Skip to content

P
Paddle

PaddlePaddle / Paddle
大约 2 年 前同步成功

通知 2325
Star 20933
Fork 5424
P
Paddle
提交 d95d66cb 编写于 作者: T Travis CI
浏览文件
操作

Deploy to GitHub Pages: d924a658

上级 97c432ee
隐藏空白更改
内联 并排
Showing with 532 addition and 4 deletion
develop/doc_cn/_sources/howto/dev/new_op_cn.md.txt
浏览文件 @ d95d66cb
......@@ -169,6 +169,8 @@ class MulKernel : public framework::OpKernel {
`MulKernel`需要重写`Compute`接口，该接口参数为`const framework::ExecutionContext& context`, `ExecutionContext`相比`InferShapeContext`增加了设备类型，同样可获取到输入输出和属性参数，`Compute`函数里写具体实现时。
注意，不同设备(CPU、GPU)共享一个Op定义，是否则共享同一个`OpKernel`，取决于`Compute`调用的函数是否支持不同设备。`MulOp`的CPU、GPU实现共享同一个`Kernel`，`OpKernel`不共享的例子可以参考[`OnehotCrossEntropyOpKernel`](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/operators/cross_entropy_op.h#L43)。
为了使得`OpKernel`的计算过程书写较为简单，CPU、GPU的代码可以复用，我们通常借助Eigen unsupported Tensor模块来实现。关于在paddle中如何使用Eigen库，请参考对应的使用[文档](https://github.com/PaddlePaddle/Paddle/blob/develop/doc/howto/dev/use_eigen_cn.md)
到此前向Op实现完成，需要在`.cc`文件中注册该op和kernel。反向Op类的定义和Kernel定义与前向Op类似，这里不再重复。但注意，反向Op没有`ProtoMaker`。
......@@ -188,9 +190,12 @@ REGISTER_OP_CPU_KERNEL(mul_grad,
- `REGISTER_OP_WITHOUT_GRADIENT` ： 用于注册没有反向的Op。
- `REGISTER_OP_CPU_KERNEL` ：注册`ops::MulKernel`类，并特化模板参数为`paddle::platform::CPUPlace`和`float`类型，同理，注册`ops::MulKernel`类。
在 `.cu`文件中注册GPU Kernel。
在 `.cu`文件中注册GPU Kernel。请注意，如果GPU Kernel的实现是基于Eigen unsupported模块，那么在 `.cu`的最前面请加上宏定义 `#define EIGEN_USE_GPU`
```c++
// if use Eigen unsupported module before include head files
#define EIGEN_USE_GPU
namespace ops = paddle::operators;
REGISTER_OP_GPU_KERNEL(mul, ops::MulKernel<paddle::platform::GPUPlace, float>);
REGISTER_OP_GPU_KERNEL(mul_grad,
......
develop/doc_cn/_sources/howto/dev/use_eigen_cn.md.txt 0 → 100644
浏览文件 @ d95d66cb
## 在Paddle中如何使用Eigen
神经网络本质上是一个计算图，计算需要的数据存放在`Tensor`中，而计算过程是由`Operartor`来描述的。在执行时，`Operator`调用对应`OpKernel`中的`Compute`接口，实现对`Tensor`的操作。
### Eigen Tensor模块
Eigen Tensor模块对element-wise计算提供了强大的支持，并且书写一份代码，可以同时在CPU、GPU执行。但Eigen Tensor是一个正在开发中的模块，因此可能测试不够完备，文档较少。
关于Eigen Tensor模块的详细介绍请参考[文档1](https://github.com/RLovelett/eigen/blob/master/unsupported/Eigen/CXX11/src/Tensor/README.md) 和[文档2](https://bitbucket.org/eigen/eigen/src/default/unsupported/Eigen/CXX11/src/Tensor/README.md)
### paddle::framework::Tensor
Paddle Tensor定义在framework目录下，其主要接口如下：
```cpp
class Tensor {
public:
/*! Return a pointer to mutable memory block. */
template <typename T>
inline T* data();
/**
* @brief Return a pointer to mutable memory block.
* @note If not exist, then allocation.
*/
template <typename T>
inline T* mutable_data(platform::Place place);
/**
* @brief Return a pointer to mutable memory block.
*
* @param[in] dims The dimensions of the memory block.
* @param[in] place The place of the memory block.
*
* @note If not exist, then allocation.
