mirror of
https://github.com/labmlai/annotated_deep_learning_paper_implementations.git
synced 2025-08-15 02:07:56 +08:00
unescape *
This commit is contained in:
Varuna Jayasiri
@ -24,6 +24,8 @@
|
||||
<link rel="shortcut icon" href="/icon.png"/>
|
||||
<link rel="stylesheet" href="../../pylit.css">
|
||||
<link rel="canonical" href="https://nn.labml.ai/transformers/basic/autoregressive_experiment.html"/>
|
||||
<link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/katex@0.13.18/dist/katex.min.css" integrity="sha384-zTROYFVGOfTw7JV7KUu8udsvW2fx4lWOsCEDqhBreBwlHI4ioVRtmIvEThzJHGET" crossorigin="anonymous">
|
||||
|
||||
<!-- Global site tag (gtag.js) - Google Analytics -->
|
||||
<script async src="https://www.googletagmanager.com/gtag/js?id=G-4V3HC8HBLH"></script>
|
||||
<script>
|
||||
@ -68,8 +70,8 @@
|
||||
<a href='#section-0'>#</a>
|
||||
</div>
|
||||
<h1>Transformer Auto-Regression Experiment</h1>
|
||||
<p>This trains a simple transformer introduced in <a href="https://papers.labml.ai/paper/1706.03762">Attention Is All You Need</a>
|
||||
on an NLP auto-regression task (with Tiny Shakespeare dataset).</p>
|
||||
<p>This trains a simple transformer introduced in <a href="https://papers.labml.ai/paper/1706.03762">Attention Is All You Need</a> on an NLP auto-regression task (with Tiny Shakespeare dataset).</p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">14</span><span></span><span class="kn">import</span> <span class="nn">torch</span>
|
||||
@ -88,6 +90,7 @@ on an NLP auto-regression task (with Tiny Shakespeare dataset).</p>
|
||||
<a href='#section-1'>#</a>
|
||||
</div>
|
||||
<h2>Auto-Regressive model</h2>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">24</span><span class="k">class</span> <span class="nc">AutoregressiveTransformer</span><span class="p">(</span><span class="n">Module</span><span class="p">):</span></pre></div>
|
||||
@ -98,12 +101,13 @@ on an NLP auto-regression task (with Tiny Shakespeare dataset).</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-2'>#</a>
|
||||
</div>
|
||||
<ul>
|
||||
<li><code>encoder</code> is the transformer <a href="../models.html#Encoder">Encoder</a></li>
|
||||
<li><code>src_embed</code> is the token
|
||||
<a href="../models.html#EmbeddingsWithLearnedPositionalEncoding">embedding module (with positional encodings)</a></li>
|
||||
<li><code>generator</code> is the <a href="../models.html#Generator">final fully connected layer</a> that gives the logits.</li>
|
||||
</ul>
|
||||
<ul><li><code>encoder</code>
|
||||
is the transformer <a href="../models.html#Encoder">Encoder</a> </li>
|
||||
<li><code>src_embed</code>
|
||||
is the token <a href="../models.html#EmbeddingsWithLearnedPositionalEncoding">embedding module (with positional encodings)</a> </li>
|
||||
<li><code>generator</code>
|
||||
is the <a href="../models.html#Generator">final fully connected layer</a> that gives the logits.</li></ul>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">28</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">encoder</span><span class="p">:</span> <span class="n">Encoder</span><span class="p">,</span> <span class="n">src_embed</span><span class="p">:</span> <span class="n">Module</span><span class="p">,</span> <span class="n">generator</span><span class="p">:</span> <span class="n">Module</span><span class="p">):</span></pre></div>
|
||||
@ -128,7 +132,8 @@ on an NLP auto-regression task (with Tiny Shakespeare dataset).</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-4'>#</a>
|
||||
</div>
|
||||
<p>The mask will be initialized on the first call</p>
|
||||
<p>The mask will be initialized on the first call </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">41</span> <span class="bp">self</span><span class="o">.</span><span class="n">mask</span> <span class="o">=</span> <span class="kc">None</span></pre></div>
|
||||
@ -150,8 +155,8 @@ on an NLP auto-regression task (with Tiny Shakespeare dataset).</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-6'>#</a>
|
||||
</div>
|
||||
<p>Create subsequent mask if mask is not initialized
|
||||
or if the size of the mask is different</p>
|
||||
<p>Create subsequent mask if mask is not initialized or if the size of the mask is different </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">46</span> <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">mask</span> <span class="ow">is</span> <span class="kc">None</span> <span class="ow">or</span> <span class="bp">self</span><span class="o">.</span><span class="n">mask</span><span class="o">.</span><span class="n">size</span><span class="p">(</span><span class="mi">0</span><span class="p">)</span> <span class="o">!=</span> <span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">):</span></pre></div>
|
||||
@ -162,7 +167,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-7'>#</a>
|
||||
</div>
|
||||
<p>Subsequent mask, will mask out tokens from seeing future tokens</p>
