virtualenv.py
--- a/python/paddle/trainer_config_helpers/evaluators.py
+++ b/python/paddle/trainer_config_helpers/evaluators.py
--- a/python/paddle/trainer_config_helpers/tests/configs/img_layers.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/img_layers.py
--- a/python/paddle/trainer_config_helpers/tests/configs/img_trans_layers.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/img_trans_layers.py
--- a/python/paddle/trainer_config_helpers/tests/configs/last_first_seq.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/last_first_seq.py
--- a/python/paddle/trainer_config_helpers/tests/configs/layer_activations.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/layer_activations.py
--- a/python/paddle/trainer_config_helpers/tests/configs/math_ops.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/math_ops.py
--- a/python/paddle/trainer_config_helpers/tests/configs/projections.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/projections.py
--- a/python/paddle/trainer_config_helpers/tests/configs/shared_fc.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/shared_fc.py
--- a/python/paddle/trainer_config_helpers/tests/configs/shared_gru.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/shared_gru.py
--- a/python/paddle/trainer_config_helpers/tests/configs/shared_lstm.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/shared_lstm.py
--- a/python/paddle/trainer_config_helpers/tests/configs/simple_rnn_layers.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/simple_rnn_layers.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_BatchNorm3D.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_BatchNorm3D.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_bi_grumemory.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_bi_grumemory.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_bilinear_interp.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_bilinear_interp.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_clip_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_clip_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_config_parser_for_non_file_config.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_config_parser_for_non_file_config.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_conv3d_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_conv3d_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_cost_layers.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_cost_layers.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_cost_layers_with_weight.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_cost_layers_with_weight.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_crop.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_crop.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_cross_entropy_over_beam.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_cross_entropy_over_beam.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_deconv3d_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_deconv3d_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_detection_output_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_detection_output_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_dot_prod_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_dot_prod_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_expand_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_expand_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_factorization_machine.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_factorization_machine.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_fc.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_fc.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_gated_unit_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_gated_unit_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_grumemory_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_grumemory_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_hsigmoid.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_hsigmoid.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_kmax_seq_socre_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_kmax_seq_socre_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_l2_distance_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_l2_distance_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_lstmemory_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_lstmemory_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_maxout.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_maxout.