*/
template <typename T>
inline T* mutable_data(DDim dims, platform::Place place);
/*! Resize the dimensions of the memory block. */
inline Tensor& Resize(const DDim& dims);
/*! Return the dimensions of the memory block. */
inline const DDim& dims() const;
private:
/*! holds the memory block if allocated. */
std::shared_ptr<Placeholder> holder_;
/*! points to dimensions of memory block. */
DDim dim_;
};
```
`Placeholder`的作用是延迟分配内存，即我们可以先定义一个Tensor，然后使用Resize接口设置Tensor的大小，最后再调用mutable_data接口分配实际的内存。
```cpp
paddle::framework::Tensor t;
paddle::platform::CPUPlace place;
// set size first
t.Resize({2, 3});
// allocate memory on CPU later
t.mutable_data(place);
```
### paddle::framework::Tensor使用样例
下面以AddOp为例说明Tensor的使用过程：
- InferShape
在运行神经网络计算图时，我们先调用每个`Operator`的`InferShape`接口，根据输入Tensor的大小来设置输出Tensor的大小，`Resize`接口会被调用。
```cpp
void InferShape(const framework::InferShapeContext &ctx) const override {
PADDLE_ENFORCE_EQ(ctx.Input<Tensor>("X")->dims(),
ctx.Input<Tensor>("Y")->dims(),
"Two input of Add Op's dimension must be same.");
ctx.Output<Tensor>("Out")->Resize(ctx.Input<Tensor>("X")->dims());
}
```
- Run
`Operator`的`Run`接口最终会调用对应`OpKernel`的`Compute`接口，在这时真正的分配内存，`mutable_data`接口会被调用。
```cpp
void Compute(const framework::ExecutionContext& context) const override {
auto* input0 = context.Input<Tensor>("X");
auto* input1 = context.Input<Tensor>("Y");
auto* output = context.Output<Tensor>("Out");
output->mutable_data<T>(context.GetPlace());
auto x = EigenVector<T>::Flatten(*input0);
auto y = EigenVector<T>::Flatten(*input1);
auto z = EigenVector<T>::Flatten(*output);
auto place = context.GetEigenDevice<Place>();
z.device(place) = x + y;
}
```
### paddle::framework::Tensor到EigenTensor的转换
如上一小节所示，在具体的计算中，我们需要先把输入Tensor和输出Tensor转换为Eigen支持的格式。我们在[eigen.h](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/framework/eigen.h)中提供了一些全局函数用来实现paddle::framework::Tensor到EigenTensor/EigenMatrix/EigenVector/EigenScalar的转换。
以EigenTensor为例，做一个介绍
```cpp
Tensor t;
float* p = t.mutable_data<float>(make_ddim({1, 2, 3}), platform::CPUPlace());
for (int i = 0; i < 1 * 2 * 3; i++) {
p[i] = static_cast<float>(i);
}
EigenTensor<float, 3>::Type et = EigenTensor<float, 3>::From(t);
```
From是EigenTensor模板提供的一个接口，可以实现从paddle::framework::Tensor到对EigenTensor的转换。由于Tensor的rank是模板参数，因此在转换时需要显示的指定。
在Eigen中，不同rank的Tensor是不同类型，Vector是rank为1的Tensor。需要额外注意的是，EigenVector<T>::From方法是把paddle中的一维Tensor转为Eigen的一维Tensor，在这里用EigenVector来表示；而EigenVector<T>::Flatten方法是把paddle中的一个Tensor进行reshape操作，压扁成为Eigen的一维Tensor，类型仍然为EigenVector。
更多的转换方法请参考eigen_test.cc中的[单元测试](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/framework/eigen_test.cc)。
### 实现计算
当需要完成计算时，我们需要等式左边的EigenTensor调用device接口。在这里需要注意的是，这里的EigenTensor之间的运算只是改变了原有Tensor中的数据，而不会改变原有Tensor的shape信息。
```cpp
auto x = EigenVector<T>::Flatten(*input0);
auto y = EigenVector<T>::Flatten(*input1);
auto z = EigenVector<T>::Flatten(*output);
auto place = context.GetEigenDevice<Place>();
z.device(place) = x + y;
```
在这段代码中，input0/input1/output可以是任意维度的Tensor。我们调用了EigenVector的Flatten接口，把任意维度的Tensor转为了一维的EigenVector。而在计算结束之后，input0/input1/output的原有shape信息不变。如果想改变原有Tensor的shape信息，可以调用Resize接口进行改变。
由于Eigen Tensor模块的文档较少，我们可以参考TensorFlow的[kernels](https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/kernels)模块下的相关`OpKernel`的计算代码。
develop/doc_cn/howto/dev/new_op_cn.html
浏览文件 @ d95d66cb
......@@ -335,6 +335,7 @@ Kernel实现 | CPU、GPU共享Kernel在<code class="docutils literal"><spa
</ul>
<p><code class="docutils literal"><span class="pre">MulKernel</span></code>需要重写<code class="docutils literal"><span class="pre">Compute</span></code>接口，该接口参数为<code class="docutils literal"><span class="pre">const</span> <span class="pre">framework::ExecutionContext&amp;</span> <span class="pre">context</span></code>, <code class="docutils literal"><span class="pre">ExecutionContext</span></code>相比<code class="docutils literal"><span class="pre">InferShapeContext</span></code>增加了设备类型，同样可获取到输入输出和属性参数，<code class="docutils literal"><span class="pre">Compute</span></code>函数里写具体实现时。</p>