|
||||
<p>Subsequent mask, will mask out tokens from seeing future tokens </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">48</span> <span class="bp">self</span><span class="o">.</span><span class="n">mask</span> <span class="o">=</span> <span class="n">subsequent_mask</span><span class="p">(</span><span class="nb">len</span><span class="p">(</span><span class="n">x</span><span class="p">))</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="n">x</span><span class="o">.</span><span class="n">device</span><span class="p">)</span></pre></div>
|
||||
@ -173,7 +179,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-8'>#</a>
|
||||
</div>
|
||||
<p>Get the token embeddings with positional encodings</p>
|
||||
<p>Get the token embeddings with positional encodings </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">50</span> <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">src_embed</span><span class="p">(</span><span class="n">x</span><span class="p">)</span></pre></div>
|
||||
@ -184,7 +191,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-9'>#</a>
|
||||
</div>
|
||||
<p>Transformer encoder</p>
|
||||
<p>Transformer encoder </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">52</span> <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">encoder</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">mask</span><span class="p">)</span></pre></div>
|
||||
@ -195,7 +203,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-10'>#</a>
|
||||
</div>
|
||||
<p>Get logits</p>
|
||||
<p>Get logits </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">54</span> <span class="n">x</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">generator</span><span class="p">(</span><span class="n">x</span><span class="p">)</span></pre></div>
|
||||
@ -206,8 +215,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-11'>#</a>
|
||||
</div>
|
||||
<p>Return results
|
||||
(second value is for state, since our trainer is used with RNNs also)</p>
|
||||
<p>Return results (second value is for state, since our trainer is used with RNNs also) </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">58</span> <span class="k">return</span> <span class="n">x</span><span class="p">,</span> <span class="kc">None</span></pre></div>
|
||||
@ -219,8 +228,9 @@ or if the size of the mask is different</p>
|
||||
<a href='#section-12'>#</a>
|
||||
</div>
|
||||
<h2>Configurations</h2>
|
||||
<p>This inherits from
|
||||
<a href="../../experiments/nlp_autoregression.html#NLPAutoRegressionConfigs"><code>NLPAutoRegressionConfigs</code></a></p>
|
||||
<p>This inherits from <a href="../../experiments/nlp_autoregression.html#NLPAutoRegressionConfigs"><code>NLPAutoRegressionConfigs</code>
|
||||
</a></p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">61</span><span class="k">class</span> <span class="nc">Configs</span><span class="p">(</span><span class="n">NLPAutoRegressionConfigs</span><span class="p">):</span></pre></div>
|
||||
@ -231,7 +241,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-13'>#</a>
|
||||
</div>
|
||||
<p>GPT model</p>
|
||||
<p>GPT model </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">70</span> <span class="n">model</span><span class="p">:</span> <span class="n">AutoregressiveTransformer</span></pre></div>
|
||||
@ -242,7 +253,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-14'>#</a>
|
||||
</div>
|
||||
<p>Transformer</p>
|
||||
<p>Transformer </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">72</span> <span class="n">transformer</span><span class="p">:</span> <span class="n">TransformerConfigs</span></pre></div>
|
||||
@ -254,6 +266,7 @@ or if the size of the mask is different</p>
|
||||
<a href='#section-15'>#</a>
|
||||
</div>
|
||||
<h3>Transformer configurations</h3>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">75</span><span class="nd">@option</span><span class="p">(</span><span class="n">Configs</span><span class="o">.</span><span class="n">transformer</span><span class="p">,</span> <span class="s1">'Transformer'</span><span class="p">)</span>
|
||||
@ -265,8 +278,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-16'>#</a>
|
||||
</div>
|
||||
<p>We use our
|
||||
<a href="../configs.html#TransformerConfigs">configurable transformer implementation</a></p>
|
||||
<p>We use our <a href="../configs.html#TransformerConfigs">configurable transformer implementation</a> </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">83</span> <span class="n">conf</span> <span class="o">=</span> <span class="n">TransformerConfigs</span><span class="p">()</span></pre></div>
|
||||
@ -277,7 +290,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-17'>#</a>
|
||||
</div>
|
||||
<p>Set the vocabulary sizes for embeddings and generating logits</p>
|
||||
