Skip to content

P
Paddle

机器未来 / Paddle
与 Fork 源项目一致

Fork自 PaddlePaddle / Paddle

通知 1
Star 1
Fork 0
P
Paddle
提交 21251d11 编写于 作者: T Travis CI
浏览文件
操作

Deploy to GitHub Pages: 4d422156

上级 5ec72b14
显示空白变更内容
内联 并排
Showing with 767 addition and 2 deletion
develop/doc/_sources/design/float16.md.txt 0 → 100644
浏览文件 @ 21251d11
# Design Doc: float16
## Why float16
Half precision (float16) is a binary floating-point format that occupies 16 bits in memory. float16 is half the size of traditional 32-bit single precision format (float) and has lower precision and smaller range.
When high precision computation is not required, using float16 data type could potentially
- reduce storage space, memory bandwidth, and power usages;
- increase the chance of data fitting into a smaller cache of lower latency;
- provide arithmetic speed up if supported by hardware.
## Survey of current float16 support
A brief survey of float16 support on different compilers, hardwares, and libraries can be found below. Interested readers can refer to [link1](https://github.com/PaddlePaddle/Paddle/issues/4853) and [link2](https://github.com/Xreki/Xreki.github.io/blob/master/multi_data_types_in_dl_framework/ppt/float16_and_quantized_type.md) for more info.
The goal of float16 is to serve as a key for the executor to find and run the correct version of compute method specialized for float16 in operator kernel. It should be compatible with various natively supported float16 implementations including `__half` for cuda, `float16_t` for ARM, and `Eigen::half` for Eigen to make writing customized float16 kernels easier.
### Compiler
- nvcc supports `__half` data type after CUDA 7.5.
- `__fp16` or `float16_t` is supported as storage type for gcc >= 6.1 and clang >= 3.4.
- `__fp16` or `float16_t` is supported as arithmetic type for gcc >= 7.1 and clang >= 3.9.
### Hardware
- `__half` is supported on GPU with compute capability >= 5.3.
- `__fp16` is supported as storage type for ARMv7-A, ARMv8-A, and above.
- `__fp16` is supported as arithmetic type after ARMv8.2-A (currently, the only microarchitecture implementing ARMv8.2-A is ARM Cortex-A75, which is announced in May 2017. There seems to be no application processors currently available on market that adopts this architecture. It is reported that Qualcomm Snapdragon 845 uses Cortex-A75 design and will be available in mobile devices in early 2018).
### Libraries
- [Eigen](https://github.com/RLovelett/eigen) >= 3.3 supports float16 calculation on both GPU and CPU using the `Eigen::half` class. It is mostly useful for Nvidia GPUs because of the overloaded arithmetic operators using cuda intrinsics. It falls back to using software emulation on CPU for calculation and there is no special treatment to ARM processors.
- [ARM compute library](https://github.com/ARM-software/ComputeLibrary) >= 17.02.01 supports NEON FP16 kernels (requires ARMv8.2-A CPU).
## Implementation
The float16 class holds a 16-bit `uint16_t` data internally.
```
struct float16 {
uint16_t x;
};
```
float16 supports the following features:
- constructors / assignment operators that take input from primitive data types including bool, integers of various length, float, and double.
- constructors / assignment operators that take input from `__half` on cuda, `float16_t` on ARM, and `Eigen::half` on Eigen.
- conversion operators to primitive data types and half precision data types on cuda, ARM and Eigen.
- overloaded arithmetic operators for cuda, arm, and non-arm cpu, respectively. These operators will take advantage of the cuda and ARM intrinsics on the corresponding hardware.
To support the above features, two fundamental conversion functions are provided:
```
float16 float_to_half_rn(float f); // convert to half precision in round-to-nearest-even mode
float half_to_float(float16 h);
```
which provides one-to-one conversion between float32 and float16. These twos functions will do different conversion routines based on the current hardware. CUDA/ARM instrinsics will be used when the corresonding hardware is available. If the hardware or compiler level does not support float32 to float16 conversion, software emulation will be performed to do the conversion.
## To do
After float16 class is available, some of the future items are below:
- Update pybind/tensor_py.h to bind c++ float16 with numpy float16.
