Built site for gh-pages
This commit is contained in:
Quarto GHA Workflow Runner
@@ -416,6 +416,7 @@ ul.task-list li input[type="checkbox"] {
|
||||
<li><a href="#core" id="toc-core" class="nav-link" data-scroll-target="#core">Core</a></li>
|
||||
<li><a href="#cli" id="toc-cli" class="nav-link" data-scroll-target="#cli">CLI</a></li>
|
||||
<li><a href="#trainers" id="toc-trainers" class="nav-link" data-scroll-target="#trainers">Trainers</a></li>
|
||||
<li><a href="#model-loading" id="toc-model-loading" class="nav-link" data-scroll-target="#model-loading">Model Loading</a></li>
|
||||
<li><a href="#mixins" id="toc-mixins" class="nav-link" data-scroll-target="#mixins">Mixins</a></li>
|
||||
<li><a href="#context-managers" id="toc-context-managers" class="nav-link" data-scroll-target="#context-managers">Context Managers</a></li>
|
||||
<li><a href="#prompt-strategies" id="toc-prompt-strategies" class="nav-link" data-scroll-target="#prompt-strategies">Prompt Strategies</a></li>
|
||||
@@ -612,6 +613,38 @@ ul.task-list li input[type="checkbox"] {
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
<section id="model-loading" class="level2">
|
||||
<h2 class="anchored" data-anchor-id="model-loading">Model Loading</h2>
|
||||
<p>Functionality for loading and patching models, tokenizers, etc.</p>
|
||||
<table class="caption-top table">
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="../../docs/api/loaders.model.html#axolotl.loaders.model">loaders.model</a></td>
|
||||
<td>Model loader class implementation for loading, configuring, and patching various</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="../../docs/api/loaders.tokenizer.html#axolotl.loaders.tokenizer">loaders.tokenizer</a></td>
|
||||
<td>Tokenizer loading functionality and associated utils</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="../../docs/api/loaders.processor.html#axolotl.loaders.processor">loaders.processor</a></td>
|
||||
<td>Processor loading functionality for multi-modal models</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="../../docs/api/loaders.adapter.html#axolotl.loaders.adapter">loaders.adapter</a></td>
|
||||
<td>Adapter loading functionality, including LoRA / QLoRA and associated utils</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="../../docs/api/loaders.patch_manager.html#axolotl.loaders.patch_manager">loaders.patch_manager</a></td>
|
||||
<td>Patch manager class implementation to complement <code>axolotl.loaders.ModelLoader</code>.</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="../../docs/api/loaders.constants.html#axolotl.loaders.constants">loaders.constants</a></td>
|
||||
<td>Shared constants for axolotl.loaders module</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
<section id="mixins" class="level2">
|
||||
<h2 class="anchored" data-anchor-id="mixins">Mixins</h2>
|
||||
<p>Mixin classes for augmenting trainers</p>
|
||||
@@ -853,6 +886,14 @@ ul.task-list li input[type="checkbox"] {
|
||||
<td><a href="../../docs/api/monkeypatch.mixtral.html#axolotl.monkeypatch.mixtral">monkeypatch.mixtral</a></td>
|
||||
<td>Patches to support multipack for mixtral</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="../../docs/api/monkeypatch.gradient_checkpointing.offload_cpu.html#axolotl.monkeypatch.gradient_checkpointing.offload_cpu">monkeypatch.gradient_checkpointing.offload_cpu</a></td>
|
||||
<td>CPU offloaded checkpointing</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="../../docs/api/monkeypatch.gradient_checkpointing.offload_disk.html#axolotl.monkeypatch.gradient_checkpointing.offload_disk">monkeypatch.gradient_checkpointing.offload_disk</a></td>
|
||||
<td>DISCO - DIsk-based Storage and Checkpointing with Optimized prefetching</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
@@ -862,25 +903,17 @@ ul.task-list li input[type="checkbox"] {
|
||||
<table class="caption-top table">
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="../../docs/api/utils.models.html#axolotl.utils.models">utils.models</a></td>
|
||||
<td>Module for models and model loading</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="../../docs/api/utils.tokenization.html#axolotl.utils.tokenization">utils.tokenization</a></td>
|
||||
<td>Module for tokenization utilities</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<tr class="even">
|
||||
<td><a href="../../docs/api/utils.chat_templates.html#axolotl.utils.chat_templates">utils.chat_templates</a></td>
|
||||
<td>This module provides functionality for selecting chat templates based on user choices.</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<tr class="odd">
