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

.copyright.hook

+from __future__ import absolute_import
+from __future__ import print_function
+from __future__ import unicode_literals
+
+import argparse
+import io, re
+import sys, os
+import subprocess
+import platform
+
+COPYRIGHT = '''
+  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.
+'''
+
+LANG_COMMENT_MARK = None
+
+NEW_LINE_MARK = None
+
+COPYRIGHT_HEADER = None
+
+if platform.system() == "Windows":
+    NEW_LINE_MARK = "\r\n"
+else:
+    NEW_LINE_MARK = '\n'
+    COPYRIGHT_HEADER = COPYRIGHT.split(NEW_LINE_MARK)[1]
+    p = re.search('(\d{4})', COPYRIGHT_HEADER).group(0)
+    process = subprocess.Popen(["date", "+%Y"], stdout=subprocess.PIPE)
+    date, err = process.communicate()
+    date = date.decode("utf-8").rstrip("\n")
+    COPYRIGHT_HEADER = COPYRIGHT_HEADER.replace(p, date)
+
+
+def generate_copyright(template, lang='C'):
+    if lang == 'Python':
+        LANG_COMMENT_MARK = '#'
+    else:
+        LANG_COMMENT_MARK = "//"
+
+    lines = template.split(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
+        if len(line)  == 0:
+            BLANK = ""
+        else:
+            BLANK = " "
+        ans += LANG_COMMENT_MARK + BLANK + line + NEW_LINE_MARK
+
+    return ans + "\n"
+
+
+def lang_type(filename):
+    if filename.endswith(".py"):
+        return "Python"
+    elif filename.endswith(".h"):
+        return "C"
+    elif filename.endswith(".c"):
+        return "C"
+    elif filename.endswith(".hpp"):
+        return "C"
+    elif filename.endswith(".cc"):
+        return "C"
+    elif filename.endswith(".cpp"):
+        return "C"
+    elif filename.endswith(".cu"):
+        return "C"
+    elif filename.endswith(".cuh"):
+        return "C"
+    elif filename.endswith(".go"):
+        return "C"
+    elif filename.endswith(".proto"):
+        return "C"
+    else:
+        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.')
+    parser.add_argument('filenames', nargs='*', help='Filenames to check')
+    args = parser.parse_args(argv)
+
+    retv = 0
+    for filename in args.filenames:
+        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))
+        retv = 1
+        with io.open(filename, 'w') as output_file:
+            output_file.write(new_contents)
+
+    return retv
+
+
+if __name__ == '__main__':
+    exit(main())

.pre-commit-config.yaml

@@ -31,3 +31,11 @@
     -   id: go-fmt
         types:
         - go
+-   repo: local
+    hooks:
+    -   id: copyright_checker
+        name: copyright_checker
+        entry: python ./.copyright.hook
+        language: system
+        files: \.(c|cc|cxx|cpp|cu|h|hpp|hxx|proto|py)$
+        exclude: (?!.*third_party)^.*$ | (?!.*book)^.*$

CODE_OF_CONDUCT.md

+# Contributor Covenant Code of Conduct
+
+## Our Pledge
+
+In the interest of fostering an open and welcoming environment, we as contributors and maintainers pledge to making participation in our project and our community a harassment-free experience for everyone, regardless of age, body size, disability, ethnicity, gender identity and expression, level of experience, nationality, personal appearance, race, religion, or sexual identity and orientation.
+
+## Our Standards
+
+Examples of behavior that contributes to creating a positive environment include:
+
+* Using welcoming and inclusive language
+* Being respectful of differing viewpoints and experiences
+* Gracefully accepting constructive criticism
+* Focusing on what is best for the community
+* Showing empathy towards other community members
+
+Examples of unacceptable behavior by participants include:
+
+* The use of sexualized language or imagery and unwelcome sexual attention or advances
+* Trolling, insulting/derogatory comments, and personal or political attacks
+* Public or private harassment
+* Publishing others' private information, such as a physical or electronic address, without explicit permission
+* Other conduct which could reasonably be considered inappropriate in a professional setting
+
+## Our Responsibilities
+
+Project maintainers are responsible for clarifying the standards of acceptable behavior and are expected to take appropriate and fair corrective action in response to any instances of unacceptable behavior.
+
+Project maintainers have the right and responsibility to remove, edit, or reject comments, commits, code, wiki edits, issues, and other contributions that are not aligned to this Code of Conduct, or to ban temporarily or permanently any contributor for other behaviors that they deem inappropriate, threatening, offensive, or harmful.
+
+## Scope
+
+This Code of Conduct applies both within project spaces and in public spaces when an individual is representing the project or its community. Examples of representing a project or community include using an official project e-mail address, posting via an official social media account, or acting as an appointed representative at an online or offline event. Representation of a project may be further defined and clarified by project maintainers.
+
+## Enforcement
+
+Instances of abusive, harassing, or otherwise unacceptable behavior may be reported by contacting the project team at paddle-dev@baidu.com. The project team will review and investigate all complaints, and will respond in a way that it deems appropriate to the circumstances. The project team is obligated to maintain confidentiality with regard to the reporter of an incident. Further details of specific enforcement policies may be posted separately.
+
+Project maintainers who do not follow or enforce the Code of Conduct in good faith may face temporary or permanent repercussions as determined by other members of the project's leadership.
+
+## Attribution
+
+This Code of Conduct is adapted from the [Contributor Covenant][homepage], version 1.4, available at [http://contributor-covenant.org/version/1/4][version]
+
+[homepage]: http://contributor-covenant.org
+[version]: http://contributor-covenant.org/version/1/4/

