mirror of
https://github.com/labmlai/annotated_deep_learning_paper_implementations.git
synced 2025-08-14 01:13:00 +08:00
📚 compressive transformer & 📇 helpers version
This commit is contained in:
Varuna Jayasiri
@ -113,7 +113,7 @@
|
||||
|
||||
<url>
|
||||
<loc>https://nn.labml.ai/normalization/batch_norm/index.html</loc>
|
||||
<lastmod>2021-02-12T16:30:00+00:00</lastmod>
|
||||
<lastmod>2021-02-15T16:30:00+00:00</lastmod>
|
||||
<priority>1.00</priority>
|
||||
</url>
|
||||
|
||||
@ -414,7 +414,21 @@
|
||||
|
||||
<url>
|
||||
<loc>https://nn.labml.ai/transformers/mha.html</loc>
|
||||
<lastmod>2021-02-05T16:30:00+00:00</lastmod>
|
||||
<lastmod>2021-02-17T16:30:00+00:00</lastmod>
|
||||
<priority>1.00</priority>
|
||||
</url>
|
||||
|
||||
|
||||
<url>
|
||||
<loc>https://nn.labml.ai/transformers/compressive/index.html</loc>
|
||||
<lastmod>2021-02-17T16:30:00+00:00</lastmod>
|
||||
<priority>1.00</priority>
|
||||
</url>
|
||||
|
||||
|
||||
<url>
|
||||
<loc>https://nn.labml.ai/transformers/compressive/experiment.html</loc>
|
||||
<lastmod>2021-02-17T16:30:00+00:00</lastmod>
|
||||
<priority>1.00</priority>
|
||||
</url>
|
||||
|
||||
@ -463,7 +477,7 @@
|
||||
|
||||
<url>
|
||||
<loc>https://nn.labml.ai/utils.html</loc>
|
||||
<lastmod>2020-12-10T16:30:00+00:00</lastmod>
|
||||
<lastmod>2021-02-17T16:30:00+00:00</lastmod>
|
||||
<priority>1.00</priority>
|
||||
</url>
|
||||
|
||||
|
||||
1121
docs/transformers/compressive/experiment.html
Normal file
1121
docs/transformers/compressive/experiment.html
Normal file
File diff suppressed because it is too large
Load Diff
684
docs/transformers/compressive/index.html
Normal file
684
docs/transformers/compressive/index.html
Normal file
@ -0,0 +1,684 @@
|
||||
<!DOCTYPE html>
|
||||
<html>
|
||||
<head>
|
||||
<meta http-equiv="content-type" content="text/html;charset=utf-8"/>
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0"/>
|
||||
<meta name="description" content=""/>
|
||||
|
||||
<meta name="twitter:card" content="summary"/>
|
||||
<meta name="twitter:image:src" content="https://avatars1.githubusercontent.com/u/64068543?s=400&v=4"/>
|
||||
<meta name="twitter:title" content="__init__.py"/>
|
||||
<meta name="twitter:description" content=""/>
|
||||
<meta name="twitter:site" content="@labmlai"/>
|
||||
<meta name="twitter:creator" content="@labmlai"/>
|
||||
|
||||
<meta property="og:url" content="https://nn.labml.ai/transformers/compressive/index.html"/>
|
||||
<meta property="og:title" content="__init__.py"/>
|
||||
<meta property="og:image" content="https://avatars1.githubusercontent.com/u/64068543?s=400&v=4"/>
|
||||
<meta property="og:site_name" content="LabML Neural Networks"/>
|
||||
<meta property="og:type" content="object"/>
|
||||
<meta property="og:title" content="__init__.py"/>
|
||||
<meta property="og:description" content=""/>
|
||||
|
||||
<title>__init__.py</title>
|
||||
<link rel="shortcut icon" href="/icon.png"/>
|
||||
<link rel="stylesheet" href="../../pylit.css">
|
||||
<link rel="canonical" href="https://nn.labml.ai/transformers/compressive/index.html"/>
|
||||
<!-- Global site tag (gtag.js) - Google Analytics -->
|
||||
<script async src="https://www.googletagmanager.com/gtag/js?id=G-4V3HC8HBLH"></script>
|
||||
<script>
|
||||
window.dataLayer = window.dataLayer || [];
|
||||
|
||||
function gtag() {
|
||||
dataLayer.push(arguments);
|
||||
}
|
||||
|
||||
gtag('js', new Date());
|
||||
|
||||
gtag('config', 'G-4V3HC8HBLH');
|
||||
</script>
|
||||
</head>
|
||||
<body>
|
||||
<div id='container'>
|
||||
<div id="background"></div>
|
||||
<div class='section'>
|
||||
<div class='docs'>
|
||||
<p>
|
||||
<a class="parent" href="/">home</a>
|
||||
<a class="parent" href="../index.html">transformers</a>
|
||||
<a class="parent" href="index.html">compressive</a>
|
||||
</p>
|
||||
<p>
|
||||
|
||||
<a href="https://github.com/lab-ml/labml_nn/tree/master/labml_nn/transformers/compressive/__init__.py">
|
||||
<img alt="Github"
|
||||
src="https://img.shields.io/github/stars/lab-ml/nn?style=social"
|
||||
style="max-width:100%;"/></a>
|
||||