|
||||
<td><a href="../../docs/api/utils.lora.html#axolotl.utils.lora">utils.lora</a></td>
|
||||
<td>module to get the state dict of a merged lora model</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="../../docs/api/utils.lora_embeddings.html#axolotl.utils.lora_embeddings">utils.lora_embeddings</a></td>
|
||||
<td>helpers for lora embeddings</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="../../docs/api/utils.model_shard_quant.html#axolotl.utils.model_shard_quant">utils.model_shard_quant</a></td>
|
||||
<td>module to handle loading model on cpu/meta device for FSDP</td>
|
||||
@@ -921,14 +954,6 @@ ul.task-list li input[type="checkbox"] {
|
||||
<td><a href="../../docs/api/utils.data.sft.html#axolotl.utils.data.sft">utils.data.sft</a></td>
|
||||
<td>data handling specific to SFT</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="../../docs/api/utils.gradient_checkpointing.offload_cpu.html#axolotl.utils.gradient_checkpointing.offload_cpu">utils.gradient_checkpointing.offload_cpu</a></td>
|
||||
<td>CPU offloaded checkpointing</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="../../docs/api/utils.gradient_checkpointing.offload_disk.html#axolotl.utils.gradient_checkpointing.offload_disk">utils.gradient_checkpointing.offload_disk</a></td>
|
||||
<td>DISCO - DIsk-based Storage and Checkpointing with Optimized prefetching</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
@@ -7,7 +7,7 @@
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
|
||||
|
||||
|
||||
<title>utils.lora_embeddings – Axolotl</title>
|
||||
<title>loaders.adapter – Axolotl</title>
|
||||
<style>
|
||||
code{white-space: pre-wrap;}
|
||||
span.smallcaps{font-variant: small-caps;}
|
||||
@@ -446,11 +446,12 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
|
||||
<h2 id="toc-title">On this page</h2>
|
||||
|
||||
<ul>
|
||||
<li><a href="#axolotl.utils.lora_embeddings" id="toc-axolotl.utils.lora_embeddings" class="nav-link active" data-scroll-target="#axolotl.utils.lora_embeddings">utils.lora_embeddings</a>
|
||||
<li><a href="#axolotl.loaders.adapter" id="toc-axolotl.loaders.adapter" class="nav-link active" data-scroll-target="#axolotl.loaders.adapter">loaders.adapter</a>
|
||||
<ul class="collapse">
|
||||
<li><a href="#functions" id="toc-functions" class="nav-link" data-scroll-target="#functions">Functions</a>
|
||||
<ul class="collapse">
|
||||
<li><a href="#axolotl.utils.lora_embeddings.get_linear_embedding_layers" id="toc-axolotl.utils.lora_embeddings.get_linear_embedding_layers" class="nav-link" data-scroll-target="#axolotl.utils.lora_embeddings.get_linear_embedding_layers">get_linear_embedding_layers</a></li>
|
||||
<li><a href="#axolotl.loaders.adapter.setup_quantized_meta_for_peft" id="toc-axolotl.loaders.adapter.setup_quantized_meta_for_peft" class="nav-link" data-scroll-target="#axolotl.loaders.adapter.setup_quantized_meta_for_peft">setup_quantized_meta_for_peft</a></li>
|
||||
<li><a href="#axolotl.loaders.adapter.setup_quantized_peft_meta_for_training" id="toc-axolotl.loaders.adapter.setup_quantized_peft_meta_for_training" class="nav-link" data-scroll-target="#axolotl.loaders.adapter.setup_quantized_peft_meta_for_training">setup_quantized_peft_meta_for_training</a></li>
|
||||
</ul></li>
|
||||
</ul></li>
|
||||
</ul>
|
||||
@@ -462,10 +463,10 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
|
||||
|
||||
|
||||
|
||||
<section id="axolotl.utils.lora_embeddings" class="level1">
|
||||
<h1>utils.lora_embeddings</h1>
|
||||
<p><code>utils.lora_embeddings</code></p>
|
||||
<p>helpers for lora embeddings</p>
|
||||
<section id="axolotl.loaders.adapter" class="level1">
|
||||
<h1>loaders.adapter</h1>
|
||||
<p><code>loaders.adapter</code></p>
|
||||
<p>Adapter loading functionality, including LoRA / QLoRA and associated utils</p>
|
||||
<section id="functions" class="level2">
|
||||
<h2 class="anchored" data-anchor-id="functions">Functions</h2>
|
||||
<table class="caption-top table">
|
||||
@@ -477,15 +478,24 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.lora_embeddings.get_linear_embedding_layers">get_linear_embedding_layers</a></td>
|
||||
<td>returns the linear embedding layers needed for loras, dependent on the model arch</td>
|
||||
<td><a href="#axolotl.loaders.adapter.setup_quantized_meta_for_peft">setup_quantized_meta_for_peft</a></td>
|
||||
<td>Replaces <code>quant_state.to</code> with a dummy function to prevent PEFT from moving <code>quant_state</code> to meta device</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.loaders.adapter.setup_quantized_peft_meta_for_training">setup_quantized_peft_meta_for_training</a></td>