CODE_OF_CONDUCT_cn.md

+# 参与者公约
+
+## 我们的保证
+
+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。
+
+## 我们的标准
+
+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
+
+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
+
+## 我们的责任
+
+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。
+
+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。
+
+## 使用范围
+
+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。
+
+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。
+
+## 强制执行
+
+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。
+
+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。
+
+## 来源
+
+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
+
+[主页]: https://www.contributor-covenant.org

README.md

@@ -37,6 +37,7 @@ Please refer to our [release announcement](https://github.com/PaddlePaddle/Paddl
 
       - Optimized math operations through SSE/AVX intrinsics, BLAS libraries
       (e.g. MKL, OpenBLAS, cuBLAS) or customized CPU/GPU kernels.
+      - Optimized CNN networks through MKL-DNN library.
       - Highly optimized recurrent networks which can handle **variable-length**
       sequence without padding.
       - Optimized local and distributed training for models with high dimensional

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.

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 
-"""

adversarial/advbox/attacks/base.py

-"""
-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

adversarial/advbox/attacks/gradientsign.py

-"""
-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

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 
-"""

adversarial/advbox/models/base.py

-"""
-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

adversarial/advbox/models/paddle.py

-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

adversarial/fluid_mnist.py

-"""
-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()

adversarial/mnist_tutorial_fgsm.py

-"""
-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()

benchmark/paddle/image/alexnet.py

-#!/usr/bin/env python
+#   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 *
 

benchmark/paddle/image/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.
+
 import io, os
 import random
 import numpy as np

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

benchmark/paddle/rnn/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.
+
 import io, os
 import random
 import numpy as np

benchmark/tensorflow/image/alexnet.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 six.moves import xrange  # pylint: disable=redefined-builtin
 from datetime import datetime
 import math

benchmark/tensorflow/image/alexnet_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.
+
 from six.moves import xrange  # pylint: disable=redefined-builtin
 from datetime import datetime
 import math

benchmark/tensorflow/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.
+
 from six.moves import xrange
 from datetime import datetime
 import math

benchmark/tensorflow/image/googlenet_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.
+
 from six.moves import xrange  # pylint: disable=redefined-builtin
 from datetime import datetime
 import math

benchmark/tensorflow/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.
+
 from six.moves import xrange  # pylint: disable=redefined-builtin
 from datetime import datetime
 import math

benchmark/tensorflow/rnn/reader.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 os.path
 import io
 import numpy as np

cmake/external/grpc.cmake

@@ -33,7 +33,7 @@ ExternalProject_Add(
     extern_grpc
     DEPENDS protobuf zlib
     GIT_REPOSITORY "https://github.com/grpc/grpc.git"
-    GIT_TAG "v1.7.x"
+    GIT_TAG "v1.8.x"
     PREFIX          ${GRPC_SOURCES_DIR}
     UPDATE_COMMAND  ""
     CONFIGURE_COMMAND ""

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})
 

cmake/make_resource.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 os
 import re
 import sys

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(

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 *
 
 

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()

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 *
 
 

doc/api/v2/fluid/layers.rst

@@ -358,6 +358,18 @@ reduce_min
 ..  autofunction:: paddle.v2.fluid.layers.reduce_min
     :noindex:
 