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_multibox_loss_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_multibox_loss_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_multiplex_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_multiplex_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_ntm_layers.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_ntm_layers.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_pad.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_pad.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_pooling3D_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_pooling3D_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_prelu_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_prelu_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_print_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_print_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_recursive_topology.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_recursive_topology.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_repeat_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_repeat_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_resize_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_resize_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_rnn_group.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_rnn_group.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_roi_pool_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_roi_pool_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_row_conv.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_row_conv.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_row_l2_norm_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_row_l2_norm_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_scale_shift_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_scale_shift_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_scale_sub_region_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_scale_sub_region_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_seq_concat_reshape.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_seq_concat_reshape.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_seq_slice_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_seq_slice_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_sequence_pooling.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_sequence_pooling.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_smooth_l1.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_smooth_l1.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_split_datasource.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_split_datasource.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_spp_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_spp_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/test_sub_nested_seq_select_layer.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/test_sub_nested_seq_select_layer.py
--- a/python/paddle/trainer_config_helpers/tests/configs/unused_layers.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/unused_layers.py
--- a/python/paddle/trainer_config_helpers/tests/configs/util_layers.py
+++ b/python/paddle/trainer_config_helpers/tests/configs/util_layers.py
--- a/python/paddle/trainer_config_helpers/tests/test_reset_hook.py
+++ b/python/paddle/trainer_config_helpers/tests/test_reset_hook.py
--- a/python/paddle/utils/image_multiproc.py
+++ b/python/paddle/utils/image_multiproc.py
--- a/python/paddle/utils/plotcurve.py
+++ b/python/paddle/utils/plotcurve.py
--- a/python/paddle/v2/dataset/__init__.py
+++ b/python/paddle/v2/dataset/__init__.py
--- a/python/paddle/v2/dataset/common.py
+++ b/python/paddle/v2/dataset/common.py
--- a/python/paddle/v2/dataset/sentiment.py
+++ b/python/paddle/v2/dataset/sentiment.py
--- a/python/paddle/v2/dataset/tests/imikolov_test.py
+++ b/python/paddle/v2/dataset/tests/imikolov_test.py
--- a/python/paddle/v2/dataset/tests/test_sentiment.py
+++ b/python/paddle/v2/dataset/tests/test_sentiment.py
--- a/python/paddle/v2/dataset/tests/wmt16_test.py
+++ b/python/paddle/v2/dataset/tests/wmt16_test.py
--- a/python/paddle/v2/dataset/wmt14.py
+++ b/python/paddle/v2/dataset/wmt14.py
--- a/python/paddle/v2/dataset/wmt16.py
+++ b/python/paddle/v2/dataset/wmt16.py
--- a/python/paddle/v2/event.py
+++ b/python/paddle/v2/event.py
--- a/python/paddle/v2/fluid/__init__.py
+++ b/python/paddle/v2/fluid/__init__.py
--- a/python/paddle/v2/fluid/backward.py
+++ b/python/paddle/v2/fluid/backward.py
--- a/python/paddle/v2/fluid/clip.py
+++ b/python/paddle/v2/fluid/clip.py
--- a/python/paddle/v2/fluid/data_feeder.py
+++ b/python/paddle/v2/fluid/data_feeder.py
--- a/python/paddle/v2/fluid/default_scope_funcs.py