- Modify `IndicateDataType()` method in `framework/operator.h` to make it compatible with float16.
- Create a type-casting operator that can convert the data type in tensor between float16 and other types.
develop/doc/design/float16.html 0 → 100644
浏览文件 @ 21251d11
<!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>Design Doc: float16 &mdash; PaddlePaddle documentation</title>
<link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
<link rel="index" title="Index"
href="../genindex.html"/>
<link rel="search" title="Search" href="../search.html"/>
<link rel="top" title="PaddlePaddle documentation" 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_en.html">GET STARTED</a></li>
<li class="toctree-l1"><a class="reference internal" href="../howto/index_en.html">HOW TO</a></li>
<li class="toctree-l1"><a class="reference internal" href="../api/index_en.html">API</a></li>
<li class="toctree-l1"><a class="reference internal" href="../mobile/index_en.html">MOBILE</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_en.html">GET STARTED</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../getstarted/build_and_install/index_en.html">Install and Build</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../getstarted/build_and_install/docker_install_en.html">PaddlePaddle in Docker Containers</a></li>
<li class="toctree-l3"><a class="reference internal" href="../getstarted/build_and_install/build_from_source_en.html">Installing from Sources</a></li>
</ul>
</li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../howto/index_en.html">HOW TO</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../howto/usage/cmd_parameter/index_en.html">Set Command-line Parameters</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../howto/usage/cmd_parameter/use_case_en.html">Use Case</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/usage/cmd_parameter/arguments_en.html">Argument Outline</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/usage/cmd_parameter/detail_introduction_en.html">Detail Description</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../howto/usage/cluster/cluster_train_en.html">PaddlePaddle Distributed Training</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/usage/k8s/k8s_en.html">Paddle On Kubernetes</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/usage/k8s/k8s_aws_en.html">Distributed PaddlePaddle Training on AWS with Kubernetes</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/dev/build_en.html">Build PaddlePaddle from Source Code and Run Unit Test</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/dev/new_layer_en.html">Write New Layers</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/dev/contribute_to_paddle_en.html">Contribute Code</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/deep_model/rnn/index_en.html">RNN Models</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../howto/deep_model/rnn/rnn_config_en.html">RNN Configuration</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../howto/optimization/gpu_profiling_en.html">Tune GPU Performance</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../api/index_en.html">API</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../api/v2/model_configs.html">Model Configuration</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">Data Reader Interface and DataSets</a></li>
<li class="toctree-l2"><a class="reference internal" href="../api/v2/run_logic.html">Training and Inference</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../mobile/index_en.html">MOBILE</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../mobile/cross_compiling_for_android_en.html">Build PaddlePaddle for Android</a></li>
<li class="toctree-l2"><a class="reference internal" href="../mobile/cross_compiling_for_raspberry_en.html">Build PaddlePaddle for Raspberry Pi</a></li>
</ul>
</li>
</ul>
</nav>
<section class="doc-content-wrap">
<div role="navigation" aria-label="breadcrumbs navigation">
<ul class="wy-breadcrumbs">
<li>Design Doc: float16</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="design-doc-float16">
<span id="design-doc-float16"></span><h1>Design Doc: float16<a class="headerlink" href="#design-doc-float16" title="Permalink to this headline"></a></h1>
<div class="section" id="why-float16">
<span id="why-float16"></span><h2>Why float16<a class="headerlink" href="#why-float16" title="Permalink to this headline"></a></h2>
<p>Half precision (float16) is a binary floating-point format that occupies 16 bits in memory. float16 is half the size of traditional 32-bit single precision format (float) and has lower precision and smaller range.</p>
<p>When high precision computation is not required, using float16 data type could potentially</p>
<ul class="simple">
<li>reduce storage space, memory bandwidth, and power usages;</li>
