Skip to content

P
PaddleSlim

PaddlePaddle / PaddleSlim
大约 1 年 前同步成功

通知 51
Star 1434
Fork 344
P
PaddleSlim
切换分支/标签
查找文件 普通视图历史永久链接Permalink
index.html 32.6 KB
Newer Older
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 
<!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 http-equiv="X-UA-Compatible" content="IE=edge">
  <meta name="viewport" content="width=device-width, initial-scale=1.0">
  
  
  <link rel="shortcut icon" href="../../img/favicon.ico">
  <title>Embedding量化 - PaddleSlim Docs</title>
  <link href='https://fonts.googleapis.com/css?family=Lato:400,700|Roboto+Slab:400,700|Inconsolata:400,700' rel='stylesheet' type='text/css'>

  <link rel="stylesheet" href="../../css/theme.css" type="text/css" />
  <link rel="stylesheet" href="../../css/theme_extra.css" type="text/css" />
  <link rel="stylesheet" href="//cdnjs.cloudflare.com/ajax/libs/highlight.js/9.12.0/styles/github.min.css">
  
  <script>
    // Current page data
    var mkdocs_page_name = "Embedding\u91cf\u5316";
    var mkdocs_page_input_path = "tutorials/quant_embedding_demo.md";
    var mkdocs_page_url = null;
  </script>
  
  <script src="../../js/jquery-2.1.1.min.js" defer></script>
  <script src="../../js/modernizr-2.8.3.min.js" defer></script>
  <script src="//cdnjs.cloudflare.com/ajax/libs/highlight.js/9.12.0/highlight.min.js"></script>
  <script>hljs.initHighlightingOnLoad();</script> 
  
</head>

<body class="wy-body-for-nav" role="document">

  <div class="wy-grid-for-nav">

    
    <nav data-toggle="wy-nav-shift" class="wy-nav-side stickynav">
      <div class="wy-side-nav-search">
        <a href="../.." class="icon icon-home"> PaddleSlim Docs</a>
        <div role="search">
  <form id ="rtd-search-form" class="wy-form" action="../../search.html" method="get">
    <input type="text" name="q" placeholder="Search docs" title="Type search term here" />
  </form>
</div>
      </div>

      <div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
	<ul class="current">
	  
          
            <li class="toctree-l1">
		
    <a class="" href="../..">Home</a>
	    </li>
          
            <li class="toctree-l1">
B
Deployed 9bcff00 with MkDocs version: 1.0.4  
baiyfbupt 已提交
58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 
    <span class="caption-text">API</span>
    <ul class="subnav">
                <li class="">
                    
    <a class="" href="../../api/quantization_api/">量化</a>
                </li>
                <li class="">
                    
    <a class="" href="../../api/prune_api/">剪枝与敏感度</a>
                </li>
                <li class="">
                    
    <a class="" href="../../api/analysis_api/">模型分析</a>
                </li>
                <li class="">
                    
    <a class="" href="../../api/single_distiller_api/">知识蒸馏</a>
                </li>
                <li class="">
                    
    <a class="" href="../../api/nas_api/">SA搜索</a>
                </li>
                <li class="">
                    
    <a class="" href="../../api/search_space/">搜索空间</a>
                </li>
                <li class="">
                    
    <a class="" href="../../table_latency/">硬件延时评估表</a>
                </li>
    </ul>
	    </li>
          
            <li class="toctree-l1">
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 
    <span class="caption-text">教程</span>
    <ul class="subnav">
                <li class="">
                    
    <a class="" href="../quant_post_demo/">离线量化</a>
                </li>
                <li class="">
                    
    <a class="" href="../quant_aware_demo/">量化训练</a>
                </li>
                <li class=" current">
                    
    <a class="current" href="./">Embedding量化</a>
    <ul class="subnav">
            
    <li class="toctree-l3"><a href="#embedding">Embedding量化示例</a></li>
    
        <ul>
        
            <li><a class="toctree-l4" href="#skip-gramword2vector">基于skip-gram的word2vector模型</a></li>
        