<p>注意，不同设备(CPU、GPU)共享一个Op定义，是否则共享同一个<code class="docutils literal"><span class="pre">OpKernel</span></code>，取决于<code class="docutils literal"><span class="pre">Compute</span></code>调用的函数是否支持不同设备。<code class="docutils literal"><span class="pre">MulOp</span></code>的CPU、GPU实现共享同一个<code class="docutils literal"><span class="pre">Kernel</span></code>，<code class="docutils literal"><span class="pre">OpKernel</span></code>不共享的例子可以参考<a class="reference external" href="https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/operators/cross_entropy_op.h#L43"><code class="docutils literal"><span class="pre">OnehotCrossEntropyOpKernel</span></code></a>。</p>
<p>为了使得<code class="docutils literal"><span class="pre">OpKernel</span></code>的计算过程书写较为简单，CPU、GPU的代码可以复用，我们通常借助Eigen unsupported Tensor模块来实现。关于在paddle中如何使用Eigen库，请参考对应的使用<a class="reference external" href="https://github.com/PaddlePaddle/Paddle/blob/develop/doc/howto/dev/use_eigen_cn.md">文档</a></p>
<p>到此前向Op实现完成，需要在<code class="docutils literal"><span class="pre">.cc</span></code>文件中注册该op和kernel。反向Op类的定义和Kernel定义与前向Op类似，这里不再重复。但注意，反向Op没有<code class="docutils literal"><span class="pre">ProtoMaker</span></code>。</p>
</div>
<div class="section" id="operator">
......@@ -352,8 +353,11 @@ Kernel实现 | CPU、GPU共享Kernel在<code class="docutils literal"><spa
<li><code class="docutils literal"><span class="pre">REGISTER_OP_WITHOUT_GRADIENT</span></code> ： 用于注册没有反向的Op。</li>
<li><code class="docutils literal"><span class="pre">REGISTER_OP_CPU_KERNEL</span></code> ：注册<code class="docutils literal"><span class="pre">ops::MulKernel</span></code>类，并特化模板参数为<code class="docutils literal"><span class="pre">paddle::platform::CPUPlace</span></code>和<code class="docutils literal"><span class="pre">float</span></code>类型，同理，注册<code class="docutils literal"><span class="pre">ops::MulKernel</span></code>类。</li>
</ul>
<p>在 <code class="docutils literal"><span class="pre">.cu</span></code>文件中注册GPU Kernel。</p>
<div class="highlight-c++"><div class="highlight"><pre><span></span><span class="k">namespace</span> <span class="n">ops</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">::</span><span class="n">operators</span><span class="p">;</span>
<p>在 <code class="docutils literal"><span class="pre">.cu</span></code>文件中注册GPU Kernel。请注意，如果GPU Kernel的实现是基于Eigen unsupported模块，那么在 <code class="docutils literal"><span class="pre">.cu</span></code>的最前面请加上宏定义 <code class="docutils literal"><span class="pre">#define</span> <span class="pre">EIGEN_USE_GPU</span></code></p>
<div class="highlight-c++"><div class="highlight"><pre><span></span><span class="c1">// if use Eigen unsupported module before include head files</span>
<span class="cp">#define EIGEN_USE_GPU</span>
<span class="k">namespace</span> <span class="n">ops</span> <span class="o">=</span> <span class="n">paddle</span><span class="o">::</span><span class="n">operators</span><span class="p">;</span>
<span class="n">REGISTER_OP_GPU_KERNEL</span><span class="p">(</span><span class="n">mul</span><span class="p">,</span> <span class="n">ops</span><span class="o">::</span><span class="n">MulKernel</span><span class="o">&lt;</span><span class="n">paddle</span><span class="o">::</span><span class="n">platform</span><span class="o">::</span><span class="n">GPUPlace</span><span class="p">,</span> <span class="kt">float</span><span class="o">&gt;</span><span class="p">);</span>
<span class="n">REGISTER_OP_GPU_KERNEL</span><span class="p">(</span><span class="n">mul_grad</span><span class="p">,</span>
<span class="n">ops</span><span class="o">::</span><span class="n">MulGradKernel</span><span class="o">&lt;</span><span class="n">paddle</span><span class="o">::</span><span class="n">platform</span><span class="o">::</span><span class="n">GPUPlace</span><span class="p">,</span> <span class="kt">float</span><span class="o">&gt;</span><span class="p">);</span>
......