<li>increase the chance of data fitting into a smaller cache of lower latency;</li>
<li>provide arithmetic speed up if supported by hardware.</li>
</ul>
</div>
<div class="section" id="survey-of-current-float16-support">
<span id="survey-of-current-float16-support"></span><h2>Survey of current float16 support<a class="headerlink" href="#survey-of-current-float16-support" title="Permalink to this headline"></a></h2>
<p>A brief survey of float16 support on different compilers, hardwares, and libraries can be found below. Interested readers can refer to <a class="reference external" href="https://github.com/PaddlePaddle/Paddle/issues/4853">link1</a> and <a class="reference external" href="https://github.com/Xreki/Xreki.github.io/blob/master/multi_data_types_in_dl_framework/ppt/float16_and_quantized_type.md">link2</a> for more info.</p>
<p>The goal of float16 is to serve as a key for the executor to find and run the correct version of compute method specialized for float16 in operator kernel. It should be compatible with various natively supported float16 implementations including <code class="docutils literal"><span class="pre">__half</span></code> for cuda, <code class="docutils literal"><span class="pre">float16_t</span></code> for ARM, and <code class="docutils literal"><span class="pre">Eigen::half</span></code> for Eigen to make writing customized float16 kernels easier.</p>
<div class="section" id="compiler">
<span id="compiler"></span><h3>Compiler<a class="headerlink" href="#compiler" title="Permalink to this headline"></a></h3>
<ul class="simple">
<li>nvcc supports <code class="docutils literal"><span class="pre">__half</span></code> data type after CUDA 7.5.</li>
<li><code class="docutils literal"><span class="pre">__fp16</span></code> or <code class="docutils literal"><span class="pre">float16_t</span></code> is supported as storage type for gcc &gt;= 6.1 and clang &gt;= 3.4.</li>
<li><code class="docutils literal"><span class="pre">__fp16</span></code> or <code class="docutils literal"><span class="pre">float16_t</span></code> is supported as arithmetic type for gcc &gt;= 7.1 and clang &gt;= 3.9.</li>
</ul>
</div>
<div class="section" id="hardware">
<span id="hardware"></span><h3>Hardware<a class="headerlink" href="#hardware" title="Permalink to this headline"></a></h3>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">__half</span></code> is supported on GPU with compute capability &gt;= 5.3.</li>
<li><code class="docutils literal"><span class="pre">__fp16</span></code> is supported as storage type for ARMv7-A, ARMv8-A, and above.</li>
<li><code class="docutils literal"><span class="pre">__fp16</span></code> is supported as arithmetic type after ARMv8.2-A (currently, the only microarchitecture implementing ARMv8.2-A is ARM Cortex-A75, which is announced in May 2017. There seems to be no application processors currently available on market that adopts this architecture. It is reported that Qualcomm Snapdragon 845 uses Cortex-A75 design and will be available in mobile devices in early 2018).</li>
</ul>
</div>
<div class="section" id="libraries">
<span id="libraries"></span><h3>Libraries<a class="headerlink" href="#libraries" title="Permalink to this headline"></a></h3>
<ul class="simple">
<li><a class="reference external" href="https://github.com/RLovelett/eigen">Eigen</a> &gt;= 3.3 supports float16 calculation on both GPU and CPU using the <code class="docutils literal"><span class="pre">Eigen::half</span></code> class. It is mostly useful for Nvidia GPUs because of the overloaded arithmetic operators using cuda intrinsics. It falls back to using software emulation on CPU for calculation and there is no special treatment to ARM processors.</li>
<li><a class="reference external" href="https://github.com/ARM-software/ComputeLibrary">ARM compute library</a> &gt;= 17.02.01 supports NEON FP16 kernels (requires ARMv8.2-A CPU).</li>
</ul>
</div>
</div>
<div class="section" id="implementation">
<span id="implementation"></span><h2>Implementation<a class="headerlink" href="#implementation" title="Permalink to this headline"></a></h2>
<p>The float16 class holds a 16-bit <code class="docutils literal"><span class="pre">uint16_t</span></code> data internally.</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">struct</span> <span class="n">float16</span> <span class="p">{</span>
<span class="n">uint16_t</span> <span class="n">x</span><span class="p">;</span>
<span class="p">};</span>
</pre></div>
</div>
<p>float16 supports the following features:</p>
<ul class="simple">
<li>constructors / assignment operators that take input from primitive data types including bool, integers of various length, float, and double.</li>