            <li><a class="toctree-l4" href="#skip-gramword2vector_1">量化基于skip-gram的word2vector模型</a></li>
        
        </ul>
    

    </ul>
                </li>
                <li class="">
                    
    <a class="" href="../nas_demo/">SA搜索</a>
                </li>
B
Deployed 0f9c3ce with MkDocs version: 1.0.4  
baiyfbupt 已提交
125 126 127 128 
                <li class="">
                    
    <a class="" href="../distillation_demo/">知识蒸馏</a>
                </li>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
129 130 131 132 133 
    </ul>
	    </li>
          
            <li class="toctree-l1">
B
Deployed 9bcff00 with MkDocs version: 1.0.4  
baiyfbupt 已提交
134 
    <a class="" href="../../algo/algo/">算法原理</a>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
135 136 
	    </li>
B
Deployed 0f9c3ce with MkDocs version: 1.0.4  
baiyfbupt 已提交
137 138 
            <li class="toctree-l1">
B
Deployed 9bcff00 with MkDocs version: 1.0.4  
baiyfbupt 已提交
139 
    <a class="" href="../../model_zoo/">模型库</a>
B
Deployed 0f9c3ce with MkDocs version: 1.0.4  
baiyfbupt 已提交
140 141 
	    </li>
B
Deployed 9bcff00 with MkDocs version: 1.0.4  
baiyfbupt 已提交
142 143 144 145 146 
            <li class="toctree-l1">
		
    <a class="" href="../../model_zoo2/">模型库2</a>
	    </li>
B
Deployed 9bcff00 with MkDocs version: 1.0.4  
baiyfbupt 已提交
147 148 149 150 151 
            <li class="toctree-l1">
		
    <a class="" href="../../model_zoo3/">模型库3</a>
	    </li>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 
        </ul>
      </div>
      &nbsp;
    </nav>

    <section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">

      
      <nav class="wy-nav-top" role="navigation" aria-label="top navigation">
        <i data-toggle="wy-nav-top" class="fa fa-bars"></i>
        <a href="../..">PaddleSlim Docs</a>
      </nav>

      
      <div class="wy-nav-content">
        <div class="rst-content">
          <div role="navigation" aria-label="breadcrumbs navigation">
  <ul class="wy-breadcrumbs">
    <li><a href="../..">Docs</a> &raquo;</li>
    
      
        
          <li>教程 &raquo;</li>
        
      
    
    <li>Embedding量化</li>
    <li class="wy-breadcrumbs-aside">
      
        <a href="https://github.com/PaddlePaddle/PaddleSlim/edit/master/docs/tutorials/quant_embedding_demo.md"
          class="icon icon-github"> Edit on GitHub</a>
      
    </li>
  </ul>
  <hr/>
</div>
          <div role="main">
            <div class="section">
              
                <h1 id="embedding">Embedding量化示例<a class="headerlink" href="#embedding" title="Permanent link">#</a></h1>
<p>本示例介绍如何使用Embedding量化的接口 <a href="">paddleslim.quant.quant_embedding</a> 。<code>quant_embedding</code>接口将网络中的Embedding参数从<code>float32</code>类型量化到 <code>8-bit</code>整数类型，在几乎不损失模型精度的情况下减少模型的存储空间和显存占用。</p>
<p>接口介绍请参考 <a href='../../../paddleslim/quant/quantization_api_doc.md'>量化API文档</a>。</p>
<p>该接口对program的修改：</p>
<p>量化前:</p>
<p align="center">
<img src="./image/before.png" height=200 width=100 hspace='10'/> <br />
<strong>图1：量化前的模型结构</strong>
</p>

<p>量化后：</p>
<p align="center">
<img src="./image/after.png" height=300 width=300 hspace='10'/> <br />
<strong>图2: 量化后的模型结构</strong>
</p>