<a href="https://join.slack.com/t/labforml/shared_invite/zt-egj9zvq9-Dl3hhZqobexgT7aVKnD14g/"
|
||||
rel="nofollow">
|
||||
<img alt="Join Slact"
|
||||
src="https://img.shields.io/badge/slack-chat-green.svg?logo=slack"
|
||||
style="max-width:100%;"/></a>
|
||||
<a href="https://twitter.com/labmlai"
|
||||
rel="nofollow">
|
||||
<img alt="Twitter"
|
||||
src="https://img.shields.io/twitter/follow/labmlai?style=social"
|
||||
style="max-width:100%;"/></a>
|
||||
</p>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-0'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-0'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">1</span><span></span><span class="kn">from</span> <span class="nn">typing</span> <span class="kn">import</span> <span class="n">Optional</span><span class="p">,</span> <span class="n">List</span>
|
||||
<span class="lineno">2</span>
|
||||
<span class="lineno">3</span><span class="kn">import</span> <span class="nn">torch</span>
|
||||
<span class="lineno">4</span><span class="kn">import</span> <span class="nn">torch.nn.functional</span> <span class="k">as</span> <span class="nn">F</span>
|
||||
<span class="lineno">5</span><span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span>
|
||||
<span class="lineno">6</span>
|
||||
<span class="lineno">7</span><span class="kn">from</span> <span class="nn">labml_helpers.module</span> <span class="kn">import</span> <span class="n">Module</span><span class="p">,</span> <span class="n">TypedModuleList</span>
|
||||
<span class="lineno">8</span><span class="kn">from</span> <span class="nn">labml_nn.transformers.feed_forward</span> <span class="kn">import</span> <span class="n">FeedForward</span>
|
||||
<span class="lineno">9</span><span class="kn">from</span> <span class="nn">labml_nn.transformers.mha</span> <span class="kn">import</span> <span class="n">PrepareForMultiHeadAttention</span>
|
||||
<span class="lineno">10</span><span class="kn">from</span> <span class="nn">labml_nn.transformers.xl.relative_mha</span> <span class="kn">import</span> <span class="n">RelativeMultiHeadAttention</span>
|
||||
<span class="lineno">11</span><span class="kn">from</span> <span class="nn">labml_nn.utils</span> <span class="kn">import</span> <span class="n">clone_module_list</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-1'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-1'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">14</span><span class="k">class</span> <span class="nc">Conv1dCompression</span><span class="p">(</span><span class="n">Module</span><span class="p">):</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-2'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-2'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">15</span> <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">compression_ratio</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">d_model</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
|
||||
<span class="lineno">16</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
||||
<span class="lineno">17</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Conv1d</span><span class="p">(</span><span class="n">d_model</span><span class="p">,</span> <span class="n">d_model</span><span class="p">,</span> <span class="n">kernel_size</span><span class="o">=</span><span class="n">compression_ratio</span><span class="p">,</span> <span class="n">stride</span><span class="o">=</span><span class="n">compression_ratio</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-3'>
|
||||
<div class='docs doc-strings'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-3'>#</a>
|
||||
</div>
|
||||
<ul>
|
||||
<li><code>mem</code> has shape <code>[seq_len, batch, d_model]</code></li>
|
||||
</ul>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">19</span> <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">mem</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-4'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-4'>#</a>
|
||||
</div>
|
||||
<p>Change the dimensions of <code>mem</code> so that we can run it through the convolution layer.