<li>constructors / assignment operators that take input from <code class="docutils literal"><span class="pre">__half</span></code> on cuda, <code class="docutils literal"><span class="pre">float16_t</span></code> on ARM, and <code class="docutils literal"><span class="pre">Eigen::half</span></code> on Eigen.</li>
<li>conversion operators to primitive data types and half precision data types on cuda, ARM and Eigen.</li>
<li>overloaded arithmetic operators for cuda, arm, and non-arm cpu, respectively. These operators will take advantage of the cuda and ARM intrinsics on the corresponding hardware.</li>
</ul>
<p>To support the above features, two fundamental conversion functions are provided:</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">float16</span> <span class="n">float_to_half_rn</span><span class="p">(</span><span class="nb">float</span> <span class="n">f</span><span class="p">);</span> <span class="o">//</span> <span class="n">convert</span> <span class="n">to</span> <span class="n">half</span> <span class="n">precision</span> <span class="ow">in</span> <span class="nb">round</span><span class="o">-</span><span class="n">to</span><span class="o">-</span><span class="n">nearest</span><span class="o">-</span><span class="n">even</span> <span class="n">mode</span>
<span class="nb">float</span> <span class="n">half_to_float</span><span class="p">(</span><span class="n">float16</span> <span class="n">h</span><span class="p">);</span>
</pre></div>
</div>
<p>which provides one-to-one conversion between float32 and float16. These twos functions will do different conversion routines based on the current hardware. CUDA/ARM instrinsics will be used when the corresonding hardware is available. If the hardware or compiler level does not support float32 to float16 conversion, software emulation will be performed to do the conversion.</p>
</div>
<div class="section" id="to-do">
<span id="to-do"></span><h2>To do<a class="headerlink" href="#to-do" title="Permalink to this headline"></a></h2>
<p>After float16 class is available, some of the future items are below:</p>
<ul class="simple">
<li>Update pybind/tensor_py.h to bind c++ float16 with numpy float16.</li>
<li>Modify <code class="docutils literal"><span class="pre">IndicateDataType()</span></code> method in <code class="docutils literal"><span class="pre">framework/operator.h</span></code> to make it compatible with float16.</li>
<li>Create a type-casting operator that can convert the data type in tensor between float16 and other types.</li>
</ul>
</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="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></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/searchindex.js
浏览文件 @ 21251d11
因为 它太大了无法显示 source diff 。你可以改为 查看blob。
develop/doc_cn/_sources/design/float16.md.txt 0 → 100644
浏览文件 @ 21251d11
# Design Doc: float16
## Why float16
Half precision (float16) is a binary floating-point format that occupies 16 bits in memory. float16 is half the size of traditional 32-bit single precision format (float) and has lower precision and smaller range.
When high precision computation is not required, using float16 data type could potentially
- reduce storage space, memory bandwidth, and power usages;
- increase the chance of data fitting into a smaller cache of lower latency;
- provide arithmetic speed up if supported by hardware.
## Survey of current float16 support
A brief survey of float16 support on different compilers, hardwares, and libraries can be found below. Interested readers can refer to [link1](https://github.com/PaddlePaddle/Paddle/issues/4853) and [link2](https://github.com/Xreki/Xreki.github.io/blob/master/multi_data_types_in_dl_framework/ppt/float16_and_quantized_type.md) for more info.
The goal of float16 is to serve as a key for the executor to find and run the correct version of compute method specialized for float16 in operator kernel. It should be compatible with various natively supported float16 implementations including `__half` for cuda, `float16_t` for ARM, and `Eigen::half` for Eigen to make writing customized float16 kernels easier.
### Compiler
- nvcc supports `__half` data type after CUDA 7.5.
- `__fp16` or `float16_t` is supported as storage type for gcc >= 6.1 and clang >= 3.4.
- `__fp16` or `float16_t` is supported as arithmetic type for gcc >= 7.1 and clang >= 3.9.
### Hardware
- `__half` is supported on GPU with compute capability >= 5.3.
- `__fp16` is supported as storage type for ARMv7-A, ARMv8-A, and above.
- `__fp16` is supported as arithmetic type after ARMv8.2-A (currently, the only microarchitecture implementing ARMv8.2-A is ARM Cortex-A75, which is announced in May 2017. There seems to be no application processors currently available on market that adopts this architecture. It is reported that Qualcomm Snapdragon 845 uses Cortex-A75 design and will be available in mobile devices in early 2018).