<p>以下将以 <code>基于skip-gram的word2vector模型</code> 为例来说明如何使用<code>quant_embedding</code>接口。首先介绍 <code>基于skip-gram的word2vector模型</code> 的正常训练和测试流程。</p>
<h2 id="skip-gramword2vector">基于skip-gram的word2vector模型<a class="headerlink" href="#skip-gramword2vector" title="Permanent link">#</a></h2>
<p>以下是本例的简要目录结构及说明：</p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
210 211 212 213 214 215 216 217 218 219 
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span> 1
 2
 3
 4
 5
 6
 7
 8
 9
10</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>.
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
220 221 222 223 224 225 226 227 228 229 
├── cluster_train.py    # 分布式训练函数
├── cluster_train.sh    # 本地模拟多机脚本
├── train.py            # 训练函数
├── infer.py            # 预测脚本
├── net.py              # 网络结构
├── preprocess.py       # 预处理脚本，包括构建词典和预处理文本
├── reader.py           # 训练阶段的文本读写
├── train.py            # 训练函数
└── utils.py            # 通用函数
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
230 
</td></tr></table>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
231 232 233 234 235 

<h3 id="_1">介绍<a class="headerlink" href="#_1" title="Permanent link">#</a></h3>
<p>本例实现了skip-gram模式的word2vector模型。</p>
<p>同时推荐用户参考<a href="https://aistudio.baidu.com/aistudio/projectDetail/124377"> IPython Notebook demo</a></p>
<h3 id="_2">数据下载<a class="headerlink" href="#_2" title="Permanent link">#</a></h3>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
236 237 238 239 240 
<p>全量数据集使用的是来自1 Billion Word Language Model Benchmark的(<a href="http://www.statmt.org/lm-benchmark">http://www.statmt.org/lm-benchmark</a>) 的数据集.</p>
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1
2
3
4</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>mkdir data
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
241 242 243 244 
wget http://www.statmt.org/lm-benchmark/1-billion-word-language-modeling-benchmark-r13output.tar.gz
tar xzvf <span class="m">1</span>-billion-word-language-modeling-benchmark-r13output.tar.gz
mv <span class="m">1</span>-billion-word-language-modeling-benchmark-r13output/training-monolingual.tokenized.shuffled/ data/
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
245 
</td></tr></table>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
246 247 

<p>备用数据地址下载命令如下</p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
248 249 250 251 
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1
2
3
4</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>mkdir data
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
252 253 254 255 
wget https://paddlerec.bj.bcebos.com/word2vec/1-billion-word-language-modeling-benchmark-r13output.tar
tar xvf <span class="m">1</span>-billion-word-language-modeling-benchmark-r13output.tar
mv <span class="m">1</span>-billion-word-language-modeling-benchmark-r13output/training-monolingual.tokenized.shuffled/ data/
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
256 
</td></tr></table>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
257 258 

<p>为了方便快速验证，我们也提供了经典的text8样例数据集，包含1700w个词。 下载命令如下</p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
259 260 261 262 
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1
2
3
4</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>mkdir data
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
263 264 265 266 
wget https://paddlerec.bj.bcebos.com/word2vec/text.tar
tar xvf text.tar
mv text data/
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
267 
</td></tr></table>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
268 269 270 271 