+++ b/python/paddle/v2/fluid/default_scope_funcs.py
--- a/python/paddle/v2/fluid/distribute_transpiler.py
+++ b/python/paddle/v2/fluid/distribute_transpiler.py
--- a/python/paddle/v2/fluid/distribute_transpiler_simple.py
+++ b/python/paddle/v2/fluid/distribute_transpiler_simple.py
--- a/python/paddle/v2/fluid/distributed_spliter.py
+++ b/python/paddle/v2/fluid/distributed_spliter.py
--- a/python/paddle/v2/fluid/evaluator.py
+++ b/python/paddle/v2/fluid/evaluator.py
--- a/python/paddle/v2/fluid/executor.py
+++ b/python/paddle/v2/fluid/executor.py
--- a/python/paddle/v2/fluid/framework.py
+++ b/python/paddle/v2/fluid/framework.py
--- a/python/paddle/v2/fluid/initializer.py
+++ b/python/paddle/v2/fluid/initializer.py
--- a/python/paddle/v2/fluid/io.py
+++ b/python/paddle/v2/fluid/io.py
--- a/python/paddle/v2/fluid/layer_helper.py
+++ b/python/paddle/v2/fluid/layer_helper.py
--- a/python/paddle/v2/fluid/layers/__init__.py
+++ b/python/paddle/v2/fluid/layers/__init__.py
--- a/python/paddle/v2/fluid/layers/control_flow.py
+++ b/python/paddle/v2/fluid/layers/control_flow.py
--- a/python/paddle/v2/fluid/layers/device.py
+++ b/python/paddle/v2/fluid/layers/device.py
--- a/python/paddle/v2/fluid/layers/io.py
+++ b/python/paddle/v2/fluid/layers/io.py
--- a/python/paddle/v2/fluid/registry.py
+++ b/python/paddle/v2/fluid/registry.py
--- a/python/paddle/v2/fluid/layers/math_op_patch.py
+++ b/python/paddle/v2/fluid/layers/math_op_patch.py
--- a/python/paddle/v2/fluid/layers/nn.py
+++ b/python/paddle/v2/fluid/layers/nn.py
--- a/python/paddle/v2/fluid/layers/ops.py
+++ b/python/paddle/v2/fluid/layers/ops.py
--- a/python/paddle/v2/fluid/layers/tensor.py
+++ b/python/paddle/v2/fluid/layers/tensor.py
--- a/python/paddle/v2/fluid/memory_optimization_transpiler.py
+++ b/python/paddle/v2/fluid/memory_optimization_transpiler.py
--- a/python/paddle/v2/fluid/net_drawer.py
+++ b/python/paddle/v2/fluid/net_drawer.py
--- a/python/paddle/v2/fluid/nets.py
+++ b/python/paddle/v2/fluid/nets.py
--- a/python/paddle/v2/fluid/op.py
+++ b/python/paddle/v2/fluid/op.py
--- a/python/paddle/v2/fluid/optimizer.py
+++ b/python/paddle/v2/fluid/optimizer.py
--- a/python/paddle/v2/fluid/param_attr.py
+++ b/python/paddle/v2/fluid/param_attr.py
--- a/python/paddle/v2/fluid/profiler.py
+++ b/python/paddle/v2/fluid/profiler.py
--- a/python/paddle/v2/fluid/regularizer.py
+++ b/python/paddle/v2/fluid/regularizer.py
--- a/python/paddle/v2/fluid/tests/CMakeLists.txt
+++ b/python/paddle/v2/fluid/tests/CMakeLists.txt
--- a/python/paddle/v2/fluid/tests/__init__.py
+++ b/python/paddle/v2/fluid/tests/__init__.py
--- a/python/paddle/v2/fluid/tests/book/test_fit_a_line.py
+++ b/python/paddle/v2/fluid/tests/book/test_fit_a_line.py
--- a/python/paddle/v2/fluid/tests/book/test_image_classification_train.py
+++ b/python/paddle/v2/fluid/tests/book/test_image_classification_train.py
--- a/python/paddle/v2/fluid/tests/book/test_label_semantic_roles.py
+++ b/python/paddle/v2/fluid/tests/book/test_label_semantic_roles.py
--- a/python/paddle/v2/fluid/tests/book/test_machine_translation.py
+++ b/python/paddle/v2/fluid/tests/book/test_machine_translation.py
--- a/python/paddle/v2/fluid/tests/book/test_recognize_digits_conv.py
+++ b/python/paddle/v2/fluid/tests/book/test_recognize_digits_conv.py
--- a/python/paddle/v2/fluid/tests/book/test_recognize_digits_mlp.py
+++ b/python/paddle/v2/fluid/tests/book/test_recognize_digits_mlp.py
--- a/python/paddle/v2/fluid/tests/book/test_recommender_system.py
+++ b/python/paddle/v2/fluid/tests/book/test_recommender_system.py
--- a/python/paddle/v2/fluid/tests/book/test_understand_sentiment_conv.py
+++ b/python/paddle/v2/fluid/tests/book/test_understand_sentiment_conv.py
--- a/python/paddle/v2/fluid/tests/book/test_understand_sentiment_dynamic_lstm.py
+++ b/python/paddle/v2/fluid/tests/book/test_understand_sentiment_dynamic_lstm.py
--- a/python/paddle/v2/fluid/tests/book/test_understand_sentiment_lstm.py
+++ b/python/paddle/v2/fluid/tests/book/test_understand_sentiment_lstm.py
--- a/python/paddle/v2/fluid/tests/book/test_word2vec.py
+++ b/python/paddle/v2/fluid/tests/book/test_word2vec.py
--- a/python/paddle/v2/fluid/tests/book_distribute/notest_dist_fit_a_line.py
+++ b/python/paddle/v2/fluid/tests/book_distribute/notest_dist_fit_a_line.py
--- a/python/paddle/v2/fluid/tests/book_distribute/notest_dist_image_classification.py
+++ b/python/paddle/v2/fluid/tests/book_distribute/notest_dist_image_classification.py
--- a/python/paddle/v2/fluid/tests/book_distribute/notest_dist_label_semantic_roles.py
+++ b/python/paddle/v2/fluid/tests/book_distribute/notest_dist_label_semantic_roles.py