|
||||
The convolution layer accepts in the form <code>[batch, features, sequence]</code></p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">26</span> <span class="n">mem</span> <span class="o">=</span> <span class="n">mem</span><span class="o">.</span><span class="n">permute</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">0</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-5'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-5'>#</a>
|
||||
</div>
|
||||
<p>Get compressed memory</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">28</span> <span class="n">c_mem</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">conv</span><span class="p">(</span><span class="n">mem</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-6'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-6'>#</a>
|
||||
</div>
|
||||
<p>Permute back to form <code>[seq_len, batch, d_model]</code></p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">30</span> <span class="k">return</span> <span class="n">c_mem</span><span class="o">.</span><span class="n">permute</span><span class="p">(</span><span class="mi">2</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-7'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-7'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">33</span><span class="k">class</span> <span class="nc">CompressiveTransformerLayer</span><span class="p">(</span><span class="n">Module</span><span class="p">):</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-8'>
|
||||
<div class='docs doc-strings'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-8'>#</a>
|
||||
</div>
|
||||
<ul>
|
||||
<li><code>d_model</code> is the token embedding size</li>
|
||||
<li><code>self_attn</code> is the <a href="relative_mha.html">self attention module</a></li>
|
||||
<li><code>feed_forward</code> is the feed forward module</li>
|
||||
<li><code>dropout_prob</code> is the probability of dropping out after self attention and FFN</li>
|
||||
</ul>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">34</span> <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span>
|
||||
<span class="lineno">35</span> <span class="n">d_model</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span>
|
||||
<span class="lineno">36</span> <span class="n">self_attn</span><span class="p">:</span> <span class="n">RelativeMultiHeadAttention</span><span class="p">,</span>
|
||||
<span class="lineno">37</span> <span class="n">feed_forward</span><span class="p">:</span> <span class="n">FeedForward</span><span class="p">,</span>
|
||||
<span class="lineno">38</span> <span class="n">dropout_prob</span><span class="p">:</span> <span class="nb">float</span><span class="p">,</span>
|
||||
<span class="lineno">39</span> <span class="n">compress</span><span class="p">:</span> <span class="n">Conv1dCompression</span><span class="p">):</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-9'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-9'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">46</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>
|
||||
<span class="lineno">47</span> <span class="bp">self</span><span class="o">.</span><span class="n">compress</span> <span class="o">=</span> <span class="n">compress</span>
|
||||
<span class="lineno">48</span> <span class="bp">self</span><span class="o">.</span><span class="n">size</span> <span class="o">=</span> <span class="n">d_model</span>
|
||||
<span class="lineno">49</span> <span class="bp">self</span><span class="o">.</span><span class="n">self_attn</span> <span class="o">=</span> <span class="n">self_attn</span>
|
||||
<span class="lineno">50</span> <span class="bp">self</span><span class="o">.</span><span class="n">feed_forward</span> <span class="o">=</span> <span class="n">feed_forward</span>
|
||||
<span class="lineno">51</span> <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">Dropout</span><span class="p">(</span><span class="n">dropout_prob</span><span class="p">)</span>
|
||||
<span class="lineno">52</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm_self_attn</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">([</span><span class="n">d_model</span><span class="p">])</span>
|
||||
<span class="lineno">53</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm_ff</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">([</span><span class="n">d_model</span><span class="p">])</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-10'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-10'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">55</span> <span class="k">def</span> <span class="nf">with_memory</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">z</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">mem</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">c_mem</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]):</span>
|
||||
<span class="lineno">56</span> <span class="k">if</span> <span class="n">mem</span> <span class="ow">is</span> <span class="kc">None</span><span class="p">:</span>
|
||||
<span class="lineno">57</span> <span class="k">return</span> <span class="n">z</span>
|
||||
<span class="lineno">58</span>
|
||||
<span class="lineno">59</span> <span class="k">if</span> <span class="n">c_mem</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
||||
<span class="lineno">60</span> <span class="n">mem</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">c_mem</span><span class="p">,</span> <span class="n">mem</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span>
|
||||
<span class="lineno">61</span>
|
||||
<span class="lineno">62</span> <span class="n">mem</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm_self_attn</span><span class="p">(</span><span class="n">mem</span><span class="p">)</span>
|
||||
<span class="lineno">63</span> <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">cat</span><span class="p">((</span><span class="n">mem</span><span class="p">,</span> <span class="n">z</span><span class="p">),</span> <span class="n">dim</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-11'>
|
||||
<div class='docs doc-strings'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-11'>#</a>
|
||||
</div>
|
||||
<ul>
|
||||
<li><code>x</code> are the token level feature vectors of shape <code>[seq_len, batch_size, d_model]</code></li>
|
||||
<li><code>mem</code> are the past token level feature vectors of shape <code>[mem_len + c_mem_len * c, batch_size, d_model]</code></li>
|
||||
<li><code>mask</code> is a matrix of shape <code>[seq_len, c_mem_len + mem_len + seq_len, batch_size]</code> or <code>[seq_len, c_mem_len + mem_len + seq_len, 1]</code>.