develop/doc_cn/howto/dev/use_eigen_cn.html 0 → 100644
浏览文件 @ d95d66cb
<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>在Paddle中如何使用Eigen &mdash; PaddlePaddle 文档</title>
<link rel="stylesheet" href="../../_static/css/theme.css" type="text/css" />
<link rel="index" title="索引"
href="../../genindex.html"/>
<link rel="search" title="搜索" href="../../search.html"/>
<link rel="top" title="PaddlePaddle 文档" href="../../index.html"/>
<link rel="stylesheet" href="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/css/perfect-scrollbar.min.css" type="text/css" />
<link rel="stylesheet" href="../../_static/css/override.css" type="text/css" />
<script>
var _hmt = _hmt || [];
(function() {
var hm = document.createElement("script");
hm.src = "//hm.baidu.com/hm.js?b9a314ab40d04d805655aab1deee08ba";
var s = document.getElementsByTagName("script")[0];
s.parentNode.insertBefore(hm, s);
})();
</script>
<script src="../../_static/js/modernizr.min.js"></script>
</head>
<body class="wy-body-for-nav" role="document">
<header class="site-header">
<div class="site-logo">
<a href="/"><img src="../../_static/images/PP_w.png"></a>
</div>
<div class="site-nav-links">
<div class="site-menu">
<a class="fork-on-github" href="https://github.com/PaddlePaddle/Paddle" target="_blank"><i class="fa fa-github"></i>Fork me on Github</a>
<div class="language-switcher dropdown">
<a type="button" data-toggle="dropdown">
<span>English</span>
<i class="fa fa-angle-up"></i>
<i class="fa fa-angle-down"></i>
</a>
<ul class="dropdown-menu">
<li><a href="/doc_cn">中文</a></li>
<li><a href="/doc">English</a></li>
</ul>
</div>
<ul class="site-page-links">
<li><a href="/">Home</a></li>
</ul>
</div>
<div class="doc-module">
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a></li>
<li class="toctree-l1"><a class="reference internal" href="../index_cn.html">进阶指南</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a></li>
<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
</ul>
<div role="search">
<form id="rtd-search-form" class="wy-form" action="../../search.html" method="get">
<input type="text" name="q" placeholder="Search docs" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
</div>
</div>
</div>
</header>
<div class="main-content-wrap">
<nav class="doc-menu-vertical" role="navigation">
<ul>
<li class="toctree-l1"><a class="reference internal" href="../../getstarted/index_cn.html">新手入门</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../getstarted/build_and_install/index_cn.html">安装与编译</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/docker_install_cn.html">PaddlePaddle的Docker容器使用方式</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../getstarted/build_and_install/cmake/build_from_source_cn.html">PaddlePaddle的编译选项</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../getstarted/concepts/use_concepts_cn.html">基本使用概念</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../index_cn.html">进阶指南</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../usage/cmd_parameter/index_cn.html">设置命令行参数</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../usage/cmd_parameter/use_case_cn.html">使用案例</a></li>
<li class="toctree-l3"><a class="reference internal" href="../usage/cmd_parameter/arguments_cn.html">参数概述</a></li>
<li class="toctree-l3"><a class="reference internal" href="../usage/cmd_parameter/detail_introduction_cn.html">细节描述</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../usage/cluster/cluster_train_cn.html">运行分布式训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="../usage/k8s/k8s_basis_cn.html">Kubernetes 简介</a></li>
<li class="toctree-l2"><a class="reference internal" href="../usage/k8s/k8s_cn.html">Kubernetes单机训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="../usage/k8s/k8s_distributed_cn.html">Kubernetes分布式训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="build_cn.html">编译PaddlePaddle和运行单元测试</a></li>
<li class="toctree-l2"><a class="reference internal" href="write_docs_cn.html">如何贡献/修改文档</a></li>
<li class="toctree-l2"><a class="reference internal" href="contribute_to_paddle_cn.html">如何贡献代码</a></li>
<li class="toctree-l2"><a class="reference internal" href="../deep_model/rnn/index_cn.html">RNN相关模型</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../deep_model/rnn/rnn_config_cn.html">RNN配置</a></li>
<li class="toctree-l3"><a class="reference internal" href="../deep_model/rnn/recurrent_group_cn.html">Recurrent Group教程</a></li>
<li class="toctree-l3"><a class="reference internal" href="../deep_model/rnn/hierarchical_layer_cn.html">支持双层序列作为输入的Layer</a></li>