--- a/python/paddle/v2/fluid/tests/book_distribute/notest_dist_word2vec.py
+++ b/python/paddle/v2/fluid/tests/book_distribute/notest_dist_word2vec.py
--- a/python/paddle/v2/fluid/tests/book_distribute/notest_machine_translation.py
+++ b/python/paddle/v2/fluid/tests/book_distribute/notest_machine_translation.py
--- a/python/paddle/v2/fluid/tests/book_distribute/notest_recognize_digits_conv_dist.py
+++ b/python/paddle/v2/fluid/tests/book_distribute/notest_recognize_digits_conv_dist.py
--- a/python/paddle/v2/fluid/tests/book_distribute/notest_recognize_digits_mlp_dist.py
+++ b/python/paddle/v2/fluid/tests/book_distribute/notest_recognize_digits_mlp_dist.py
--- a/python/paddle/v2/fluid/tests/book_distribute/notest_understand_sentiment_conv_dist.py
+++ b/python/paddle/v2/fluid/tests/book_distribute/notest_understand_sentiment_conv_dist.py
--- a/python/paddle/v2/fluid/tests/book_distribute/test_split_var.py
+++ b/python/paddle/v2/fluid/tests/book_distribute/test_split_var.py
--- a/python/paddle/v2/fluid/tests/book_memory_optimization/CMakeLists.txt
+++ b/python/paddle/v2/fluid/tests/book_memory_optimization/CMakeLists.txt
--- a/python/paddle/v2/fluid/tests/book_memory_optimization/test_memopt_fit_a_line.py
+++ b/python/paddle/v2/fluid/tests/book_memory_optimization/test_memopt_fit_a_line.py
--- a/python/paddle/v2/fluid/tests/book_memory_optimization/test_memopt_image_classification_train.py
+++ b/python/paddle/v2/fluid/tests/book_memory_optimization/test_memopt_image_classification_train.py
--- a/python/paddle/v2/fluid/tests/book_memory_optimization/test_memopt_machine_translation.py
+++ b/python/paddle/v2/fluid/tests/book_memory_optimization/test_memopt_machine_translation.py
--- a/python/paddle/v2/fluid/tests/decorators.py
+++ b/python/paddle/v2/fluid/tests/decorators.py
--- a/python/paddle/v2/fluid/tests/demo/fc_gan.py
+++ b/python/paddle/v2/fluid/tests/demo/fc_gan.py
--- a/python/paddle/v2/fluid/tests/op_test.py
+++ b/python/paddle/v2/fluid/tests/op_test.py
--- a/python/paddle/v2/fluid/tests/test_accuracy_op.py
+++ b/python/paddle/v2/fluid/tests/test_accuracy_op.py
--- a/python/paddle/v2/fluid/tests/test_activation_op.py
+++ b/python/paddle/v2/fluid/tests/test_activation_op.py
--- a/python/paddle/v2/fluid/tests/test_adadelta_op.py
+++ b/python/paddle/v2/fluid/tests/test_adadelta_op.py
--- a/python/paddle/v2/fluid/tests/test_adagrad_op.py
+++ b/python/paddle/v2/fluid/tests/test_adagrad_op.py
--- a/python/paddle/v2/fluid/tests/test_adam_op.py
+++ b/python/paddle/v2/fluid/tests/test_adam_op.py
--- a/python/paddle/v2/fluid/tests/test_adamax_op.py
+++ b/python/paddle/v2/fluid/tests/test_adamax_op.py
--- a/python/paddle/v2/fluid/tests/test_array_read_write_op.py
+++ b/python/paddle/v2/fluid/tests/test_array_read_write_op.py
--- a/python/paddle/v2/fluid/tests/test_assign_op.py
+++ b/python/paddle/v2/fluid/tests/test_assign_op.py
--- a/python/paddle/v2/fluid/tests/test_assign_value_op.py
+++ b/python/paddle/v2/fluid/tests/test_assign_value_op.py
--- a/python/paddle/v2/fluid/tests/test_auc_op.py
+++ b/python/paddle/v2/fluid/tests/test_auc_op.py
--- a/python/paddle/v2/fluid/tests/test_batch_norm_op.py
+++ b/python/paddle/v2/fluid/tests/test_batch_norm_op.py
--- a/python/paddle/v2/fluid/tests/test_beam_search_decode_op.py
+++ b/python/paddle/v2/fluid/tests/test_beam_search_decode_op.py
--- a/python/paddle/v2/fluid/tests/test_beam_search_op.py
+++ b/python/paddle/v2/fluid/tests/test_beam_search_op.py
--- a/python/paddle/v2/fluid/tests/test_bilinear_tensor_product_op.py
+++ b/python/paddle/v2/fluid/tests/test_bilinear_tensor_product_op.py
--- a/python/paddle/v2/fluid/tests/test_bipartite_match_op.py
+++ b/python/paddle/v2/fluid/tests/test_bipartite_match_op.py
--- a/python/paddle/v2/fluid/tests/test_calc_gradient.py
+++ b/python/paddle/v2/fluid/tests/test_calc_gradient.py
--- a/python/paddle/v2/fluid/tests/test_cast_op.py
+++ b/python/paddle/v2/fluid/tests/test_cast_op.py
--- a/python/paddle/v2/fluid/tests/test_chunk_eval_op.py
+++ b/python/paddle/v2/fluid/tests/test_chunk_eval_op.py
--- a/python/paddle/v2/fluid/tests/test_clip_by_norm_op.py
+++ b/python/paddle/v2/fluid/tests/test_clip_by_norm_op.py
--- a/python/paddle/v2/fluid/tests/test_clip_op.py
+++ b/python/paddle/v2/fluid/tests/test_clip_op.py
--- a/python/paddle/v2/fluid/tests/test_compare_op.py
+++ b/python/paddle/v2/fluid/tests/test_compare_op.py
--- a/python/paddle/v2/fluid/tests/test_concat_op.py
+++ b/python/paddle/v2/fluid/tests/test_concat_op.py
--- a/python/paddle/v2/fluid/tests/test_cond_op.py
+++ b/python/paddle/v2/fluid/tests/test_cond_op.py
--- a/python/paddle/v2/fluid/tests/test_conditional_block.py
+++ b/python/paddle/v2/fluid/tests/test_conditional_block.py
