mirror of
https://github.com/labmlai/annotated_deep_learning_paper_implementations.git
synced 2025-08-15 10:11:39 +08:00
172 lines
9.6 KiB
HTML
172 lines
9.6 KiB
HTML
<!DOCTYPE html>
|
||
<html lang="zh">
|
||
<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="这是变压器和相关技术的 PyTorch 实现/教程的集合。"/>
|
||
|
||
<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="变压器"/>
|
||
<meta name="twitter:description" content="这是变压器和相关技术的 PyTorch 实现/教程的集合。"/>
|
||
<meta name="twitter:site" content="@labmlai"/>
|
||
<meta name="twitter:creator" content="@labmlai"/>
|
||
|
||
<meta property="og:url" content="https://nn.labml.ai/transformers/index.html"/>
|
||
<meta property="og:title" content="变压器"/>
|
||
<meta property="og:image" content="https://avatars1.githubusercontent.com/u/64068543?s=400&v=4"/>
|
||
<meta property="og:site_name" content="变压器"/>
|
||
<meta property="og:type" content="object"/>
|
||
<meta property="og:title" content="变压器"/>
|
||
<meta property="og:description" content="这是变压器和相关技术的 PyTorch 实现/教程的集合。"/>
|
||
|
||
<title>变压器</title>
|
||
<link rel="shortcut icon" href="/icon.png"/>
|
||
<link rel="stylesheet" href="../pylit.css?v=1">
|
||
<link rel="canonical" href="https://nn.labml.ai/transformers/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>
|
||
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>
|
||
</p>
|
||
<p>
|
||
<a href="https://github.com/labmlai/annotated_deep_learning_paper_implementations" target="_blank">
|
||
<img alt="Github"
|
||
src="https://img.shields.io/github/stars/labmlai/annotated_deep_learning_paper_implementations?style=social"
|
||
style="max-width:100%;"/></a>
|
||
<a href="https://twitter.com/labmlai" rel="nofollow" target="_blank">
|
||
<img alt="Twitter"
|
||
src="https://img.shields.io/twitter/follow/labmlai?style=social"
|
||
style="max-width:100%;"/></a>
|
||
</p>
|
||
<p>
|
||
<a href="https://github.com/labmlai/annotated_deep_learning_paper_implementations/tree/master/labml_nn/transformers/__init__.py" target="_blank">
|
||
View code on Github</a>
|
||
</p>
|
||
</div>
|
||
</div>
|
||
<div class='section' id='section-0'>
|
||
<div class='docs doc-strings'>
|
||
<div class='section-link'>
|
||
<a href='#section-0'>#</a>
|
||
</div>
|
||
<h1>变压器</h1>
|
||
</a><p>本模块包含 <a href="https://pytorch.org/">PyTorch 实现和论文 Attronger Is <a href="https://papers.labml.ai/paper/1706.03762">All You Need</a> 中对原创变压器的解释，以及它的衍生品和增强功能。</p>
|
||
<ul><li><a href="mha.html">多头关注</a></li>
|
||
<li><a href="models.html">变压器编码器和解码器型号</a></li>
|
||
<li><a href="feed_forward.html">位置前馈网络 (FFN)</a></li>
|
||
<li><a href="positional_encoding.html">固定位置编码</a></li></ul>
|
||
<h2><a href="xl/index.html">变压器 XL</a></h2>
|
||
<p>这使用<a href="xl/relative_mha.html">相对的多头注意力</a>实现了变形金刚 XL 模型</p>
|
||
<h2><a href="rope/index.html">旋转位置嵌入</a></h2>
|
||
<p>这实现了旋转位置嵌入 (roPE)</p>
|
||
<h2><a href="alibi/index.html">注意线性偏差</a></h2>
|
||
<p>这实现了线性偏差注意力（AliBI）。</p>
|
||
<h2><a href="retro/index.html">复古</a></h2>
|
||
<p>这实现了检索增强型转换器（RETRO）。</p>
|
||
<h2><a href="compressive/index.html">压缩变压器</a></h2>
|
||
<p>这是一种压缩变压器的实现，它通过压缩最古老的存储<a href="xl/index.html">器来延长注意力跨度，从而在Transformer XL</a> 上扩展。</p>
|
||
<h2><a href="gpt/index.html">GPT 架构</a></h2>
|
||
<p>这是 GPT-2 体系结构的实现。</p>
|
||
<h2><a href="glu_variants/simple.html">GLU 变体</a></h2>
|
||
<p>这是论文 <a href="https://papers.labml.ai/paper/2002.05202">GLU 变体改进变压器的</a>实现。</p>
|
||
<h2><a href="knn/index.html">knn-lm</a></h2>
|
||
<p>这是论文《<a href="https://papers.labml.ai/paper/1911.00172">通过记忆推广：最近邻语言模型</a>》的实现。</p>
|
||
<h2><a href="feedback/index.html">反馈变压器</a></h2>
|
||
<p>这是一篇论文《使用<a href="https://papers.labml.ai/paper/2002.09402">反馈存储器访问顺序变压器中的更高层次表示》的</a>实现。</p>
|
||
<h2><a href="switch/index.html">开关变压器</a></h2>