<h3 id="_3">数据预处理<a class="headerlink" href="#_3" title="Permanent link">#</a></h3>
<p>以样例数据集为例进行预处理。全量数据集注意解压后以training-monolingual.tokenized.shuffled 目录为预处理目录，和样例数据集的text目录并列。</p>
<p>词典格式: 词&lt;空格&gt;词频。注意低频词用'UNK'表示</p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
272 273 274 275 276 277 278 279 280 281 282 
<p>可以按格式自建词典，如果自建词典跳过第一步。
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span> 1
 2
 3
 4
 5
 6
 7
 8
 9
10</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="n">the</span> <span class="mi">1061396</span>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
283 284 285 286 287 288 289 290 291 292 
<span class="k">of</span> <span class="mi">593677</span>
<span class="k">and</span> <span class="mi">416629</span>
<span class="n">one</span> <span class="mi">411764</span>
<span class="k">in</span> <span class="mi">372201</span>
<span class="n">a</span> <span class="mi">325873</span>
<span class="o">&lt;</span><span class="n">UNK</span><span class="o">&gt;</span> <span class="mi">324608</span>
<span class="k">to</span> <span class="mi">316376</span>
<span class="n">zero</span> <span class="mi">264975</span>
<span class="n">nine</span> <span class="mi">250430</span>
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
293 
</td></tr></table></p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
294 
<p>第一步根据英文语料生成词典，中文语料可以通过修改text_strip方法自定义处理方法。</p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
295 
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>python preprocess.py --build_dict --build_dict_corpus_dir data/text/ --dict_path data/test_build_dict
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
296 
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
297 
</td></tr></table>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
298 299 

<p>第二步根据词典将文本转成id, 同时进行downsample，按照概率过滤常见词, 同时生成word和id映射的文件，文件名为词典+"<em>word_to_id</em>"。</p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
300 
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>python preprocess.py --filter_corpus --dict_path data/test_build_dict --input_corpus_dir data/text --output_corpus_dir data/convert_text8 --min_count <span class="m">5</span> --downsample <span class="m">0</span>.001
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
301 
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
302 
</td></tr></table>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
303 304 305 

<h3 id="_4">训练<a class="headerlink" href="#_4" title="Permanent link">#</a></h3>
<p>具体的参数配置可运行</p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
306 
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>python train.py -h
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
307 
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
308 
</td></tr></table>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
309
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
310 311 
<p>单机多线程训练
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="nv">OPENBLAS_NUM_THREADS</span><span class="o">=</span><span class="m">1</span> <span class="nv">CPU_NUM</span><span class="o">=</span><span class="m">5</span> python train.py --train_data_dir data/convert_text8 --dict_path data/test_build_dict --num_passes <span class="m">10</span> --batch_size <span class="m">100</span> --model_output_dir v1_cpu5_b100_lr1dir --base_lr <span class="m">1</span>.0 --print_batch <span class="m">1000</span> --with_speed --is_sparse
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
312 
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
313 
</td></tr></table></p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
314 
<p>本地单机模拟多机训练</p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
315 
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>sh cluster_train.sh
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
316 
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
317 
</td></tr></table>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
318
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
319 320 
<p>本示例中按照单机多线程训练的命令进行训练，训练完毕后，可看到在当前文件夹下保存模型的路径为:     <code>v1_cpu5_b100_lr1dir</code>, 运行 <code>ls v1_cpu5_b100_lr1dir</code>可看到该文件夹下保存了训练的10个epoch的模型文件。
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="n">pass</span><span class="o">-</span><span class="mi">0</span>  <span class="n">pass</span><span class="o">-</span><span class="mi">1</span>  <span class="n">pass</span><span class="o">-</span><span class="mi">2</span>  <span class="n">pass</span><span class="o">-</span><span class="mi">3</span>  <span class="n">pass</span><span class="o">-</span><span class="mi">4</span>  <span class="n">pass</span><span class="o">-</span><span class="mi">5</span>  <span class="n">pass</span><span class="o">-</span><span class="mi">6</span>  <span class="n">pass</span><span class="o">-</span><span class="mi">7</span>  <span class="n">pass</span><span class="o">-</span><span class="mi">8</span>  <span class="n">pass</span><span class="o">-</span><span class="mi">9</span>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
321 
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
322 
</td></tr></table></p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
323 324 
<h3 id="_5">预测<a class="headerlink" href="#_5" title="Permanent link">#</a></h3>
<p>测试集下载命令如下</p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
325 326 327 328 
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1
2
3
4</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="c1">#全量数据集测试集</span>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
329 330 331 332 
wget https://paddlerec.bj.bcebos.com/word2vec/test_dir.tar
<span class="c1">#样本数据集测试集</span>
wget https://paddlerec.bj.bcebos.com/word2vec/test_mid_dir.tar
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
333 
</td></tr></table>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
334
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
335 336 
<p>预测命令，注意词典名称需要加后缀"<em>word_to_id</em>", 此文件是预处理阶段生成的。
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>python infer.py --infer_epoch --test_dir data/test_mid_dir --dict_path data/test_build_dict_word_to_id_ --batch_size <span class="m">20000</span> --model_dir v1_cpu5_b100_lr1dir/  --start_index <span class="m">0</span> --last_index <span class="m">9</span>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
337 
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 
</td></tr></table>
运行该预测命令, 可看到如下输出
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span> 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="p">(</span><span class="s1">&#39;start index: &#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="s1">&#39; last_index:&#39;</span><span class="p">,</span> <span class="mi">9</span><span class="p">)</span>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 
<span class="p">(</span><span class="s1">&#39;vocab_size:&#39;</span><span class="p">,</span> <span class="mi">63642</span><span class="p">)</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">249</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">0</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">014</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">590</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">1</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">033</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">982</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">2</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">055</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">1338</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">3</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">075</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">1653</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">4</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">093</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">1914</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">5</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">107</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">2204</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">6</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">124</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">2416</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">7</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">136</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">2606</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">8</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">146</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">2722</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">9</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">153</span>
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
384 
</td></tr></table></p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
385 
<h2 id="skip-gramword2vector_1">量化<code>基于skip-gram的word2vector模型</code><a class="headerlink" href="#skip-gramword2vector_1" title="Permanent link">#</a></h2>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
386 387 388 389 390 
<p>量化配置为:
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1
2
3
4</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="n">config</span> <span class="o">=</span> <span class="err">{</span>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
391 392 393 394 
        <span class="s1">&#39;params_name&#39;</span><span class="p">:</span> <span class="s1">&#39;emb&#39;</span><span class="p">,</span>
        <span class="s1">&#39;quantize_type&#39;</span><span class="p">:</span> <span class="s1">&#39;abs_max&#39;</span>
        <span class="err">}</span>
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
395 
</td></tr></table></p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
396 
<p>运行命令为：</p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
397 
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span>1</pre></div></td><td class="code"><div class="codehilite"><pre><span></span>python infer.py --infer_epoch --test_dir data/test_mid_dir --dict_path data/test_build_dict_word_to_id_ --batch_size <span class="m">20000</span> --model_dir v1_cpu5_b100_lr1dir/  --start_index <span class="m">0</span> --last_index <span class="m">9</span> --emb_quant True
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
398 
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
399 
</td></tr></table>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
400 401 