### Libraries
- [Eigen](https://github.com/RLovelett/eigen) >= 3.3 supports float16 calculation on both GPU and CPU using the `Eigen::half` class. It is mostly useful for Nvidia GPUs because of the overloaded arithmetic operators using cuda intrinsics. It falls back to using software emulation on CPU for calculation and there is no special treatment to ARM processors.
- [ARM compute library](https://github.com/ARM-software/ComputeLibrary) >= 17.02.01 supports NEON FP16 kernels (requires ARMv8.2-A CPU).
## Implementation
The float16 class holds a 16-bit `uint16_t` data internally.
```
struct float16 {
uint16_t x;
};
```
float16 supports the following features:
- constructors / assignment operators that take input from primitive data types including bool, integers of various length, float, and double.
- constructors / assignment operators that take input from `__half` on cuda, `float16_t` on ARM, and `Eigen::half` on Eigen.
- conversion operators to primitive data types and half precision data types on cuda, ARM and Eigen.
- overloaded arithmetic operators for cuda, arm, and non-arm cpu, respectively. These operators will take advantage of the cuda and ARM intrinsics on the corresponding hardware.
To support the above features, two fundamental conversion functions are provided:
```
float16 float_to_half_rn(float f); // convert to half precision in round-to-nearest-even mode
float half_to_float(float16 h);
```
which provides one-to-one conversion between float32 and float16. These twos functions will do different conversion routines based on the current hardware. CUDA/ARM instrinsics will be used when the corresonding hardware is available. If the hardware or compiler level does not support float32 to float16 conversion, software emulation will be performed to do the conversion.
## To do
After float16 class is available, some of the future items are below:
- Update pybind/tensor_py.h to bind c++ float16 with numpy float16.
- Modify `IndicateDataType()` method in `framework/operator.h` to make it compatible with float16.
- Create a type-casting operator that can convert the data type in tensor between float16 and other types.
develop/doc_cn/design/float16.html 0 → 100644
浏览文件 @ 21251d11
<!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>Design Doc: float16 &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="../howto/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>
<li class="toctree-l1"><a class="reference internal" href="../mobile/index_cn.html">MOBILE</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="../howto/index_cn.html">进阶指南</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../howto/usage/cmd_parameter/index_cn.html">设置命令行参数</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../howto/usage/cmd_parameter/use_case_cn.html">使用案例</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/usage/cmd_parameter/arguments_cn.html">参数概述</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/usage/cmd_parameter/detail_introduction_cn.html">细节描述</a></li>
</ul>
</li>
<li class="toctree-l2"><a class="reference internal" href="../howto/usage/cluster/cluster_train_cn.html">PaddlePaddle分布式训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/usage/k8s/k8s_basis_cn.html">Kubernetes 简介</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/usage/k8s/k8s_cn.html">Kubernetes单机训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/usage/k8s/k8s_distributed_cn.html">Kubernetes分布式训练</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/dev/build_cn.html">编译PaddlePaddle和运行单元测试</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/dev/write_docs_cn.html">如何贡献/修改文档</a></li>
<li class="toctree-l2"><a class="reference internal" href="../howto/deep_model/rnn/index_cn.html">RNN相关模型</a><ul>
<li class="toctree-l3"><a class="reference internal" href="../howto/deep_model/rnn/rnn_config_cn.html">RNN配置</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/deep_model/rnn/recurrent_group_cn.html">Recurrent Group教程</a></li>
<li class="toctree-l3"><a class="reference internal" href="../howto/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="../howto/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="../howto/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><ul>
<li class="toctree-l2"><a class="reference internal" href="../faq/build_and_install/index_cn.html">编译安装与单元测试</a></li>
<li class="toctree-l2"><a class="reference internal" href="../faq/model/index_cn.html">模型配置</a></li>