|
||||
<code>mask[i, j]</code> is true if token at <code>i</code> can see token at <code>j</code>.</li>
|
||||
</ul>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">65</span> <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="o">*</span><span class="p">,</span>
|
||||
<span class="lineno">66</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span>
|
||||
<span class="lineno">67</span> <span class="n">mem</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span>
|
||||
<span class="lineno">68</span> <span class="n">c_mem</span><span class="p">:</span> <span class="n">Optional</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span>
|
||||
<span class="lineno">69</span> <span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-12'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-12'>#</a>
|
||||
</div>
|
||||
<p>Normalize the vectors before doing self attention</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">78</span> <span class="n">z</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm_self_attn</span><span class="p">(</span><span class="n">x</span><span class="p">)</span>
|
||||
<span class="lineno">79</span> <span class="n">m_z</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">with_memory</span><span class="p">(</span><span class="n">z</span><span class="p">,</span> <span class="n">mem</span><span class="p">,</span> <span class="n">c_mem</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-13'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-13'>#</a>
|
||||
</div>
|
||||
<p>Attention</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">81</span> <span class="n">self_attn</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">self_attn</span><span class="p">(</span><span class="n">query</span><span class="o">=</span><span class="n">z</span><span class="p">,</span> <span class="n">key</span><span class="o">=</span><span class="n">m_z</span><span class="p">,</span> <span class="n">value</span><span class="o">=</span><span class="n">m_z</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">mask</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-14'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-14'>#</a>
|
||||
</div>
|
||||
<p>Add the attention results</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">83</span> <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">self_attn</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-15'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-15'>#</a>
|
||||
</div>
|
||||
<p>Normalize for feed-forward</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">86</span> <span class="n">z</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm_ff</span><span class="p">(</span><span class="n">x</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-16'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-16'>#</a>
|
||||
</div>
|
||||
<p>Pass through the feed-forward network</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">88</span> <span class="n">ff</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">feed_forward</span><span class="p">(</span><span class="n">z</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-17'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-17'>#</a>
|
||||
</div>
|
||||
<p>Add the feed-forward results back</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">90</span> <span class="n">x</span> <span class="o">=</span> <span class="n">x</span> <span class="o">+</span> <span class="bp">self</span><span class="o">.</span><span class="n">dropout</span><span class="p">(</span><span class="n">ff</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-18'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-18'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">93</span> <span class="k">return</span> <span class="n">x</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-19'>
|
||||
<div class='docs doc-strings'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-19'>#</a>
|
||||
</div>
|
||||
<h2>Transformer XL Model</h2>
|
||||
<p>This consists of multiple transformer XL layers</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">96</span><span class="k">class</span> <span class="nc">CompressiveTransformer</span><span class="p">(</span><span class="n">Module</span><span class="p">):</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-20'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-20'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">103</span> <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">layer</span><span class="p">:</span> <span class="n">CompressiveTransformerLayer</span><span class="p">,</span> <span class="n">n_layers</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span>
|
||||
<span class="lineno">104</span> <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-21'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-21'>#</a>
|
||||
</div>
|
||||
<p>Make copies of the transformer layer</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">106</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers</span> <span class="o">=</span> <span class="n">clone_module_list</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="n">n_layers</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-22'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-22'>#</a>
|
||||
</div>
|
||||
<p>Final normalization layer</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">108</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">([</span><span class="n">layer</span><span class="o">.</span><span class="n">size</span><span class="p">])</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-23'>
|
||||
<div class='docs doc-strings'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-23'>#</a>
|
||||
</div>
|
||||
<ul>
|
||||
<li><code>x</code> are the token embeddings vectors of shape <code>[seq_len, batch_size, d_model]</code></li>
|
||||
<li><code>mem</code> are the past token level feature vectors of shape <code>[mem_len, batch_size, d_model]</code> for each layer</li>