<li class="toctree-l3"><a class="reference internal" href="../deep_model/rnn/hrnn_rnn_api_compare_cn.html">单双层RNN API对比介绍</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../optimization/gpu_profiling_cn.html">GPU性能分析与调优</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../api/index_cn.html">API</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../../api/v2/model_configs.html">模型配置</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/activation.html">Activation</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/layer.html">Layers</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/evaluators.html">Evaluators</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/optimizer.html">Optimizer</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/pooling.html">Pooling</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/networks.html">Networks</a></li>
<li class="toctree-l3"><a class="reference internal" href="../../api/v2/config/attr.html">Parameter Attribute</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../../api/v2/data.html">数据访问</a></li>
<li class="toctree-l2"><a class="reference internal" href="../../api/v2/run_logic.html">训练与应用</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../../faq/index_cn.html">FAQ</a></li>
</ul>
</nav>
<section class="doc-content-wrap">
<div role="navigation" aria-label="breadcrumbs navigation">
<ul class="wy-breadcrumbs">
<li>在Paddle中如何使用Eigen</li>
</ul>
</div>
<div class="wy-nav-content" id="doc-content">
<div class="rst-content">
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
<div itemprop="articleBody">
<div class="section" id="paddleeigen">
<span id="paddleeigen"></span><h1>在Paddle中如何使用Eigen<a class="headerlink" href="#paddleeigen" title="永久链接至标题"></a></h1>
<p>神经网络本质上是一个计算图，计算需要的数据存放在<code class="docutils literal"><span class="pre">Tensor</span></code>中，而计算过程是由<code class="docutils literal"><span class="pre">Operartor</span></code>来描述的。在执行时，<code class="docutils literal"><span class="pre">Operator</span></code>调用对应<code class="docutils literal"><span class="pre">OpKernel</span></code>中的<code class="docutils literal"><span class="pre">Compute</span></code>接口，实现对<code class="docutils literal"><span class="pre">Tensor</span></code>的操作。</p>
<div class="section" id="eigen-tensor">
<span id="eigen-tensor"></span><h2>Eigen Tensor模块<a class="headerlink" href="#eigen-tensor" title="永久链接至标题"></a></h2>
<p>Eigen Tensor模块对element-wise计算提供了强大的支持，并且书写一份代码，可以同时在CPU、GPU执行。但Eigen Tensor是一个正在开发中的模块，因此可能测试不够完备，文档较少。</p>
<p>关于Eigen Tensor模块的详细介绍请参考<a class="reference external" href="https://github.com/RLovelett/eigen/blob/master/unsupported/Eigen/CXX11/src/Tensor/README.md">文档1</a> 和<a class="reference external" href="https://bitbucket.org/eigen/eigen/src/default/unsupported/Eigen/CXX11/src/Tensor/README.md">文档2</a></p>
</div>
<div class="section" id="paddle-framework-tensor">
<span id="paddle-framework-tensor"></span><h2>paddle::framework::Tensor<a class="headerlink" href="#paddle-framework-tensor" title="永久链接至标题"></a></h2>
<p>Paddle Tensor定义在framework目录下，其主要接口如下：</p>
<div class="highlight-cpp"><div class="highlight"><pre><span></span><span class="k">class</span> <span class="nc">Tensor</span> <span class="p">{</span>
<span class="k">public</span><span class="o">:</span>
<span class="cm">/*! Return a pointer to mutable memory block. */</span>
<span class="k">template</span> <span class="o">&lt;</span><span class="k">typename</span> <span class="n">T</span><span class="o">&gt;</span>
<span class="kr">inline</span> <span class="n">T</span><span class="o">*</span> <span class="n">data</span><span class="p">();</span>
<span class="cm">/**</span>
<span class="cm"> * @brief Return a pointer to mutable memory block.</span>
<span class="cm"> * @note If not exist, then allocation.</span>
<span class="cm"> */</span>
<span class="k">template</span> <span class="o">&lt;</span><span class="k">typename</span> <span class="n">T</span><span class="o">&gt;</span>
<span class="kr">inline</span> <span class="n">T</span><span class="o">*</span> <span class="n">mutable_data</span><span class="p">(</span><span class="n">platform</span><span class="o">::</span><span class="n">Place</span> <span class="n">place</span><span class="p">);</span>
<span class="cm">/**</span>
<span class="cm"> * @brief Return a pointer to mutable memory block.</span>
<span class="cm"> *</span>
<span class="cm"> * @param[in] dims The dimensions of the memory block.</span>
<span class="cm"> * @param[in] place The place of the memory block.</span>
<span class="cm"> *</span>
<span class="cm"> * @note If not exist, then allocation.</span>
<span class="cm"> */</span>
<span class="k">template</span> <span class="o">&lt;</span><span class="k">typename</span> <span class="n">T</span><span class="o">&gt;</span>
<span class="kr">inline</span> <span class="n">T</span><span class="o">*</span> <span class="n">mutable_data</span><span class="p">(</span><span class="n">DDim</span> <span class="n">dims</span><span class="p">,</span> <span class="n">platform</span><span class="o">::</span><span class="n">Place</span> <span class="n">place</span><span class="p">);</span>
<span class="cm">/*! Resize the dimensions of the memory block. */</span>