--- a/python/paddle/v2/fluid/tests/test_const_value.py
+++ b/python/paddle/v2/fluid/tests/test_const_value.py
--- a/python/paddle/v2/fluid/tests/test_conv2d_op.py
+++ b/python/paddle/v2/fluid/tests/test_conv2d_op.py
--- a/python/paddle/v2/fluid/tests/test_conv2d_transpose_op.py
+++ b/python/paddle/v2/fluid/tests/test_conv2d_transpose_op.py
--- a/python/paddle/v2/fluid/tests/test_conv3d_op.py
+++ b/python/paddle/v2/fluid/tests/test_conv3d_op.py
--- a/python/paddle/v2/fluid/tests/test_conv3d_transpose_op.py
+++ b/python/paddle/v2/fluid/tests/test_conv3d_transpose_op.py
--- a/python/paddle/v2/fluid/tests/test_conv_shift_op.py
+++ b/python/paddle/v2/fluid/tests/test_conv_shift_op.py
--- a/python/paddle/v2/fluid/tests/test_cos_sim_op.py
+++ b/python/paddle/v2/fluid/tests/test_cos_sim_op.py
--- a/python/paddle/v2/fluid/tests/test_create_op_doc_string.py
+++ b/python/paddle/v2/fluid/tests/test_create_op_doc_string.py
--- a/python/paddle/v2/fluid/tests/test_crf_decoding_op.py
+++ b/python/paddle/v2/fluid/tests/test_crf_decoding_op.py
--- a/python/paddle/v2/fluid/tests/test_crop_op.py
+++ b/python/paddle/v2/fluid/tests/test_crop_op.py
--- a/python/paddle/v2/fluid/tests/test_cross_entropy_op.py
+++ b/python/paddle/v2/fluid/tests/test_cross_entropy_op.py
--- a/python/paddle/v2/fluid/tests/test_ctc_align.py
+++ b/python/paddle/v2/fluid/tests/test_ctc_align.py
--- a/python/paddle/v2/fluid/tests/test_data_feeder.py
+++ b/python/paddle/v2/fluid/tests/test_data_feeder.py
--- a/python/paddle/v2/fluid/tests/test_decayed_adagrad_op.py
+++ b/python/paddle/v2/fluid/tests/test_decayed_adagrad_op.py
--- a/python/paddle/v2/fluid/tests/test_default_scope_funcs.py
+++ b/python/paddle/v2/fluid/tests/test_default_scope_funcs.py
--- a/python/paddle/v2/fluid/tests/test_detection_output_op.py
+++ b/python/paddle/v2/fluid/tests/test_detection_output_op.py
--- a/python/paddle/v2/fluid/tests/test_dropout_op.py
+++ b/python/paddle/v2/fluid/tests/test_dropout_op.py
--- a/python/paddle/v2/fluid/tests/test_dyn_rnn.py
+++ b/python/paddle/v2/fluid/tests/test_dyn_rnn.py
--- a/python/paddle/v2/fluid/tests/test_dynrnn_gradient_check.py
+++ b/python/paddle/v2/fluid/tests/test_dynrnn_gradient_check.py
--- a/python/paddle/v2/fluid/tests/test_dynrnn_static_input.py
+++ b/python/paddle/v2/fluid/tests/test_dynrnn_static_input.py
--- a/python/paddle/v2/fluid/tests/test_edit_distance_op.py
+++ b/python/paddle/v2/fluid/tests/test_edit_distance_op.py
--- a/python/paddle/v2/fluid/tests/test_elementwise_add_op.py
+++ b/python/paddle/v2/fluid/tests/test_elementwise_add_op.py
--- a/python/paddle/v2/fluid/tests/test_elementwise_div_op.py
+++ b/python/paddle/v2/fluid/tests/test_elementwise_div_op.py
--- a/python/paddle/v2/fluid/tests/test_elementwise_max_op.py
+++ b/python/paddle/v2/fluid/tests/test_elementwise_max_op.py
--- a/python/paddle/v2/fluid/tests/test_elementwise_min_op.py
+++ b/python/paddle/v2/fluid/tests/test_elementwise_min_op.py
--- a/python/paddle/v2/fluid/tests/test_elementwise_mul_op.py
+++ b/python/paddle/v2/fluid/tests/test_elementwise_mul_op.py
--- a/python/paddle/v2/fluid/tests/test_elementwise_sub_op.py
+++ b/python/paddle/v2/fluid/tests/test_elementwise_sub_op.py
--- a/python/paddle/v2/fluid/tests/test_clip.py
+++ b/python/paddle/v2/fluid/tests/test_clip.py
--- a/python/paddle/v2/fluid/tests/test_exception.py
+++ b/python/paddle/v2/fluid/tests/test_exception.py
--- a/python/paddle/v2/fluid/tests/test_executor_and_mul.py
+++ b/python/paddle/v2/fluid/tests/test_executor_and_mul.py
--- a/python/paddle/v2/fluid/tests/test_expand_op.py
+++ b/python/paddle/v2/fluid/tests/test_expand_op.py
--- a/python/paddle/v2/fluid/tests/test_feed_fetch_method.py
+++ b/python/paddle/v2/fluid/tests/test_feed_fetch_method.py
--- a/python/paddle/v2/fluid/tests/test_fill_constant_batch_size_like_op.py
+++ b/python/paddle/v2/fluid/tests/test_fill_constant_batch_size_like_op.py
--- a/python/paddle/v2/fluid/tests/test_fill_constant_op.py
+++ b/python/paddle/v2/fluid/tests/test_fill_constant_op.py
--- a/python/paddle/v2/fluid/tests/test_fill_op.py
+++ b/python/paddle/v2/fluid/tests/test_fill_op.py
--- a/python/paddle/v2/fluid/tests/test_fill_zeros_like_op.py
+++ b/python/paddle/v2/fluid/tests/test_fill_zeros_like_op.py
--- a/python/paddle/v2/fluid/tests/test_framework_debug_str.py
+++ b/python/paddle/v2/fluid/tests/test_framework_debug_str.py
--- a/python/paddle/v2/fluid/tests/test_ftrl_op.py
+++ b/python/paddle/v2/fluid/tests/test_ftrl_op.py
--- a/python/paddle/v2/fluid/tests/test_gather_op.py
+++ b/python/paddle/v2/fluid/tests/test_gather_op.py
--- a/python/paddle/v2/fluid/tests/test_gaussian_random_op.py
+++ b/python/paddle/v2/fluid/tests/test_gaussian_random_op.py