<p>Set the vocabulary sizes for embeddings and generating logits </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">85</span> <span class="n">conf</span><span class="o">.</span><span class="n">n_src_vocab</span> <span class="o">=</span> <span class="n">c</span><span class="o">.</span><span class="n">n_tokens</span>
|
||||
@ -289,7 +303,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-18'>#</a>
|
||||
</div>
|
||||
|
||||
<p> </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">88</span> <span class="n">conf</span><span class="o">.</span><span class="n">d_model</span> <span class="o">=</span> <span class="n">c</span><span class="o">.</span><span class="n">d_model</span></pre></div>
|
||||
@ -300,7 +315,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-19'>#</a>
|
||||
</div>
|
||||
|
||||
<p> </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">91</span> <span class="k">return</span> <span class="n">conf</span></pre></div>
|
||||
@ -311,7 +327,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-20'>#</a>
|
||||
</div>
|
||||
<p>Create GPT model and initialize weights</p>
|
||||
<p> Create GPT model and initialize weights</p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">94</span><span class="nd">@option</span><span class="p">(</span><span class="n">Configs</span><span class="o">.</span><span class="n">model</span><span class="p">)</span>
|
||||
@ -349,7 +366,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-23'>#</a>
|
||||
</div>
|
||||
<p>Create experiment</p>
|
||||
<p>Create experiment </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">108</span> <span class="n">experiment</span><span class="o">.</span><span class="n">create</span><span class="p">(</span><span class="n">name</span><span class="o">=</span><span class="s2">"transformer"</span><span class="p">)</span></pre></div>
|
||||
@ -360,7 +378,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-24'>#</a>
|
||||
</div>
|
||||
<p>Create configs</p>
|
||||
<p>Create configs </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">110</span> <span class="n">conf</span> <span class="o">=</span> <span class="n">Configs</span><span class="p">()</span></pre></div>
|
||||
@ -371,7 +390,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-25'>#</a>
|
||||
</div>
|
||||
<p>Override configurations</p>
|
||||
<p>Override configurations </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">112</span> <span class="n">experiment</span><span class="o">.</span><span class="n">configs</span><span class="p">(</span><span class="n">conf</span><span class="p">,</span> <span class="p">{</span>
|
||||
@ -383,7 +403,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-26'>#</a>
|
||||
</div>
|
||||
<p>Use character level tokenizer</p>
|
||||
<p>Use character level tokenizer </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">116</span> <span class="s1">'tokenizer'</span><span class="p">:</span> <span class="s1">'character'</span><span class="p">,</span></pre></div>
|
||||
@ -394,7 +415,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-27'>#</a>
|
||||
</div>
|
||||
<p>Prompt separator is blank</p>
|
||||
<p>Prompt separator is blank </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">118</span> <span class="s1">'prompt_separator'</span><span class="p">:</span> <span class="s1">''</span><span class="p">,</span></pre></div>
|
||||
@ -405,7 +427,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-28'>#</a>
|
||||
</div>
|
||||
<p>Starting prompt for sampling</p>
|
||||
<p>Starting prompt for sampling </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">120</span> <span class="s1">'prompt'</span><span class="p">:</span> <span class="s1">'It is '</span><span class="p">,</span></pre></div>
|
||||
@ -416,7 +439,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-29'>#</a>
|
||||
</div>
|
||||
<p>Use Tiny Shakespeare dataset</p>
|
||||
<p>Use Tiny Shakespeare dataset </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">122</span> <span class="s1">'text'</span><span class="p">:</span> <span class="s1">'tiny_shakespeare'</span><span class="p">,</span></pre></div>
|
||||
@ -427,7 +451,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-30'>#</a>
|
||||
</div>
|
||||
<p>Use a context size of $256$</p>
|
||||
<p>Use a context size of <span class="katex"><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.64444em;vertical-align:0em;"></span><span class="mord">256</span></span></span></span> </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">125</span> <span class="s1">'seq_len'</span><span class="p">:</span> <span class="mi">256</span><span class="p">,</span></pre></div>
|
||||
@ -438,7 +463,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-31'>#</a>
|
||||
</div>
|
||||
<p>Train for $128$ epochs</p>
|
||||
<p>Train for <span class="katex"><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.64444em;vertical-align:0em;"></span><span class="mord">128</span></span></span></span> epochs </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">127</span> <span class="s1">'epochs'</span><span class="p">:</span> <span class="mi">128</span><span class="p">,</span></pre></div>