<span class="kr">inline</span> <span class="n">Tensor</span><span class="o">&amp;</span> <span class="n">Resize</span><span class="p">(</span><span class="k">const</span> <span class="n">DDim</span><span class="o">&amp;</span> <span class="n">dims</span><span class="p">);</span>
<span class="cm">/*! Return the dimensions of the memory block. */</span>
<span class="kr">inline</span> <span class="k">const</span> <span class="n">DDim</span><span class="o">&amp;</span> <span class="n">dims</span><span class="p">()</span> <span class="k">const</span><span class="p">;</span>
<span class="k">private</span><span class="o">:</span>
<span class="cm">/*! holds the memory block if allocated. */</span>
<span class="n">std</span><span class="o">::</span><span class="n">shared_ptr</span><span class="o">&lt;</span><span class="n">Placeholder</span><span class="o">&gt;</span> <span class="n">holder_</span><span class="p">;</span>
<span class="cm">/*! points to dimensions of memory block. */</span>
<span class="n">DDim</span> <span class="n">dim_</span><span class="p">;</span>
<span class="p">};</span>
</pre></div>
</div>
<p><code class="docutils literal"><span class="pre">Placeholder</span></code>的作用是延迟分配内存，即我们可以先定义一个Tensor，然后使用Resize接口设置Tensor的大小，最后再调用mutable_data接口分配实际的内存。</p>
<div class="highlight-cpp"><div class="highlight"><pre><span></span><span class="n">paddle</span><span class="o">::</span><span class="n">framework</span><span class="o">::</span><span class="n">Tensor</span> <span class="n">t</span><span class="p">;</span>
<span class="n">paddle</span><span class="o">::</span><span class="n">platform</span><span class="o">::</span><span class="n">CPUPlace</span> <span class="n">place</span><span class="p">;</span>
<span class="c1">// set size first</span>
<span class="n">t</span><span class="p">.</span><span class="n">Resize</span><span class="p">({</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">});</span>
<span class="c1">// allocate memory on CPU later</span>
<span class="n">t</span><span class="p">.</span><span class="n">mutable_data</span><span class="p">(</span><span class="n">place</span><span class="p">);</span>
</pre></div>
</div>
</div>
<div class="section" id="paddle-framework-tensor">
<span id="id1"></span><h2>paddle::framework::Tensor使用样例<a class="headerlink" href="#paddle-framework-tensor" title="永久链接至标题"></a></h2>
<p>下面以AddOp为例说明Tensor的使用过程：</p>
<ul class="simple">
<li>InferShape</li>
</ul>
<p>在运行神经网络计算图时，我们先调用每个<code class="docutils literal"><span class="pre">Operator</span></code>的<code class="docutils literal"><span class="pre">InferShape</span></code>接口，根据输入Tensor的大小来设置输出Tensor的大小，<code class="docutils literal"><span class="pre">Resize</span></code>接口会被调用。</p>
<div class="highlight-cpp"><div class="highlight"><pre><span></span><span class="kt">void</span> <span class="nf">InferShape</span><span class="p">(</span><span class="k">const</span> <span class="n">framework</span><span class="o">::</span><span class="n">InferShapeContext</span> <span class="o">&amp;</span><span class="n">ctx</span><span class="p">)</span> <span class="k">const</span> <span class="k">override</span> <span class="p">{</span>
<span class="n">PADDLE_ENFORCE_EQ</span><span class="p">(</span><span class="n">ctx</span><span class="p">.</span><span class="n">Input</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;X&quot;</span><span class="p">)</span><span class="o">-&gt;</span><span class="n">dims</span><span class="p">(),</span>
<span class="n">ctx</span><span class="p">.</span><span class="n">Input</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;Y&quot;</span><span class="p">)</span><span class="o">-&gt;</span><span class="n">dims</span><span class="p">(),</span>
<span class="s">&quot;Two input of Add Op&#39;s dimension must be same.&quot;</span><span class="p">);</span>
<span class="n">ctx</span><span class="p">.</span><span class="n">Output</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;Out&quot;</span><span class="p">)</span><span class="o">-&gt;</span><span class="n">Resize</span><span class="p">(</span><span class="n">ctx</span><span class="p">.</span><span class="n">Input</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;X&quot;</span><span class="p">)</span><span class="o">-&gt;</span><span class="n">dims</span><span class="p">());</span>
<span class="p">}</span>
</pre></div>
</div>
<ul class="simple">
<li>Run</li>
</ul>
<p><code class="docutils literal"><span class="pre">Operator</span></code>的<code class="docutils literal"><span class="pre">Run</span></code>接口最终会调用对应<code class="docutils literal"><span class="pre">OpKernel</span></code>的<code class="docutils literal"><span class="pre">Compute</span></code>接口，在这时真正的分配内存，<code class="docutils literal"><span class="pre">mutable_data</span></code>接口会被调用。</p>