+
+split
+-----
+..  autofunction:: paddle.v2.fluid.layers.split
+    :noindex:
+
+
+matmul
+------
+..  autofunction:: paddle.v2.fluid.layers.matmul
+    :noindex:
+
 logsigmoid
 ----------
 ..  autofunction:: paddle.v2.fluid.layers.logsigmoid
@@ -487,3 +499,13 @@ swish
 ------
 ..  autofunction:: paddle.v2.fluid.layers.swish
     :noindex:
+
+l2_normalize
+------------
+..  autofunction:: paddle.v2.fluid.layers.l2_normalize
+    :noindex:
+
+sequence_reshape
+----------------
+..  autofunction:: paddle.v2.fluid.layers.sequence_reshape
+    :noindex:

doc/api/v2/fluid/nets.rst

@@ -20,3 +20,14 @@ sequence_conv_pool
     :noindex:
 
 
+glu
+---
+..  autofunction:: paddle.v2.fluid.nets.glu
+    :noindex:
+
+
+dot_product_attention
+---------------------
+..  autofunction:: paddle.v2.fluid.nets.dot_product_attention
+    :noindex:
+

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)
 ```

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)

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).

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)
+```

doc/faq/local/src/reduce_min_pool_size.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.
+
+
 @provider(min_pool_size=0, ...)
 def process(settings, filename):
     os.system('shuf %s > %s.shuf' % (filename, filename))  # shuffle before.

doc/faq/local/src/word2vec_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.
+
 ...  # the settings and define data provider is omitted.
 DICT_DIM = 3000  # dictionary dimension.
 word_ids = data_layer('word_ids', size=DICT_DIM)

doc/faq/local/src/word2vec_dataprovider.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.
+
 DICT_DIM = 3000
 
 

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。

doc/getstarted/concepts/src/infer.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 numpy as np
 

doc/getstarted/concepts/src/train.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 numpy as np
 

doc/howto/dev/new_op_cn.md

@@ -24,7 +24,7 @@
 - `framework::OperatorWithKernel`：继承自OperatorBase，Op有计算函数，称作有Kernel。
 - `class OpProtoAndCheckerMaker`：描述该Op的输入、输出、属性、注释,主要用于Python API接口生成
 
-依据是否包含kernel，可以将Op分为两种：包含Kernel的Op和不包含kernel的Op，前者Op的定义继承自`OperatorBase`，后者继承自`OperatorWithKernel`。本教程主要介绍带Kernel的Op如何写，简单总结Op需要包含的内容如下：
+依据是否包含kernel，可以将Op分为两种：包含Kernel的Op和不包含kernel的Op，前者Op的定义继承自`OperatorWithKernel`，后者继承自`OperatorBase`。本教程主要介绍带Kernel的Op如何写，简单总结Op需要包含的内容如下：
 
 
  内容            | 定义位置

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
 ```
 

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`.

doc/howto/usage/capi/organization_of_the_inputs_cn.md

@@ -19,7 +19,7 @@
 
 ### 基本使用概念
 
-- 在PaddlePaddle内部，神经网络中一个计算层的输入/输出被组织为一个 `Argument` 结构体，如果神经网络有多个输入或者多个输入，每一个输入/输入都会对应有自己的`Argument`。
+- 在PaddlePaddle内部，神经网络中一个计算层的输入/输出被组织为一个 `Argument` 结构体，如果神经网络有多个输入或者多个输出，每一个输入/输出都会对应有自己的`Argument`。
 - `Argument` 并不真正“存储”数据，而是将输入/输出信息有机地组织在一起。
 - 在`Argument`内部由`IVector`（对应着上文提到的一维整型数组）和`Matrix`（对应着上文提到的二维浮点型矩阵）来实际存储数据；由 `Sequence Start Positions` (下文详细解释) 来描述输入/输出的序列信息。
 