|
||||
<li><code>mask</code> is the masking matrix</li>
|
||||
</ul>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">110</span> <span class="k">def</span> <span class="nf">forward</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">mem</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">c_mem</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">mask</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-24'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-24'>#</a>
|
||||
</div>
|
||||
<p>List to store token level feature vectors,
|
||||
which will be the memories for the next sequential batch.</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">118</span> <span class="n">new_mem</span> <span class="o">=</span> <span class="p">[]</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-25'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-25'>#</a>
|
||||
</div>
|
||||
<p>Run through each transformer layer</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">120</span> <span class="k">for</span> <span class="n">i</span><span class="p">,</span> <span class="n">layer</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">layers</span><span class="p">):</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-26'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-26'>#</a>
|
||||
</div>
|
||||
<p>Add to the list of feature vectors</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">122</span> <span class="n">new_mem</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">detach</span><span class="p">())</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-27'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-27'>#</a>
|
||||
</div>
|
||||
<p>Memory</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">124</span> <span class="n">m</span> <span class="o">=</span> <span class="n">mem</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="k">if</span> <span class="n">mem</span> <span class="k">else</span> <span class="kc">None</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-28'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-28'>#</a>
|
||||
</div>
|
||||
<p>Memory</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">126</span> <span class="n">cm</span> <span class="o">=</span> <span class="n">c_mem</span><span class="p">[</span><span class="n">i</span><span class="p">]</span> <span class="k">if</span> <span class="n">c_mem</span> <span class="k">else</span> <span class="kc">None</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-29'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-29'>#</a>
|
||||
</div>
|
||||
<p>Run through the transformer XL layer</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">128</span> <span class="n">x</span> <span class="o">=</span> <span class="n">layer</span><span class="p">(</span><span class="n">x</span><span class="o">=</span><span class="n">x</span><span class="p">,</span> <span class="n">mem</span><span class="o">=</span><span class="n">m</span><span class="p">,</span> <span class="n">c_mem</span><span class="o">=</span><span class="n">cm</span><span class="p">,</span> <span class="n">mask</span><span class="o">=</span><span class="n">mask</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-30'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-30'>#</a>
|
||||
</div>
|
||||
<p>Finally, normalize the vectors</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">130</span> <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">x</span><span class="p">),</span> <span class="n">new_mem</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-31'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-31'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">133</span><span class="k">class</span> <span class="nc">AttentionReconstructionLoss</span><span class="p">:</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-32'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-32'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">134</span> <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">layers</span><span class="p">:</span> <span class="n">TypedModuleList</span><span class="p">[</span><span class="n">CompressiveTransformerLayer</span><span class="p">]):</span>
|
||||
<span class="lineno">135</span> <span class="bp">self</span><span class="o">.</span><span class="n">layers</span> <span class="o">=</span> <span class="n">layers</span>
|
||||
<span class="lineno">136</span> <span class="bp">self</span><span class="o">.</span><span class="n">loss_func</span> <span class="o">=</span> <span class="n">nn</span><span class="o">.</span><span class="n">MSELoss</span><span class="p">()</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-33'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-33'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">138</span> <span class="k">def</span> <span class="nf">prepare_for_attn</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">pmha</span><span class="p">:</span> <span class="n">PrepareForMultiHeadAttention</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
|
||||
<span class="lineno">139</span> <span class="n">head_shape</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">shape</span><span class="p">[:</span><span class="o">-</span><span class="mi">1</span><span class="p">]</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-34'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-34'>#</a>
|
||||
</div>
|
||||
<p>Linear transform</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">142</span> <span class="n">weight</span> <span class="o">=</span> <span class="n">pmha</span><span class="o">.</span><span class="n">linear</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span>
|
||||
<span class="lineno">143</span> <span class="n">bias</span> <span class="o">=</span> <span class="n">pmha</span><span class="o">.</span><span class="n">linear</span><span class="o">.</span><span class="n">bias</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span> <span class="k">if</span> <span class="n">pmha</span><span class="o">.</span><span class="n">linear</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="kc">None</span>