<div class="highlight-cpp"><div class="highlight"><pre><span></span><span class="kt">void</span> <span class="nf">Compute</span><span class="p">(</span><span class="k">const</span> <span class="n">framework</span><span class="o">::</span><span class="n">ExecutionContext</span><span class="o">&amp;</span> <span class="n">context</span><span class="p">)</span> <span class="k">const</span> <span class="k">override</span> <span class="p">{</span>
<span class="k">auto</span><span class="o">*</span> <span class="n">input0</span> <span class="o">=</span> <span class="n">context</span><span class="p">.</span><span class="n">Input</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;X&quot;</span><span class="p">);</span>
<span class="k">auto</span><span class="o">*</span> <span class="n">input1</span> <span class="o">=</span> <span class="n">context</span><span class="p">.</span><span class="n">Input</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;Y&quot;</span><span class="p">);</span>
<span class="k">auto</span><span class="o">*</span> <span class="n">output</span> <span class="o">=</span> <span class="n">context</span><span class="p">.</span><span class="n">Output</span><span class="o">&lt;</span><span class="n">Tensor</span><span class="o">&gt;</span><span class="p">(</span><span class="s">&quot;Out&quot;</span><span class="p">);</span>
<span class="n">output</span><span class="o">-&gt;</span><span class="n">mutable_data</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;</span><span class="p">(</span><span class="n">context</span><span class="p">.</span><span class="n">GetPlace</span><span class="p">());</span>
<span class="k">auto</span> <span class="n">x</span> <span class="o">=</span> <span class="n">EigenVector</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;::</span><span class="n">Flatten</span><span class="p">(</span><span class="o">*</span><span class="n">input0</span><span class="p">);</span>
<span class="k">auto</span> <span class="n">y</span> <span class="o">=</span> <span class="n">EigenVector</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;::</span><span class="n">Flatten</span><span class="p">(</span><span class="o">*</span><span class="n">input1</span><span class="p">);</span>
<span class="k">auto</span> <span class="n">z</span> <span class="o">=</span> <span class="n">EigenVector</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;::</span><span class="n">Flatten</span><span class="p">(</span><span class="o">*</span><span class="n">output</span><span class="p">);</span>
<span class="k">auto</span> <span class="n">place</span> <span class="o">=</span> <span class="n">context</span><span class="p">.</span><span class="n">GetEigenDevice</span><span class="o">&lt;</span><span class="n">Place</span><span class="o">&gt;</span><span class="p">();</span>
<span class="n">z</span><span class="p">.</span><span class="n">device</span><span class="p">(</span><span class="n">place</span><span class="p">)</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="n">y</span><span class="p">;</span>
<span class="p">}</span>
</pre></div>
</div>
</div>
<div class="section" id="paddle-framework-tensoreigentensor">
<span id="paddle-framework-tensoreigentensor"></span><h2>paddle::framework::Tensor到EigenTensor的转换<a class="headerlink" href="#paddle-framework-tensoreigentensor" title="永久链接至标题"></a></h2>
<p>如上一小节所示，在具体的计算中，我们需要先把输入Tensor和输出Tensor转换为Eigen支持的格式。我们在<a class="reference external" href="https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/framework/eigen.h">eigen.h</a>中提供了一些全局函数用来实现paddle::framework::Tensor到EigenTensor/EigenMatrix/EigenVector/EigenScalar的转换。</p>
<p>以EigenTensor为例，做一个介绍</p>
<div class="highlight-cpp"><div class="highlight"><pre><span></span><span class="n">Tensor</span> <span class="n">t</span><span class="p">;</span>
<span class="kt">float</span><span class="o">*</span> <span class="n">p</span> <span class="o">=</span> <span class="n">t</span><span class="p">.</span><span class="n">mutable_data</span><span class="o">&lt;</span><span class="kt">float</span><span class="o">&gt;</span><span class="p">(</span><span class="n">make_ddim</span><span class="p">({</span><span class="mi">1</span><span class="p">,</span> <span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">}),</span> <span class="n">platform</span><span class="o">::</span><span class="n">CPUPlace</span><span class="p">());</span>
<span class="k">for</span> <span class="p">(</span><span class="kt">int</span> <span class="n">i</span> <span class="o">=</span> <span class="mi">0</span><span class="p">;</span> <span class="n">i</span> <span class="o">&lt;</span> <span class="mi">1</span> <span class="o">*</span> <span class="mi">2</span> <span class="o">*</span> <span class="mi">3</span><span class="p">;</span> <span class="n">i</span><span class="o">++</span><span class="p">)</span> <span class="p">{</span>
<span class="n">p</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="o">=</span> <span class="k">static_cast</span><span class="o">&lt;</span><span class="kt">float</span><span class="o">&gt;</span><span class="p">(</span><span class="n">i</span><span class="p">);</span>