doc/howto/usage/cluster/fluid_cluster_train_en.md

+# Fluid Distributed Training
+
+## Introduction
+
+In this article, we'll explain how to config and run distributed training jobs with PaddlePaddle Fluid in a bare metal cluster.
+
+## Preparations
+
+### Get your 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.
+
+### 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).
+
+### Update 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:
+
+``` python
+import paddle.v2 as paddle
+import paddle.v2.fluid as fluid
+
+x = fluid.layers.data(name='x', shape=[13], dtype='float32')
+y_predict = fluid.layers.fc(input=x, size=1, act=None)
+y = fluid.layers.data(name='y', shape=[1], dtype='float32')
+
+cost = fluid.layers.square_error_cost(input=y_predict, label=y)
+avg_cost = fluid.layers.mean(x=cost)
+
+sgd_optimizer = fluid.optimizer.SGD(learning_rate=0.001)
+sgd_optimizer.minimize(avg_cost)
+
+BATCH_SIZE = 20
+
+train_reader = paddle.batch(
+    paddle.reader.shuffle(
+        paddle.dataset.uci_housing.train(), buf_size=500),
+    batch_size=BATCH_SIZE)
+
+place = fluid.CPUPlace()
+feeder = fluid.DataFeeder(place=place, feed_list=[x, y])
+exe = fluid.Executor(place)
+
+exe.run(fluid.default_startup_program())
+
+PASS_NUM = 100
+for pass_id in range(PASS_NUM):
+    fluid.io.save_persistables(exe, "./fit_a_line.model/")
+    fluid.io.load_persistables(exe, "./fit_a_line.model/")
+    for data in train_reader():
+        avg_loss_value, = exe.run(fluid.default_main_program(),
+                                  feed=feeder.feed(data),
+                                  fetch_list=[avg_cost])
+
+        if avg_loss_value[0] < 10.0:
+            exit(0)  # if avg cost less than 10.0, we think our code is good.
+exit(1)
+```
+
+We created a simple fully connected neural networks 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.
+
+#### 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.
+
+![parameter server architect](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).
+
+Now we need to create program 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.
+
+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.
+
+Optimize OPs and gradient parameters can be found from the return values of optimizer's minimize function.
+
+To put them together:
+
+``` python
+... #define the program, cost, and create sgd optimizer
+
+optimize_ops, params_grads = sgd_optimizer.minimize(avg_cost) #get optimize OPs and gradient parameters
+
+t = fluid.DistributeTranspiler() # create 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) 
+
+... #create executor
+
+# in pserver, run this
+#current_endpoint here means current pserver IP:PORT you wish to run on
+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
+exe.run(fluid.default_startup_program())
+for pass_id in range(100):
+    for data in train_reader():
+        exe.run(t.get_trainer_program())
+
+
+```
+
+### 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:
+
+``` bash
+PSERVERS=192.168.1.2:6174 SERVER_ENDPOINT=192.168.1.2:6174 TRAINING_ROLE=PSERVER python notest_dist_fit_a_line.py
+```
+
+*please note we assume that your parameter server runs at 192.168.1.2:6174*
+
+Wait until the prompt `Server listening on 192.168.1.2:6174`
+
+Then in 2 of your trainer node 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*
+
+Now you have 2 trainers and 1 parameter server up and running.

doc/howto/usage/cluster/src/word2vec/api_train_v2.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 gzip
 import math
 

doc/howto/usage/cluster/src/word2vec/api_train_v2_cluster.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 math
 import os
 import paddle.v2 as paddle

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

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.

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

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 ','

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>

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
-```

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
-```

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
-    ===========================================  ==============================================================

go/pserver/client/c/test/test_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.
+
 import paddle.v2 as paddle
 import gzip
 

go/pserver/client/c/test/test_train.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 paddle.v2.dataset.uci_housing as uci_housing
 import paddle.v2.master as master

paddle/api/test/testTrainConfig.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 *
 
 settings(batch_size=100, learning_method=AdamOptimizer())

paddle/capi/examples/model_inference/common/common.h

+//  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.
 #ifndef __CAPI_EXAMPLE_COMMON_H__
 #define __CAPI_EXAMPLE_COMMON_H__
 #include <stdio.h>

paddle/capi/examples/model_inference/dense/main.c

+//   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.
+
 #include <paddle/capi.h>
 #include <time.h>
 

paddle/capi/examples/model_inference/dense/merge_v2_model.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.utils.merge_model import merge_v2_model
 
 from mnist_v2 import network

paddle/capi/examples/model_inference/dense/mnist_v2.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 os
 import sys
 import gzip

paddle/capi/examples/model_inference/dense/trainer_config.py

-from paddle.trainer_config_helpers import *
-
-img = data_layer(name='pixel', size=784)
-
-hidden = fc_layer(
-    input=img,
-    size=200,
-    param_attr=ParamAttr(name='hidden.w'),
-    bias_attr=ParamAttr(name='hidden.b'))
-
-prob = fc_layer(
-    input=hidden,
-    size=10,
-    act=SoftmaxActivation(),
-    param_attr=ParamAttr(name='prob.w'),
-    bias_attr=ParamAttr(name='prob.b'))
-
-outputs(prob)
+#   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.