<p>运行输出为:</p>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 
<table class="codehilitetable"><tr><td class="linenos"><div class="linenodiv"><pre><span></span> 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32</pre></div></td><td class="code"><div class="codehilite"><pre><span></span><span class="p">(</span><span class="s1">&#39;start index: &#39;</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="s1">&#39; last_index:&#39;</span><span class="p">,</span> <span class="mi">9</span><span class="p">)</span>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 
<span class="p">(</span><span class="s1">&#39;vocab_size:&#39;</span><span class="p">,</span> <span class="mi">63642</span><span class="p">)</span>
<span class="n">quant_embedding</span> <span class="n">config</span> <span class="err">{</span><span class="s1">&#39;quantize_type&#39;</span><span class="p">:</span> <span class="s1">&#39;abs_max&#39;</span><span class="p">,</span> <span class="s1">&#39;params_name&#39;</span><span class="p">:</span> <span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="s1">&#39;quantize_bits&#39;</span><span class="p">:</span> <span class="mi">8</span><span class="p">,</span> <span class="s1">&#39;dtype&#39;</span><span class="p">:</span> <span class="s1">&#39;int8&#39;</span><span class="err">}</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">253</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">0</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">014</span>
<span class="n">quant_embedding</span> <span class="n">config</span> <span class="err">{</span><span class="s1">&#39;quantize_type&#39;</span><span class="p">:</span> <span class="s1">&#39;abs_max&#39;</span><span class="p">,</span> <span class="s1">&#39;params_name&#39;</span><span class="p">:</span> <span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="s1">&#39;quantize_bits&#39;</span><span class="p">:</span> <span class="mi">8</span><span class="p">,</span> <span class="s1">&#39;dtype&#39;</span><span class="p">:</span> <span class="s1">&#39;int8&#39;</span><span class="err">}</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">586</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">1</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">033</span>
<span class="n">quant_embedding</span> <span class="n">config</span> <span class="err">{</span><span class="s1">&#39;quantize_type&#39;</span><span class="p">:</span> <span class="s1">&#39;abs_max&#39;</span><span class="p">,</span> <span class="s1">&#39;params_name&#39;</span><span class="p">:</span> <span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="s1">&#39;quantize_bits&#39;</span><span class="p">:</span> <span class="mi">8</span><span class="p">,</span> <span class="s1">&#39;dtype&#39;</span><span class="p">:</span> <span class="s1">&#39;int8&#39;</span><span class="err">}</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">970</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">2</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">054</span>
<span class="n">quant_embedding</span> <span class="n">config</span> <span class="err">{</span><span class="s1">&#39;quantize_type&#39;</span><span class="p">:</span> <span class="s1">&#39;abs_max&#39;</span><span class="p">,</span> <span class="s1">&#39;params_name&#39;</span><span class="p">:</span> <span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="s1">&#39;quantize_bits&#39;</span><span class="p">:</span> <span class="mi">8</span><span class="p">,</span> <span class="s1">&#39;dtype&#39;</span><span class="p">:</span> <span class="s1">&#39;int8&#39;</span><span class="err">}</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">1364</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">3</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">077</span>
<span class="n">quant_embedding</span> <span class="n">config</span> <span class="err">{</span><span class="s1">&#39;quantize_type&#39;</span><span class="p">:</span> <span class="s1">&#39;abs_max&#39;</span><span class="p">,</span> <span class="s1">&#39;params_name&#39;</span><span class="p">:</span> <span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="s1">&#39;quantize_bits&#39;</span><span class="p">:</span> <span class="mi">8</span><span class="p">,</span> <span class="s1">&#39;dtype&#39;</span><span class="p">:</span> <span class="s1">&#39;int8&#39;</span><span class="err">}</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">1642</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">4</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">092</span>