<span class="p">}</span>
<span class="n">EigenTensor</span><span class="o">&lt;</span><span class="kt">float</span><span class="p">,</span> <span class="mi">3</span><span class="o">&gt;::</span><span class="n">Type</span> <span class="n">et</span> <span class="o">=</span> <span class="n">EigenTensor</span><span class="o">&lt;</span><span class="kt">float</span><span class="p">,</span> <span class="mi">3</span><span class="o">&gt;::</span><span class="n">From</span><span class="p">(</span><span class="n">t</span><span class="p">);</span>
</pre></div>
</div>
<p>From是EigenTensor模板提供的一个接口，可以实现从paddle::framework::Tensor到对EigenTensor的转换。由于Tensor的rank是模板参数，因此在转换时需要显示的指定。</p>
<p>在Eigen中，不同rank的Tensor是不同类型，Vector是rank为1的Tensor。需要额外注意的是，EigenVector<T>::From方法是把paddle中的一维Tensor转为Eigen的一维Tensor，在这里用EigenVector来表示；而EigenVector<T>::Flatten方法是把paddle中的一个Tensor进行reshape操作，压扁成为Eigen的一维Tensor，类型仍然为EigenVector。</p>
<p>更多的转换方法请参考eigen_test.cc中的<a class="reference external" href="https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/framework/eigen_test.cc">单元测试</a>。</p>
</div>
<div class="section" id="">
<span id="id2"></span><h2>实现计算<a class="headerlink" href="#" title="永久链接至标题"></a></h2>
<p>当需要完成计算时，我们需要等式左边的EigenTensor调用device接口。在这里需要注意的是，这里的EigenTensor之间的运算只是改变了原有Tensor中的数据，而不会改变原有Tensor的shape信息。</p>
<div class="highlight-cpp"><div class="highlight"><pre><span></span><span class="k">auto</span> <span class="n">x</span> <span class="o">=</span> <span class="n">EigenVector</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;::</span><span class="n">Flatten</span><span class="p">(</span><span class="o">*</span><span class="n">input0</span><span class="p">);</span>
<span class="k">auto</span> <span class="n">y</span> <span class="o">=</span> <span class="n">EigenVector</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;::</span><span class="n">Flatten</span><span class="p">(</span><span class="o">*</span><span class="n">input1</span><span class="p">);</span>
<span class="k">auto</span> <span class="n">z</span> <span class="o">=</span> <span class="n">EigenVector</span><span class="o">&lt;</span><span class="n">T</span><span class="o">&gt;::</span><span class="n">Flatten</span><span class="p">(</span><span class="o">*</span><span class="n">output</span><span class="p">);</span>
<span class="k">auto</span> <span class="n">place</span> <span class="o">=</span> <span class="n">context</span><span class="p">.</span><span class="n">GetEigenDevice</span><span class="o">&lt;</span><span class="n">Place</span><span class="o">&gt;</span><span class="p">();</span>
<span class="n">z</span><span class="p">.</span><span class="n">device</span><span class="p">(</span><span class="n">place</span><span class="p">)</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="n">y</span><span class="p">;</span>
</pre></div>
</div>
<p>在这段代码中，input0/input1/output可以是任意维度的Tensor。我们调用了EigenVector的Flatten接口，把任意维度的Tensor转为了一维的EigenVector。而在计算结束之后，input0/input1/output的原有shape信息不变。如果想改变原有Tensor的shape信息，可以调用Resize接口进行改变。</p>
<p>由于Eigen Tensor模块的文档较少，我们可以参考TensorFlow的<a class="reference external" href="https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/kernels">kernels</a>模块下的相关<code class="docutils literal"><span class="pre">OpKernel</span></code>的计算代码。</p>
</div>
</div>
</div>
</div>
<footer>
<hr/>
<div role="contentinfo">
<p>
&copy; Copyright 2016, PaddlePaddle developers.
</p>
</div>
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/snide/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
</footer>
</div>
</div>
</section>
</div>
<script type="text/javascript">
var DOCUMENTATION_OPTIONS = {
URL_ROOT:'../../',
VERSION:'',
COLLAPSE_INDEX:false,
FILE_SUFFIX:'.html',
HAS_SOURCE: true,
SOURCELINK_SUFFIX: ".txt",
};
</script>
<script type="text/javascript" src="../../_static/jquery.js"></script>
<script type="text/javascript" src="../../_static/underscore.js"></script>
<script type="text/javascript" src="../../_static/doctools.js"></script>
<script type="text/javascript" src="../../_static/translations.js"></script>
<script type="text/javascript" src="https://cdn.bootcss.com/mathjax/2.7.0/MathJax.js"></script>
<script type="text/javascript" src="../../_static/js/theme.js"></script>
<script src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.7/js/bootstrap.min.js" integrity="sha384-Tc5IQib027qvyjSMfHjOMaLkfuWVxZxUPnCJA7l2mCWNIpG9mGCD8wGNIcPD7Txa" crossorigin="anonymous"></script>
<script src="https://cdn.jsdelivr.net/perfect-scrollbar/0.6.14/js/perfect-scrollbar.jquery.min.js"></script>
<script src="../../_static/js/paddle_doc_init.js"></script>
</body>
</html>
\ No newline at end of file
develop/doc_cn/searchindex.js
浏览文件 @ d95d66cb
因为 它太大了无法显示 source diff 。你可以改为 查看blob。
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或