<li class="toctree-l2"><a class="reference internal" href="../faq/parameter/index_cn.html">参数设置</a></li>
<li class="toctree-l2"><a class="reference internal" href="../faq/local/index_cn.html">本地训练与预测</a></li>
<li class="toctree-l2"><a class="reference internal" href="../faq/cluster/index_cn.html">集群训练与预测</a></li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../mobile/index_cn.html">MOBILE</a><ul>
<li class="toctree-l2"><a class="reference internal" href="../mobile/cross_compiling_for_android_cn.html">构建Android平台上的PaddlePaddle库</a></li>
<li class="toctree-l2"><a class="reference internal" href="../mobile/cross_compiling_for_ios_cn.html">构建iOS平台上的PaddlePaddle库</a></li>
<li class="toctree-l2"><a class="reference internal" href="../mobile/cross_compiling_for_raspberry_cn.html">构建Raspberry Pi平台上的PaddlePaddle库</a></li>
</ul>
</li>
</ul>
</nav>
<section class="doc-content-wrap">
<div role="navigation" aria-label="breadcrumbs navigation">
<ul class="wy-breadcrumbs">
<li>Design Doc: float16</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="design-doc-float16">
<span id="design-doc-float16"></span><h1>Design Doc: float16<a class="headerlink" href="#design-doc-float16" title="永久链接至标题"></a></h1>
<div class="section" id="why-float16">
<span id="why-float16"></span><h2>Why float16<a class="headerlink" href="#why-float16" title="永久链接至标题"></a></h2>
<p>Half precision (float16) is a binary floating-point format that occupies 16 bits in memory. float16 is half the size of traditional 32-bit single precision format (float) and has lower precision and smaller range.</p>
<p>When high precision computation is not required, using float16 data type could potentially</p>
<ul class="simple">
<li>reduce storage space, memory bandwidth, and power usages;</li>
<li>increase the chance of data fitting into a smaller cache of lower latency;</li>
<li>provide arithmetic speed up if supported by hardware.</li>
</ul>
</div>
<div class="section" id="survey-of-current-float16-support">
<span id="survey-of-current-float16-support"></span><h2>Survey of current float16 support<a class="headerlink" href="#survey-of-current-float16-support" title="永久链接至标题"></a></h2>
<p>A brief survey of float16 support on different compilers, hardwares, and libraries can be found below. Interested readers can refer to <a class="reference external" href="https://github.com/PaddlePaddle/Paddle/issues/4853">link1</a> and <a class="reference external" href="https://github.com/Xreki/Xreki.github.io/blob/master/multi_data_types_in_dl_framework/ppt/float16_and_quantized_type.md">link2</a> for more info.</p>
<p>The goal of float16 is to serve as a key for the executor to find and run the correct version of compute method specialized for float16 in operator kernel. It should be compatible with various natively supported float16 implementations including <code class="docutils literal"><span class="pre">__half</span></code> for cuda, <code class="docutils literal"><span class="pre">float16_t</span></code> for ARM, and <code class="docutils literal"><span class="pre">Eigen::half</span></code> for Eigen to make writing customized float16 kernels easier.</p>
<div class="section" id="compiler">
<span id="compiler"></span><h3>Compiler<a class="headerlink" href="#compiler" title="永久链接至标题"></a></h3>
<ul class="simple">
<li>nvcc supports <code class="docutils literal"><span class="pre">__half</span></code> data type after CUDA 7.5.</li>
<li><code class="docutils literal"><span class="pre">__fp16</span></code> or <code class="docutils literal"><span class="pre">float16_t</span></code> is supported as storage type for gcc &gt;= 6.1 and clang &gt;= 3.4.</li>
<li><code class="docutils literal"><span class="pre">__fp16</span></code> or <code class="docutils literal"><span class="pre">float16_t</span></code> is supported as arithmetic type for gcc &gt;= 7.1 and clang &gt;= 3.9.</li>
</ul>
</div>
<div class="section" id="hardware">
<span id="hardware"></span><h3>Hardware<a class="headerlink" href="#hardware" title="永久链接至标题"></a></h3>
<ul class="simple">
<li><code class="docutils literal"><span class="pre">__half</span></code> is supported on GPU with compute capability &gt;= 5.3.</li>
<li><code class="docutils literal"><span class="pre">__fp16</span></code> is supported as storage type for ARMv7-A, ARMv8-A, and above.</li>