<span class="n">quant_embedding</span> <span class="n">config</span> <span class="err">{</span><span class="s1">&#39;quantize_type&#39;</span><span class="p">:</span> <span class="s1">&#39;abs_max&#39;</span><span class="p">,</span> <span class="s1">&#39;params_name&#39;</span><span class="p">:</span> <span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="s1">&#39;quantize_bits&#39;</span><span class="p">:</span> <span class="mi">8</span><span class="p">,</span> <span class="s1">&#39;dtype&#39;</span><span class="p">:</span> <span class="s1">&#39;int8&#39;</span><span class="err">}</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">1936</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">5</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">109</span>
<span class="n">quant_embedding</span> <span class="n">config</span> <span class="err">{</span><span class="s1">&#39;quantize_type&#39;</span><span class="p">:</span> <span class="s1">&#39;abs_max&#39;</span><span class="p">,</span> <span class="s1">&#39;params_name&#39;</span><span class="p">:</span> <span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="s1">&#39;quantize_bits&#39;</span><span class="p">:</span> <span class="mi">8</span><span class="p">,</span> <span class="s1">&#39;dtype&#39;</span><span class="p">:</span> <span class="s1">&#39;int8&#39;</span><span class="err">}</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">2216</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">6</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">124</span>
<span class="n">quant_embedding</span> <span class="n">config</span> <span class="err">{</span><span class="s1">&#39;quantize_type&#39;</span><span class="p">:</span> <span class="s1">&#39;abs_max&#39;</span><span class="p">,</span> <span class="s1">&#39;params_name&#39;</span><span class="p">:</span> <span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="s1">&#39;quantize_bits&#39;</span><span class="p">:</span> <span class="mi">8</span><span class="p">,</span> <span class="s1">&#39;dtype&#39;</span><span class="p">:</span> <span class="s1">&#39;int8&#39;</span><span class="err">}</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">2419</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">7</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">136</span>
<span class="n">quant_embedding</span> <span class="n">config</span> <span class="err">{</span><span class="s1">&#39;quantize_type&#39;</span><span class="p">:</span> <span class="s1">&#39;abs_max&#39;</span><span class="p">,</span> <span class="s1">&#39;params_name&#39;</span><span class="p">:</span> <span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="s1">&#39;quantize_bits&#39;</span><span class="p">:</span> <span class="mi">8</span><span class="p">,</span> <span class="s1">&#39;dtype&#39;</span><span class="p">:</span> <span class="s1">&#39;int8&#39;</span><span class="err">}</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">2603</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">8</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">146</span>
<span class="n">quant_embedding</span> <span class="n">config</span> <span class="err">{</span><span class="s1">&#39;quantize_type&#39;</span><span class="p">:</span> <span class="s1">&#39;abs_max&#39;</span><span class="p">,</span> <span class="s1">&#39;params_name&#39;</span><span class="p">:</span> <span class="s1">&#39;emb&#39;</span><span class="p">,</span> <span class="s1">&#39;quantize_bits&#39;</span><span class="p">:</span> <span class="mi">8</span><span class="p">,</span> <span class="s1">&#39;dtype&#39;</span><span class="p">:</span> <span class="s1">&#39;int8&#39;</span><span class="err">}</span>
<span class="n">step</span><span class="p">:</span><span class="mi">1</span> <span class="mi">2719</span>
<span class="n">epoch</span><span class="p">:</span><span class="mi">9</span>          <span class="n">acc</span><span class="p">:</span><span class="mi">0</span><span class="p">.</span><span class="mi">153</span>
</pre></div>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
466 
</td></tr></table>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 