paddle/capi/examples/model_inference/multi_thread/main.c

+//   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.
+
 #include <paddle/capi.h>
 #include <pthread.h>
 #include <time.h>

paddle/capi/examples/model_inference/multi_thread/main_gpu.c

+//   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.
+
 #include <paddle/capi.h>
 #include <pthread.h>
 #include <time.h>

paddle/capi/examples/model_inference/sequence/main.c

+//   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.
+
 #include <paddle/capi.h>
 #include <time.h>
 #include "../common/common.h"

paddle/capi/examples/model_inference/sequence/trainer_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 *
 
 WORD_DIM = 3000

paddle/capi/examples/model_inference/sparse_binary/main.c

+//   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.
+
 #include <paddle/capi.h>
 #include <time.h>
 

paddle/capi/tests/test_predict_network.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 *
 
 settings(batch_size=100)

paddle/cuda/src/avx_mathfun.h

+//  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.
 /*
    AVX implementation of sin, cos, sincos, exp and log
 

paddle/framework/CMakeLists.txt

@@ -33,8 +33,14 @@ cc_library(scope SRCS scope.cc DEPS glog threadpool)
 cc_test(scope_test SRCS scope_test.cc DEPS scope)
 
 cc_library(data_device_transform SRCS data_device_transform.cc DEPS tensor)
+nv_test(data_device_transform_test SRCS data_device_transform_test.cu
+        DEPS operator op_registry init math_function)
+
 cc_library(data_type_transform SRCS data_type_transform.cc DEPS tensor)
+cc_test(data_type_transform_test SRCS data_type_transform_test.cc DEPS data_type_transform)
+
 cc_library(data_layout_transform SRCS data_layout_transform.cc DEPS tensor math_function)
+cc_test(data_layout_transform_test SRCS data_layout_transform_test.cc DEPS data_layout_transform)
 
 cc_library(data_transform SRCS data_transform.cc DEPS math_function tensor
         framework_proto selected_rows data_device_transform data_type_transform data_layout_transform)
@@ -83,5 +89,3 @@ cc_test(init_test SRCS init_test.cc DEPS init)
 
 cc_test(op_kernel_type_test SRCS op_kernel_type_test.cc DEPS place device_context framework_proto)
 cc_test(cow_ptr_tests SRCS details/cow_ptr_test.cc)
-nv_test(data_device_transform_test SRCS data_device_transform_test.cu
-        DEPS operator op_registry init math_function)

paddle/framework/backward_test.cc

-/* 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. */
+//  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.
 
 #include "paddle/framework/backward.h"
 

paddle/framework/data_device_transform.cc

@@ -31,15 +31,14 @@ static const platform::DeviceContext* GetDeviceContext(
   }
 }
 
-Tensor* DeviceTransform(const Tensor& in, const platform::Place& dst_place) {
+void TransDataDevice(const Tensor& in, const platform::Place& dst_place,
+                     Tensor* out) {
   VLOG(3) << "DeviceTransform in, src_place " << in.place()
           << " dst_place: " << dst_place;
-  Tensor* out = new Tensor();
   auto* dev_ctx = GetDeviceContext(in.place(), dst_place);
   dev_ctx->Wait();
   Copy(in, dst_place, *dev_ctx, out);
   dev_ctx->Wait();
-  return out;
 }
 
 }  // namespace framework

paddle/framework/data_device_transform.h

@@ -21,7 +21,8 @@ limitations under the License. */
 namespace paddle {
 namespace framework {
 
-Tensor* DeviceTransform(const Tensor& in, const platform::Place& dst_place);
+void TransDataDevice(const Tensor& in, const platform::Place& dst_place,
+                     Tensor* out);
 
 }  // namespace framework
 }  // namespace paddle

paddle/framework/data_device_transform_test.cu

@@ -150,6 +150,7 @@ TEST(Operator, CPUtoGPU) {
   // get output
   auto* output2 = scope.Var("OUT2");
   gpu_op->Run(scope, cuda_place);
+  VLOG(3) << "after gpu_op run";
 
   // auto* output2_ptr = output2->Get<LoDTensor>().data<float>();
   DeviceContextPool& pool = DeviceContextPool::Instance();