<li><code class="docutils literal"><span class="pre">__fp16</span></code> is supported as arithmetic type after ARMv8.2-A (currently, the only microarchitecture implementing ARMv8.2-A is ARM Cortex-A75, which is announced in May 2017. There seems to be no application processors currently available on market that adopts this architecture. It is reported that Qualcomm Snapdragon 845 uses Cortex-A75 design and will be available in mobile devices in early 2018).</li>
</ul>
</div>
<div class="section" id="libraries">
<span id="libraries"></span><h3>Libraries<a class="headerlink" href="#libraries" title="永久链接至标题"></a></h3>
<ul class="simple">
<li><a class="reference external" href="https://github.com/RLovelett/eigen">Eigen</a> &gt;= 3.3 supports float16 calculation on both GPU and CPU using the <code class="docutils literal"><span class="pre">Eigen::half</span></code> class. It is mostly useful for Nvidia GPUs because of the overloaded arithmetic operators using cuda intrinsics. It falls back to using software emulation on CPU for calculation and there is no special treatment to ARM processors.</li>
<li><a class="reference external" href="https://github.com/ARM-software/ComputeLibrary">ARM compute library</a> &gt;= 17.02.01 supports NEON FP16 kernels (requires ARMv8.2-A CPU).</li>
</ul>
</div>
</div>
<div class="section" id="implementation">
<span id="implementation"></span><h2>Implementation<a class="headerlink" href="#implementation" title="永久链接至标题"></a></h2>
<p>The float16 class holds a 16-bit <code class="docutils literal"><span class="pre">uint16_t</span></code> data internally.</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">struct</span> <span class="n">float16</span> <span class="p">{</span>
<span class="n">uint16_t</span> <span class="n">x</span><span class="p">;</span>
<span class="p">};</span>
</pre></div>
</div>
<p>float16 supports the following features:</p>
<ul class="simple">
<li>constructors / assignment operators that take input from primitive data types including bool, integers of various length, float, and double.</li>
<li>constructors / assignment operators that take input from <code class="docutils literal"><span class="pre">__half</span></code> on cuda, <code class="docutils literal"><span class="pre">float16_t</span></code> on ARM, and <code class="docutils literal"><span class="pre">Eigen::half</span></code> on Eigen.</li>
<li>conversion operators to primitive data types and half precision data types on cuda, ARM and Eigen.</li>
<li>overloaded arithmetic operators for cuda, arm, and non-arm cpu, respectively. These operators will take advantage of the cuda and ARM intrinsics on the corresponding hardware.</li>
</ul>
<p>To support the above features, two fundamental conversion functions are provided:</p>
<div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">float16</span> <span class="n">float_to_half_rn</span><span class="p">(</span><span class="nb">float</span> <span class="n">f</span><span class="p">);</span> <span class="o">//</span> <span class="n">convert</span> <span class="n">to</span> <span class="n">half</span> <span class="n">precision</span> <span class="ow">in</span> <span class="nb">round</span><span class="o">-</span><span class="n">to</span><span class="o">-</span><span class="n">nearest</span><span class="o">-</span><span class="n">even</span> <span class="n">mode</span>
<span class="nb">float</span> <span class="n">half_to_float</span><span class="p">(</span><span class="n">float16</span> <span class="n">h</span><span class="p">);</span>
</pre></div>
</div>
<p>which provides one-to-one conversion between float32 and float16. These twos functions will do different conversion routines based on the current hardware. CUDA/ARM instrinsics will be used when the corresonding hardware is available. If the hardware or compiler level does not support float32 to float16 conversion, software emulation will be performed to do the conversion.</p>
</div>
<div class="section" id="to-do">
<span id="to-do"></span><h2>To do<a class="headerlink" href="#to-do" title="永久链接至标题"></a></h2>
<p>After float16 class is available, some of the future items are below:</p>
<ul class="simple">
<li>Update pybind/tensor_py.h to bind c++ float16 with numpy float16.</li>
<li>Modify <code class="docutils literal"><span class="pre">IndicateDataType()</span></code> method in <code class="docutils literal"><span class="pre">framework/operator.h</span></code> to make it compatible with float16.</li>
<li>Create a type-casting operator that can convert the data type in tensor between float16 and other types.</li>
</ul>
</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 @ 21251d11
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