|
||||
@ -449,7 +475,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-32'>#</a>
|
||||
</div>
|
||||
<p>Batch size $32$</p>
|
||||
<p>Batch size <span class="katex"><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.64444em;vertical-align:0em;"></span><span class="mord">32</span></span></span></span> </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">129</span> <span class="s1">'batch_size'</span><span class="p">:</span> <span class="mi">32</span><span class="p">,</span></pre></div>
|
||||
@ -460,8 +487,8 @@ or if the size of the mask is different</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-33'>#</a>
|
||||
</div>
|
||||
<p>Switch between training and validation for $10$ times
|
||||
per epoch</p>
|
||||
<p>Switch between training and validation for <span class="katex"><span class="katex-html" aria-hidden="true"><span class="base"><span class="strut" style="height:0.64444em;vertical-align:0em;"></span><span class="mord">10</span></span></span></span> times per epoch </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">132</span> <span class="s1">'inner_iterations'</span><span class="p">:</span> <span class="mi">10</span><span class="p">,</span></pre></div>
|
||||
@ -472,7 +499,8 @@ per epoch</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-34'>#</a>
|
||||
</div>
|
||||
<p>Model size</p>
|
||||
<p>Model size </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">135</span> <span class="s1">'d_model'</span><span class="p">:</span> <span class="mi">128</span><span class="p">,</span>
|
||||
@ -484,7 +512,8 @@ per epoch</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-35'>#</a>
|
||||
</div>
|
||||
<p>Use <a href="../../optimizers/noam.html">Noam optimizer</a></p>
|
||||
<p>Use <a href="../../optimizers/noam.html">Noam optimizer</a> </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">139</span> <span class="s1">'optimizer.optimizer'</span><span class="p">:</span> <span class="s1">'Noam'</span><span class="p">,</span>
|
||||
@ -497,7 +526,8 @@ per epoch</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-36'>#</a>
|
||||
</div>
|
||||
<p>Set models for saving and loading</p>
|
||||
<p>Set models for saving and loading </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">144</span> <span class="n">experiment</span><span class="o">.</span><span class="n">add_pytorch_models</span><span class="p">({</span><span class="s1">'model'</span><span class="p">:</span> <span class="n">conf</span><span class="o">.</span><span class="n">model</span><span class="p">})</span></pre></div>
|
||||
@ -508,7 +538,8 @@ per epoch</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-37'>#</a>
|
||||
</div>
|
||||
<p>Start the experiment</p>
|
||||
<p>Start the experiment </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">147</span> <span class="k">with</span> <span class="n">experiment</span><span class="o">.</span><span class="n">start</span><span class="p">():</span></pre></div>
|
||||
@ -519,7 +550,8 @@ per epoch</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-38'>#</a>
|
||||
</div>
|
||||
<p>Run training</p>
|
||||
<p>Run training </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">149</span> <span class="n">conf</span><span class="o">.</span><span class="n">run</span><span class="p">()</span></pre></div>
|
||||
@ -530,7 +562,8 @@ per epoch</p>
|
||||
<div class='section-link'>
|
||||
<a href='#section-39'>#</a>
|
||||
</div>
|
||||
|
||||
<p> </p>
|
||||
|
||||
</div>
|
||||
<div class='code'>
|
||||
<div class="highlight"><pre><span class="lineno">153</span><span class="k">if</span> <span class="vm">__name__</span> <span class="o">==</span> <span class="s1">'__main__'</span><span class="p">:</span>
|
||||
@ -542,24 +575,6 @@ per epoch</p>
|
||||
<a href="https://labml.ai">labml.ai</a>
|
||||
</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>
|
||||
<script>
|
||||
function handleImages() {
|
||||
var images = document.querySelectorAll('p>img')
|
||||
|
||||
@ -24,6 +24,8 @@
|
||||
<link rel="shortcut icon" href="/icon.png"/>
|
||||
<link rel="stylesheet" href="../../pylit.css">
|
||||
<link rel="canonical" href="https://nn.labml.ai/transformers/basic/index.html"/>
|
||||
<link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/katex@0.13.18/dist/katex.min.css" integrity="sha384-zTROYFVGOfTw7JV7KUu8udsvW2fx4lWOsCEDqhBreBwlHI4ioVRtmIvEThzJHGET" crossorigin="anonymous">
|
||||
|
||||
<!-- Global site tag (gtag.js) - Google Analytics -->
|
||||
<script async src="https://www.googletagmanager.com/gtag/js?id=G-4V3HC8HBLH"></script>
|
||||
<script>
|
||||
@ -67,24 +69,6 @@
|
||||
<a href="https://labml.ai">labml.ai</a>
|
||||
</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>
|
||||
<script>
|
||||
function handleImages() {
|
||||
var images = document.querySelectorAll('p>img')
|
||||
|
||||
Reference in New Issue
Block a user