<p>量化后的模型保存在<code>./output_quant</code>中，可看到量化后的参数<code>'emb.int8'</code>的大小为3.9M, 在<code>./v1_cpu5_b100_lr1dir</code>中可看到量化前的参数<code>'emb'</code>的大小为16M。</p>
              
            </div>
          </div>
          <footer>
  
    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
      
        <a href="../nas_demo/" class="btn btn-neutral float-right" title="SA搜索">Next <span class="icon icon-circle-arrow-right"></span></a>
      
      
        <a href="../quant_aware_demo/" class="btn btn-neutral" title="量化训练"><span class="icon icon-circle-arrow-left"></span> Previous</a>
      
    </div>
  

  <hr/>

  <div role="contentinfo">
    <!-- Copyright etc -->
    
  </div>

  Built with <a href="http://www.mkdocs.org">MkDocs</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>

  <div class="rst-versions" role="note" style="cursor: pointer">
    <span class="rst-current-version" data-toggle="rst-current-version">
      
          <a href="https://github.com/PaddlePaddle/PaddleSlim/" class="fa fa-github" style="float: left; color: #fcfcfc"> GitHub</a>
      
      
        <span><a href="../quant_aware_demo/" style="color: #fcfcfc;">&laquo; Previous</a></span>
      
      
        <span style="margin-left: 15px"><a href="../nas_demo/" style="color: #fcfcfc">Next &raquo;</a></span>
      
    </span>
</div>
    <script>var base_url = '../..';</script>
    <script src="../../js/theme.js" defer></script>
      <script src="../../mathjax-config.js" defer></script>
B
Deployed 0f9c3ce with MkDocs version: 1.0.4  
baiyfbupt 已提交
517 
      <script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.0/MathJax.js?config=TeX-AMS-MML_HTMLorMML" defer></script>
B
Deployed 09f99c9 with MkDocs version: 1.0.4  
baiyfbupt 已提交
518 519 520 521 
      <script src="../../search/main.js" defer></script>

</body>
</html>