|
||||
<span class="lineno">144</span> <span class="n">x</span> <span class="o">=</span> <span class="n">F</span><span class="o">.</span><span class="n">linear</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">weight</span><span class="p">,</span> <span class="n">bias</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-35'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-35'>#</a>
|
||||
</div>
|
||||
<p>Split last dimension into heads</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">147</span> <span class="n">x</span> <span class="o">=</span> <span class="n">x</span><span class="o">.</span><span class="n">view</span><span class="p">(</span><span class="o">*</span><span class="n">head_shape</span><span class="p">,</span> <span class="n">pmha</span><span class="o">.</span><span class="n">heads</span><span class="p">,</span> <span class="n">pmha</span><span class="o">.</span><span class="n">d_k</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-36'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-36'>#</a>
|
||||
</div>
|
||||
<p>Output has shape <code>[seq_len, batch_size, heads, d_k]</code> or <code>[batch_size, d_model]</code></p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">150</span> <span class="k">return</span> <span class="n">x</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-37'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-37'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">152</span> <span class="k">def</span> <span class="nf">attn</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">layer</span><span class="p">:</span> <span class="n">RelativeMultiHeadAttention</span><span class="p">,</span> <span class="n">query</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">key</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">value</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
|
||||
<span class="lineno">153</span> <span class="n">query</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">prepare_for_attn</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">query</span><span class="p">,</span> <span class="n">query</span><span class="p">)</span>
|
||||
<span class="lineno">154</span> <span class="n">key</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">prepare_for_attn</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">key</span><span class="p">,</span> <span class="n">key</span><span class="p">)</span>
|
||||
<span class="lineno">155</span> <span class="n">value</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">prepare_for_attn</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">value</span><span class="p">,</span> <span class="n">value</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-38'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-38'>#</a>
|
||||
</div>
|
||||
<p>Compute attention scores $Q K^\top$.
|
||||
This gives a tensor of shape <code>[seq_len, seq_len, batch_size, heads]</code>.</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">159</span> <span class="n">scores</span> <span class="o">=</span> <span class="n">torch</span><span class="o">.</span><span class="n">einsum</span><span class="p">(</span><span class="s1">'ibhd,jbhd->ijbh'</span><span class="p">,</span> <span class="n">query</span><span class="p">,</span> <span class="n">key</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-39'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-39'>#</a>
|
||||
</div>
|
||||
<p>Scale scores $\frac{Q K^\top}{\sqrt{d_k}}$</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">162</span> <span class="n">scores</span> <span class="o">*=</span> <span class="n">layer</span><span class="o">.</span><span class="n">scale</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-40'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-40'>#</a>
|
||||
</div>
|
||||
<p>$softmax$ attention along the key sequence dimension
|
||||
$\underset{seq}{softmax}\Bigg(\frac{Q K^\top}{\sqrt{d_k}}\Bigg)$</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">166</span> <span class="n">attn</span> <span class="o">=</span> <span class="n">layer</span><span class="o">.</span><span class="n">softmax</span><span class="p">(</span><span class="n">scores</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-41'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-41'>#</a>
|
||||
</div>
|
||||
<p>Multiply by values
|
||||
<script type="math/tex; mode=display">\underset{seq}{softmax}\Bigg(\frac{Q K^\top}{\sqrt{d_k}}\Bigg)V</script>
|
||||
</p>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">170</span> <span class="k">return</span> <span class="n">torch</span><span class="o">.</span><span class="n">einsum</span><span class="p">(</span><span class="s2">"ijbh,jbhd->ibhd"</span><span class="p">,</span> <span class="n">attn</span><span class="p">,</span> <span class="n">value</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-42'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-42'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">172</span> <span class="k">def</span> <span class="nf">norm</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">ln</span><span class="p">:</span> <span class="n">nn</span><span class="o">.</span><span class="n">LayerNorm</span><span class="p">,</span> <span class="n">x</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
|
||||
<span class="lineno">173</span> <span class="n">weight</span> <span class="o">=</span> <span class="n">ln</span><span class="o">.</span><span class="n">weight</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span> <span class="k">if</span> <span class="n">ln</span><span class="o">.</span><span class="n">weight</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="kc">None</span>
|
||||