--- a/python/paddle/v2/fluid/tests/test_get_places_op.py
+++ b/python/paddle/v2/fluid/tests/test_get_places_op.py
--- a/python/paddle/v2/fluid/tests/test_gradient_clip.py
+++ b/python/paddle/v2/fluid/tests/test_gradient_clip.py
--- a/python/paddle/v2/fluid/tests/test_gru_op.py
+++ b/python/paddle/v2/fluid/tests/test_gru_op.py
--- a/python/paddle/v2/fluid/tests/test_gru_unit_op.py
+++ b/python/paddle/v2/fluid/tests/test_gru_unit_op.py
--- a/python/paddle/v2/fluid/tests/test_hinge_loss_op.py
+++ b/python/paddle/v2/fluid/tests/test_hinge_loss_op.py
--- a/python/paddle/v2/fluid/tests/test_huber_loss_op.py
+++ b/python/paddle/v2/fluid/tests/test_huber_loss_op.py
--- a/python/paddle/v2/fluid/tests/test_im2sequence_op.py
+++ b/python/paddle/v2/fluid/tests/test_im2sequence_op.py
--- a/python/paddle/v2/fluid/tests/test_image_classification_layer.py
+++ b/python/paddle/v2/fluid/tests/test_image_classification_layer.py
--- a/python/paddle/v2/fluid/tests/test_infer_shape.py
+++ b/python/paddle/v2/fluid/tests/test_infer_shape.py
--- a/python/paddle/v2/fluid/tests/test_inference_model_io.py
+++ b/python/paddle/v2/fluid/tests/test_inference_model_io.py
--- a/python/paddle/v2/fluid/tests/test_initializer.py
+++ b/python/paddle/v2/fluid/tests/test_initializer.py
--- a/python/paddle/v2/fluid/tests/test_iou_similarity_op.py
+++ b/python/paddle/v2/fluid/tests/test_iou_similarity_op.py
--- a/python/paddle/v2/fluid/tests/test_is_empty_op.py
+++ b/python/paddle/v2/fluid/tests/test_is_empty_op.py
--- a/python/paddle/v2/fluid/tests/test_l1_norm_op.py
+++ b/python/paddle/v2/fluid/tests/test_l1_norm_op.py
--- a/python/paddle/v2/fluid/tests/test_layers.py
+++ b/python/paddle/v2/fluid/tests/test_layers.py
--- a/python/paddle/v2/fluid/tests/test_linear_chain_crf_op.py
+++ b/python/paddle/v2/fluid/tests/test_linear_chain_crf_op.py
--- a/python/paddle/v2/fluid/tests/test_lod_array_length_op.py
+++ b/python/paddle/v2/fluid/tests/test_lod_array_length_op.py
--- a/python/paddle/v2/fluid/tests/test_lod_rank_table.py
+++ b/python/paddle/v2/fluid/tests/test_lod_rank_table.py
--- a/python/paddle/v2/fluid/tests/test_lod_reset_op.py
+++ b/python/paddle/v2/fluid/tests/test_lod_reset_op.py
--- a/python/paddle/v2/fluid/tests/test_lod_tensor_array.py
+++ b/python/paddle/v2/fluid/tests/test_lod_tensor_array.py
--- a/python/paddle/v2/fluid/tests/test_lod_tensor_array_ops.py
+++ b/python/paddle/v2/fluid/tests/test_lod_tensor_array_ops.py
--- a/python/paddle/v2/fluid/tests/test_log_loss_op.py
+++ b/python/paddle/v2/fluid/tests/test_log_loss_op.py
--- a/python/paddle/v2/fluid/tests/test_logical_op.py
+++ b/python/paddle/v2/fluid/tests/test_logical_op.py
--- a/python/paddle/v2/fluid/tests/test_lookup_table_op.py
+++ b/python/paddle/v2/fluid/tests/test_lookup_table_op.py
--- a/python/paddle/v2/fluid/tests/test_lrn_op.py
+++ b/python/paddle/v2/fluid/tests/test_lrn_op.py
--- a/python/paddle/v2/fluid/tests/test_lstm_op.py
+++ b/python/paddle/v2/fluid/tests/test_lstm_op.py
--- a/python/paddle/v2/fluid/tests/test_lstm_unit_op.py
+++ b/python/paddle/v2/fluid/tests/test_lstm_unit_op.py
--- a/python/paddle/v2/fluid/tests/test_margin_rank_loss_op.py
+++ b/python/paddle/v2/fluid/tests/test_margin_rank_loss_op.py
--- a/python/paddle/v2/fluid/tests/test_math_op_patch.py
+++ b/python/paddle/v2/fluid/tests/test_math_op_patch.py
--- a/python/paddle/v2/fluid/tests/test_matmul_op.py
+++ b/python/paddle/v2/fluid/tests/test_matmul_op.py
--- a/python/paddle/v2/fluid/tests/test_maxout_op.py
+++ b/python/paddle/v2/fluid/tests/test_maxout_op.py
--- a/python/paddle/v2/fluid/tests/test_mean_op.py
+++ b/python/paddle/v2/fluid/tests/test_mean_op.py
--- a/python/paddle/v2/fluid/tests/test_memory_optimization_transpiler.py
+++ b/python/paddle/v2/fluid/tests/test_memory_optimization_transpiler.py
--- a/python/paddle/v2/fluid/tests/test_minus_op.py
+++ b/python/paddle/v2/fluid/tests/test_minus_op.py
--- a/python/paddle/v2/fluid/tests/test_mnist_if_else_op.py
+++ b/python/paddle/v2/fluid/tests/test_mnist_if_else_op.py
--- a/python/paddle/v2/fluid/tests/test_modified_huber_loss_op.py
+++ b/python/paddle/v2/fluid/tests/test_modified_huber_loss_op.py
--- a/python/paddle/v2/fluid/tests/test_momentum_op.py
+++ b/python/paddle/v2/fluid/tests/test_momentum_op.py
--- a/python/paddle/v2/fluid/tests/test_mul_op.py
+++ b/python/paddle/v2/fluid/tests/test_mul_op.py
--- a/python/paddle/v2/fluid/tests/test_multiplex_op.py
+++ b/python/paddle/v2/fluid/tests/test_multiplex_op.py
--- a/python/paddle/v2/fluid/tests/test_nce.py
+++ b/python/paddle/v2/fluid/tests/test_nce.py
--- a/python/paddle/v2/fluid/tests/test_net.py
+++ b/python/paddle/v2/fluid/tests/test_net.py
--- a/python/paddle/v2/fluid/tests/test_norm_op.py