|
||
<p>这是论文《<a href="https://papers.labml.ai/paper/2101.03961">开关变压器：以简单高效的稀疏度缩放到万亿参数模型</a>》的微型实现。我们的实现只有几百万个参数，不对并行分布式训练进行建模。它进行单个 GPU 训练，但我们实现了白皮书中描述的切换概念。</p>
|
||
<h2><a href="fast_weights/index.html">快速重量变压器</a></h2>
|
||
<p>这是 <a href="https://papers.labml.ai/paper/2102.11174">PyTorch 中线性变压器是秘密的快速重量存储系统论文的</a>实现。</p>
|
||
<h2><a href="fnet/index.html">FNet：将令牌与傅里叶变换混合</a></h2>
|
||
<p>这是论文《<a href="https://papers.labml.ai/paper/2105.03824">FNet：将令牌与傅里叶变换混合</a>》的实现。</p>
|
||
<h2><a href="aft/index.html">免注意变压器</a></h2>
|
||
<p>这是论文《<a href="https://papers.labml.ai/paper/2105.14103">无注意力变压器》的</a>实现。</p>
|
||
<h2><a href="mlm/index.html">屏蔽语言模型</a></h2>
|
||
<p>这是在论文《B <a href="https://papers.labml.ai/paper/1810.04805">ERT：用于语言理解的深度双向变换器的预训练》中用于预训练的蒙面语言模型的</a>实现。</p>
|
||
<h2><a href="mlp_mixer/index.html">MLP 混音器：面向视觉的全 MLP 架构</a></h2>
|
||
<p>这是论文 <a href="https://papers.labml.ai/paper/2105.01601">MLP-Mixer：视觉的全 MLP 架构的</a>实现。</p>
|
||
<h2><a href="gmlp/index.html">注意 MLP (gMLP)</a></h2>
|
||
<p>这是 “<a href="https://papers.labml.ai/paper/2105.08050">注意 MLP” 一文的</a>实现。</p>
|
||
<h2><a href="vit/index.html">视觉变压器 (ViT)</a></h2>
|
||
<p>这是论文《<a href="https://papers.labml.ai/paper/2010.11929">图像值得 16x16 Words：大规模图像识别的变形金刚》的</a>实现。</p>
|
||
<h2><a href="primer_ez/index.html">Primer</a></h2>
|
||
<p>这是论文《入<a href="https://papers.labml.ai/paper/2109.08668">门：为语言建模寻找高效的变换器》的</a>实现。</p>
|
||
<h2><a href="hour_glass/index.html">沙漏</a></h2>
|
||
<p>这是论文《<a href="https://papers.labml.ai/paper/2110.13711">分层变换器是更有效的语言模型</a>》的实现</p>
|
||
|
||
</div>
|
||
<div class='code'>
|
||
<div class="highlight"><pre><span class="lineno">112</span><span></span><span class="kn">from</span> <span class="nn">.configs</span> <span class="kn">import</span> <span class="n">TransformerConfigs</span>
|
||
<span class="lineno">113</span><span class="kn">from</span> <span class="nn">.models</span> <span class="kn">import</span> <span class="n">TransformerLayer</span><span class="p">,</span> <span class="n">Encoder</span><span class="p">,</span> <span class="n">Decoder</span><span class="p">,</span> <span class="n">Generator</span><span class="p">,</span> <span class="n">EncoderDecoder</span>
|
||
<span class="lineno">114</span><span class="kn">from</span> <span class="nn">.mha</span> <span class="kn">import</span> <span class="n">MultiHeadAttention</span>
|
||
<span class="lineno">115</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></pre></div>
|
||
</div>
|
||
</div>
|
||
<div class='footer'>
|
||
<a href="https://papers.labml.ai">Trending Research Papers</a>
|
||
<a href="https://labml.ai">labml.ai</a>
|
||
</div>
|
||
</div>
|
||
<script src=../interactive.js?v=1"></script>
|
||
<script>
|
||
function handleImages() {
|
||
var images = document.querySelectorAll('p>img')
|
||
|
||
for (var i = 0; i < images.length; ++i) {
|
||
handleImage(images[i])
|
||
}
|
||
}
|
||
|
||
function handleImage(img) {
|
||
img.parentElement.style.textAlign = 'center'
|
||
|
||
var modal = document.createElement('div')
|
||
modal.id = 'modal'
|
||
|
||
var modalContent = document.createElement('div')
|
||
modal.appendChild(modalContent)
|
||
|
||
var modalImage = document.createElement('img')
|
||
modalContent.appendChild(modalImage)
|
||
|
||
var span = document.createElement('span')
|
||
span.classList.add('close')
|
||
span.textContent = 'x'
|
||
modal.appendChild(span)
|
||
|
||
img.onclick = function () {
|
||
console.log('clicked')
|
||
document.body.appendChild(modal)
|
||
modalImage.src = img.src
|
||
}
|
||
|
||
span.onclick = function () {
|
||
document.body.removeChild(modal)
|
||
}
|
||
}
|
||
|
||
handleImages()
|
||
</script>
|
||
</body>
|
||
</html> |