<span class="lineno">174</span> <span class="n">bias</span> <span class="o">=</span> <span class="n">ln</span><span class="o">.</span><span class="n">bias</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span> <span class="k">if</span> <span class="n">ln</span><span class="o">.</span><span class="n">bias</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span> <span class="k">else</span> <span class="kc">None</span>
|
||||
<span class="lineno">175</span>
|
||||
<span class="lineno">176</span> <span class="k">return</span> <span class="n">F</span><span class="o">.</span><span class="n">layer_norm</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">ln</span><span class="o">.</span><span class="n">normalized_shape</span><span class="p">,</span> <span class="n">weight</span><span class="p">,</span> <span class="n">bias</span><span class="p">,</span> <span class="n">ln</span><span class="o">.</span><span class="n">eps</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-43'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-43'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">178</span> <span class="k">def</span> <span class="nf">calc_loss</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">layer</span><span class="p">:</span> <span class="n">CompressiveTransformerLayer</span><span class="p">,</span> <span class="n">h</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">,</span> <span class="n">mem</span><span class="p">:</span> <span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">):</span>
|
||||
<span class="lineno">179</span> <span class="n">h</span> <span class="o">=</span> <span class="n">h</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span>
|
||||
<span class="lineno">180</span> <span class="n">mem</span> <span class="o">=</span> <span class="n">mem</span><span class="o">.</span><span class="n">detach</span><span class="p">()</span>
|
||||
<span class="lineno">181</span>
|
||||
<span class="lineno">182</span> <span class="n">c_mem</span> <span class="o">=</span> <span class="n">layer</span><span class="o">.</span><span class="n">compress</span><span class="p">(</span><span class="n">mem</span><span class="p">)</span>
|
||||
<span class="lineno">183</span>
|
||||
<span class="lineno">184</span> <span class="n">h</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">norm_self_attn</span><span class="p">,</span> <span class="n">h</span><span class="p">)</span>
|
||||
<span class="lineno">185</span> <span class="n">mem</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">norm_self_attn</span><span class="p">,</span> <span class="n">mem</span><span class="p">)</span>
|
||||
<span class="lineno">186</span> <span class="n">c_mem</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">norm</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">norm_self_attn</span><span class="p">,</span> <span class="n">c_mem</span><span class="p">)</span>
|
||||
<span class="lineno">187</span>
|
||||
<span class="lineno">188</span> <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">loss_func</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">attn</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">self_attn</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">mem</span><span class="p">,</span> <span class="n">mem</span><span class="p">),</span>
|
||||
<span class="lineno">189</span> <span class="bp">self</span><span class="o">.</span><span class="n">attn</span><span class="p">(</span><span class="n">layer</span><span class="o">.</span><span class="n">self_attn</span><span class="p">,</span> <span class="n">h</span><span class="p">,</span> <span class="n">c_mem</span><span class="p">,</span> <span class="n">c_mem</span><span class="p">))</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-44'>
|
||||
<div class='docs'>
|
||||
<div class='section-link'>
|
||||
<a href='#section-44'>#</a>
|
||||
</div>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">191</span> <span class="k">def</span> <span class="fm">__call__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">h</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">],</span> <span class="n">mem</span><span class="p">:</span> <span class="n">List</span><span class="p">[</span><span class="n">torch</span><span class="o">.</span><span class="n">Tensor</span><span class="p">]):</span>
|
||||
<span class="lineno">192</span> <span class="n">losses</span> <span class="o">=</span> <span class="p">[</span><span class="bp">self</span><span class="o">.</span><span class="n">calc_loss</span><span class="p">(</span><span class="n">layer</span><span class="p">,</span> <span class="n">h</span><span class="p">[</span><span class="n">n</span><span class="p">],</span> <span class="n">mem</span><span class="p">[</span><span class="n">n</span><span class="p">])</span> <span class="k">for</span> <span class="n">n</span><span class="p">,</span> <span class="n">layer</span> <span class="ow">in</span> <span class="nb">enumerate</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">layers</span><span class="p">)]</span>
|
||||
<span class="lineno">193</span> <span class="k">return</span> <span class="nb">sum</span><span class="p">(</span><span class="n">losses</span><span class="p">)</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
<script src="https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.4/MathJax.js?config=TeX-AMS_HTML">
|
||||
</script>
|
||||
<!-- MathJax configuration -->
|
||||
<script type="text/x-mathjax-config">
|
||||
MathJax.Hub.Config({
|
||||
tex2jax: {
|
||||
inlineMath: [ ['$','$'] ],
|
||||
displayMath: [ ['$$','$$'] ],
|
||||
processEscapes: true,
|
||||
processEnvironments: true
|
||||
},
|
||||
// Center justify equations in code and markdown cells. Elsewhere
|
||||
// we use CSS to left justify single line equations in code cells.
|
||||
displayAlign: 'center',
|
||||
"HTML-CSS": { fonts: ["TeX"] }
|
||||
});
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
</script>
|
||||
</body>
|
||||
</html>
|
||||
@ -469,7 +469,7 @@ This gives a tensor of shape <code>[seq_len, seq_len, batch_size, heads]</code>.