+++ b/python/paddle/v2/fluid/tests/test_norm_op.py
--- a/python/paddle/v2/fluid/tests/test_normalization_wrapper.py
+++ b/python/paddle/v2/fluid/tests/test_normalization_wrapper.py
--- a/python/paddle/v2/fluid/tests/test_op_support_gpu.py
+++ b/python/paddle/v2/fluid/tests/test_op_support_gpu.py
--- a/python/paddle/v2/fluid/tests/test_operator.py
+++ b/python/paddle/v2/fluid/tests/test_operator.py
--- a/python/paddle/v2/fluid/tests/test_operator_desc.py
+++ b/python/paddle/v2/fluid/tests/test_operator_desc.py
--- a/python/paddle/v2/fluid/tests/test_optimizer.py
+++ b/python/paddle/v2/fluid/tests/test_optimizer.py
--- a/python/paddle/v2/fluid/tests/test_pad_op.py
+++ b/python/paddle/v2/fluid/tests/test_pad_op.py
--- a/python/paddle/v2/fluid/tests/test_parallel_op.py
+++ b/python/paddle/v2/fluid/tests/test_parallel_op.py
--- a/python/paddle/v2/fluid/tests/test_parameter.py
+++ b/python/paddle/v2/fluid/tests/test_parameter.py
--- a/python/paddle/v2/fluid/tests/test_pool2d_op.py
+++ b/python/paddle/v2/fluid/tests/test_pool2d_op.py
--- a/python/paddle/v2/fluid/tests/test_pool3d_op.py
+++ b/python/paddle/v2/fluid/tests/test_pool3d_op.py
--- a/python/paddle/v2/fluid/tests/test_pool_max_op.py
+++ b/python/paddle/v2/fluid/tests/test_pool_max_op.py
--- a/python/paddle/v2/fluid/tests/test_positive_negative_pair_op.py
+++ b/python/paddle/v2/fluid/tests/test_positive_negative_pair_op.py
--- a/python/paddle/v2/fluid/tests/test_precision_recall_op.py
+++ b/python/paddle/v2/fluid/tests/test_precision_recall_op.py
--- a/python/paddle/v2/fluid/tests/test_prelu_op.py
+++ b/python/paddle/v2/fluid/tests/test_prelu_op.py
--- a/python/paddle/v2/fluid/tests/test_print_op.py
+++ b/python/paddle/v2/fluid/tests/test_print_op.py
--- a/python/paddle/v2/fluid/tests/test_prior_box_op.py
+++ b/python/paddle/v2/fluid/tests/test_prior_box_op.py
--- a/python/paddle/v2/fluid/tests/test_profiler.py
+++ b/python/paddle/v2/fluid/tests/test_profiler.py
--- a/python/paddle/v2/fluid/tests/test_program.py
+++ b/python/paddle/v2/fluid/tests/test_program.py
--- a/python/paddle/v2/fluid/tests/test_protobuf.py
+++ b/python/paddle/v2/fluid/tests/test_protobuf.py
--- a/python/paddle/v2/fluid/tests/test_protobuf_descs.py
+++ b/python/paddle/v2/fluid/tests/test_protobuf_descs.py
--- a/python/paddle/v2/fluid/tests/test_proximal_adagrad_op.py
+++ b/python/paddle/v2/fluid/tests/test_proximal_adagrad_op.py
--- a/python/paddle/v2/fluid/tests/test_proximal_gd_op.py
+++ b/python/paddle/v2/fluid/tests/test_proximal_gd_op.py
--- a/python/paddle/v2/fluid/tests/test_rank_loss_op.py
+++ b/python/paddle/v2/fluid/tests/test_rank_loss_op.py
--- a/python/paddle/v2/fluid/tests/test_recurrent_op.py
+++ b/python/paddle/v2/fluid/tests/test_recurrent_op.py
--- a/python/paddle/v2/fluid/tests/test_reduce_op.py
+++ b/python/paddle/v2/fluid/tests/test_reduce_op.py
--- a/python/paddle/v2/fluid/tests/test_registry.py
+++ b/python/paddle/v2/fluid/tests/test_registry.py
--- a/python/paddle/v2/fluid/tests/test_regularizer.py
+++ b/python/paddle/v2/fluid/tests/test_regularizer.py
--- a/python/paddle/v2/fluid/tests/test_reorder_lod_tensor.py
+++ b/python/paddle/v2/fluid/tests/test_reorder_lod_tensor.py
--- a/python/paddle/v2/fluid/tests/test_reshape_op.py
+++ b/python/paddle/v2/fluid/tests/test_reshape_op.py
--- a/python/paddle/v2/fluid/tests/test_rmsprop_op.py
+++ b/python/paddle/v2/fluid/tests/test_rmsprop_op.py
--- a/python/paddle/v2/fluid/tests/test_rnn_memory_helper_op.py
+++ b/python/paddle/v2/fluid/tests/test_rnn_memory_helper_op.py
--- a/python/paddle/v2/fluid/tests/test_roi_pool_op.py
+++ b/python/paddle/v2/fluid/tests/test_roi_pool_op.py
--- a/python/paddle/v2/fluid/tests/test_row_conv_op.py
+++ b/python/paddle/v2/fluid/tests/test_row_conv_op.py
--- a/python/paddle/v2/fluid/tests/test_scale_op.py
+++ b/python/paddle/v2/fluid/tests/test_scale_op.py
--- a/python/paddle/v2/fluid/tests/test_scatter_op.py
+++ b/python/paddle/v2/fluid/tests/test_scatter_op.py
--- a/python/paddle/v2/fluid/tests/test_scope.py
+++ b/python/paddle/v2/fluid/tests/test_scope.py
--- a/python/paddle/v2/fluid/tests/test_selected_rows.py
+++ b/python/paddle/v2/fluid/tests/test_selected_rows.py
--- a/python/paddle/v2/fluid/tests/test_seq_concat_op.py
+++ b/python/paddle/v2/fluid/tests/test_seq_concat_op.py
--- a/python/paddle/v2/fluid/tests/test_seq_conv.py
+++ b/python/paddle/v2/fluid/tests/test_seq_conv.py
--- a/python/paddle/v2/fluid/tests/test_seq_pool.py
+++ b/python/paddle/v2/fluid/tests/test_seq_pool.py
--- a/python/paddle/v2/fluid/tests/test_sequence_erase_op.py
+++ b/python/paddle/v2/fluid/tests/test_sequence_erase_op.py
--- a/python/paddle/v2/fluid/tests/test_sequence_expand.py