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">164</span> <span class="k">if</span> <span class="n">mask</span> <span class="ow">is</span> <span class="ow">not</span> <span class="kc">None</span><span class="p">:</span>
|
||||
<span class="lineno">165</span> <span class="n">scores</span> <span class="o">=</span> <span class="n">scores</span><span class="o">.</span><span class="n">masked_fill</span><span class="p">(</span><span class="n">mask</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span> <span class="o">-</span><span class="mf">1e9</span><span class="p">)</span></pre></div>
|
||||
<span class="lineno">165</span> <span class="n">scores</span> <span class="o">=</span> <span class="n">scores</span><span class="o">.</span><span class="n">masked_fill</span><span class="p">(</span><span class="n">mask</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span> <span class="nb">float</span><span class="p">(</span><span class="s1">'-inf'</span><span class="p">))</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-32'>
|
||||
|
||||
@ -75,9 +75,7 @@
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">10</span><span></span><span class="kn">import</span> <span class="nn">copy</span>
|
||||
<span class="lineno">11</span>
|
||||
<span class="lineno">12</span><span class="kn">from</span> <span class="nn">torch</span> <span class="kn">import</span> <span class="n">nn</span>
|
||||
<span class="lineno">13</span>
|
||||
<span class="lineno">14</span><span class="kn">from</span> <span class="nn">labml_helpers.module</span> <span class="kn">import</span> <span class="n">Module</span></pre></div>
|
||||
<span class="lineno">12</span><span class="kn">from</span> <span class="nn">labml_helpers.module</span> <span class="kn">import</span> <span class="n">M</span><span class="p">,</span> <span class="n">TypedModuleList</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-1'>
|
||||
@ -88,7 +86,7 @@
|
||||
<h2>Make a <code>nn.ModuleList</code> with clones of a given layer</h2>
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">17</span><span class="k">def</span> <span class="nf">clone_module_list</span><span class="p">(</span><span class="n">module</span><span class="p">:</span> <span class="n">Module</span><span class="p">,</span> <span class="n">n</span><span class="p">:</span> <span class="nb">int</span><span class="p">):</span></pre></div>
|
||||
<div class="highlight"><pre><span class="lineno">15</span><span class="k">def</span> <span class="nf">clone_module_list</span><span class="p">(</span><span class="n">module</span><span class="p">:</span> <span class="n">M</span><span class="p">,</span> <span class="n">n</span><span class="p">:</span> <span class="nb">int</span><span class="p">)</span> <span class="o">-></span> <span class="n">TypedModuleList</span><span class="p">[</span><span class="n">M</span><span class="p">]:</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
<div class='section' id='section-2'>
|
||||
@ -99,7 +97,7 @@
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">21</span> <span class="k">return</span> <span class="n">nn</span><span class="o">.</span><span class="n">ModuleList</span><span class="p">([</span><span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">module</span><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">n</span><span class="p">)])</span></pre></div>
|
||||
<div class="highlight"><pre><span class="lineno">19</span> <span class="k">return</span> <span class="n">TypedModuleList</span><span class="p">([</span><span class="n">copy</span><span class="o">.</span><span class="n">deepcopy</span><span class="p">(</span><span class="n">module</span><span class="p">)</span> <span class="k">for</span> <span class="n">_</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="n">n</span><span class="p">)])</span></pre></div>
|
||||
</div>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
@ -1,6 +1,6 @@
|
||||
torch>=1.7
|
||||
labml>=0.4.94
|
||||
labml-helpers>=0.4.74
|
||||
labml-helpers>=0.4.76
|
||||
torchvision
|
||||
numpy>=1.16.3
|
||||
matplotlib>=3.0.3
|
||||
|
||||
Reference in New Issue
Block a user