+++ b/python/paddle/v2/fluid/tests/test_sequence_expand.py
--- a/python/paddle/v2/fluid/tests/test_sequence_reshape.py
+++ b/python/paddle/v2/fluid/tests/test_sequence_reshape.py
--- a/python/paddle/v2/fluid/tests/test_sequence_slice_op.py
+++ b/python/paddle/v2/fluid/tests/test_sequence_slice_op.py
--- a/python/paddle/v2/fluid/tests/test_sequence_softmax_op.py
+++ b/python/paddle/v2/fluid/tests/test_sequence_softmax_op.py
--- a/python/paddle/v2/fluid/tests/test_sgd_op.py
+++ b/python/paddle/v2/fluid/tests/test_sgd_op.py
--- a/python/paddle/v2/fluid/tests/test_shrink_rnn_memory.py
+++ b/python/paddle/v2/fluid/tests/test_shrink_rnn_memory.py
--- a/python/paddle/v2/fluid/tests/test_sigmoid_cross_entropy_with_logits_op.py
+++ b/python/paddle/v2/fluid/tests/test_sigmoid_cross_entropy_with_logits_op.py
--- a/python/paddle/v2/fluid/tests/test_sign_op.py
+++ b/python/paddle/v2/fluid/tests/test_sign_op.py
--- a/python/paddle/v2/fluid/tests/test_smooth_l1_loss_op.py
+++ b/python/paddle/v2/fluid/tests/test_smooth_l1_loss_op.py
--- a/python/paddle/v2/fluid/tests/test_softmax_op.py
+++ b/python/paddle/v2/fluid/tests/test_softmax_op.py
--- a/python/paddle/v2/fluid/tests/test_softmax_with_cross_entropy_op.py
+++ b/python/paddle/v2/fluid/tests/test_softmax_with_cross_entropy_op.py
--- a/python/paddle/v2/fluid/tests/test_split_and_merge_lod_tensor_op.py
+++ b/python/paddle/v2/fluid/tests/test_split_and_merge_lod_tensor_op.py
--- a/python/paddle/v2/fluid/tests/test_split_op.py
+++ b/python/paddle/v2/fluid/tests/test_split_op.py
--- a/python/paddle/v2/fluid/tests/test_split_selected_rows_op.py
+++ b/python/paddle/v2/fluid/tests/test_split_selected_rows_op.py
--- a/python/paddle/v2/fluid/tests/test_spp_op.py
+++ b/python/paddle/v2/fluid/tests/test_spp_op.py
--- a/python/paddle/v2/fluid/tests/test_squared_l2_distance_op.py
+++ b/python/paddle/v2/fluid/tests/test_squared_l2_distance_op.py
--- a/python/paddle/v2/fluid/tests/test_squared_l2_norm_op.py
+++ b/python/paddle/v2/fluid/tests/test_squared_l2_norm_op.py
--- a/python/paddle/v2/fluid/tests/test_sum_op.py
+++ b/python/paddle/v2/fluid/tests/test_sum_op.py
--- a/python/paddle/v2/fluid/tests/test_tensor.py
+++ b/python/paddle/v2/fluid/tests/test_tensor.py
--- a/python/paddle/v2/fluid/tests/test_top_k_op.py
+++ b/python/paddle/v2/fluid/tests/test_top_k_op.py
--- a/python/paddle/v2/fluid/tests/test_transpose_op.py
+++ b/python/paddle/v2/fluid/tests/test_transpose_op.py
--- a/python/paddle/v2/fluid/tests/test_uniform_random_op.py
+++ b/python/paddle/v2/fluid/tests/test_uniform_random_op.py
--- a/python/paddle/v2/fluid/tests/test_unpool_op.py
+++ b/python/paddle/v2/fluid/tests/test_unpool_op.py
--- a/python/paddle/v2/fluid/tests/test_variable.py
+++ b/python/paddle/v2/fluid/tests/test_variable.py
--- a/python/paddle/v2/fluid/tests/test_warpctc_op.py
+++ b/python/paddle/v2/fluid/tests/test_warpctc_op.py
--- a/python/paddle/v2/fluid/tests/test_while_op.py
+++ b/python/paddle/v2/fluid/tests/test_while_op.py
--- a/python/paddle/v2/image.py
+++ b/python/paddle/v2/image.py
--- a/python/paddle/v2/inference.py
+++ b/python/paddle/v2/inference.py
--- a/python/paddle/v2/master/__init__.py
+++ b/python/paddle/v2/master/__init__.py
--- a/python/paddle/v2/master/client.py
+++ b/python/paddle/v2/master/client.py
--- a/python/paddle/v2/reader/tests/__init__.py
+++ b/python/paddle/v2/reader/tests/__init__.py
--- a/python/paddle/v2/reader/tests/creator_test.py
+++ b/python/paddle/v2/reader/tests/creator_test.py
--- a/python/paddle/v2/reader/tests/decorator_test.py
+++ b/python/paddle/v2/reader/tests/decorator_test.py
--- a/python/paddle/v2/tests/test_image.py
+++ b/python/paddle/v2/tests/test_image.py
--- a/python/paddle/v2/tests/test_layer.py
+++ b/python/paddle/v2/tests/test_layer.py
--- a/python/paddle/v2/tests/test_op.py
+++ b/python/paddle/v2/tests/test_op.py
--- a/python/paddle/v2/tests/test_paramconf_order.py
+++ b/python/paddle/v2/tests/test_paramconf_order.py
--- a/python/paddle/v2/tests/test_parameters.py
+++ b/python/paddle/v2/tests/test_parameters.py
--- a/python/paddle/v2/tests/test_rnn_layer.py
+++ b/python/paddle/v2/tests/test_rnn_layer.py
--- a/python/paddle/v2/tests/test_topology.py
+++ b/python/paddle/v2/tests/test_topology.py
--- a/python/paddle/v2/trainer.py
+++ b/python/paddle/v2/trainer.py
--- a/tools/manylinux1/Dockerfile.x64
+++ b/tools/manylinux1/Dockerfile.x64
--- a/tools/manylinux1/build_scripts/manylinux1-check.py
+++ b/tools/manylinux1/build_scripts/manylinux1-check.py
--- a/tools/manylinux1/build_scripts/python-tag-abi-tag.py
+++ b/tools/manylinux1/build_scripts/python-tag-abi-tag.py
--- a/tools/manylinux1/build_scripts/ssl-check.py
+++ b/tools/manylinux1/build_scripts/ssl-check.py