|
||||
<td>Replaces dummy <code>quant_state.to</code> method with the original function to allow training to continue</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
<section id="axolotl.utils.lora_embeddings.get_linear_embedding_layers" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.lora_embeddings.get_linear_embedding_layers">get_linear_embedding_layers</h3>
|
||||
<div class="sourceCode" id="cb1"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a>utils.lora_embeddings.get_linear_embedding_layers(model_type)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>returns the linear embedding layers needed for loras, dependent on the model arch</p>
|
||||
<section id="axolotl.loaders.adapter.setup_quantized_meta_for_peft" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.loaders.adapter.setup_quantized_meta_for_peft">setup_quantized_meta_for_peft</h3>
|
||||
<div class="sourceCode" id="cb1"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a>loaders.adapter.setup_quantized_meta_for_peft(model)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Replaces <code>quant_state.to</code> with a dummy function to prevent PEFT from moving <code>quant_state</code> to meta device</p>
|
||||
</section>
|
||||
<section id="axolotl.loaders.adapter.setup_quantized_peft_meta_for_training" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.loaders.adapter.setup_quantized_peft_meta_for_training">setup_quantized_peft_meta_for_training</h3>
|
||||
<div class="sourceCode" id="cb2"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a>loaders.adapter.setup_quantized_peft_meta_for_training(model)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Replaces dummy <code>quant_state.to</code> method with the original function to allow training to continue</p>
|
||||
|
||||
|
||||
</section>
|
||||
837
docs/api/loaders.constants.html
Normal file
837
docs/api/loaders.constants.html
Normal file
@@ -0,0 +1,837 @@
|
||||
<!DOCTYPE html>
|
||||
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en"><head>
|
||||
|
||||
<meta charset="utf-8">
|
||||
<meta name="generator" content="quarto-1.7.31">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
|
||||
|
||||
|
||||
<title>loaders.constants – Axolotl</title>
|
||||
<style>
|
||||
code{white-space: pre-wrap;}
|
||||
span.smallcaps{font-variant: small-caps;}
|
||||
div.columns{display: flex; gap: min(4vw, 1.5em);}
|
||||
div.column{flex: auto; overflow-x: auto;}
|
||||
div.hanging-indent{margin-left: 1.5em; text-indent: -1.5em;}
|
||||
ul.task-list{list-style: none;}
|
||||
ul.task-list li input[type="checkbox"] {
|
||||
width: 0.8em;
|
||||
margin: 0 0.8em 0.2em -1em; /* quarto-specific, see https://github.com/quarto-dev/quarto-cli/issues/4556 */
|
||||
vertical-align: middle;
|
||||
}
|
||||
</style>
|
||||
|
||||
|
||||
<script src="../../site_libs/quarto-nav/quarto-nav.js"></script>
|
||||
<script src="../../site_libs/clipboard/clipboard.min.js"></script>
|
||||
<script src="../../site_libs/quarto-search/autocomplete.umd.js"></script>
|
||||
<script src="../../site_libs/quarto-search/fuse.min.js"></script>
|
||||
<script src="../../site_libs/quarto-search/quarto-search.js"></script>
|
||||
<meta name="quarto:offset" content="../../">
|
||||
<link href="../../favicon.jpg" rel="icon" type="image/jpeg">
|
||||
<script src="../../site_libs/quarto-html/quarto.js" type="module"></script>
|
||||
<script src="../../site_libs/quarto-html/tabsets/tabsets.js" type="module"></script>
|
||||
<script src="../../site_libs/quarto-html/popper.min.js"></script>
|
||||
<script src="../../site_libs/quarto-html/tippy.umd.min.js"></script>
|
||||
<script src="../../site_libs/quarto-html/anchor.min.js"></script>
|
||||
<link href="../../site_libs/quarto-html/tippy.css" rel="stylesheet">
|
||||
<link href="../../site_libs/quarto-html/quarto-syntax-highlighting-dark-8ef56b68f8fa1e9d2ba328e99e439f80.css" rel="stylesheet" id="quarto-text-highlighting-styles">
|
||||
<script src="../../site_libs/bootstrap/bootstrap.min.js"></script>
|
||||
<link href="../../site_libs/bootstrap/bootstrap-icons.css" rel="stylesheet">
|
||||
<link href="../../site_libs/bootstrap/bootstrap-2288ecdcbf81d2ab6432743cedd71d9a.min.css" rel="stylesheet" append-hash="true" id="quarto-bootstrap" data-mode="dark">
|
||||
<script id="quarto-search-options" type="application/json">{
|
||||
"location": "navbar",
|
||||
"copy-button": false,
|
||||
"collapse-after": 3,
|
||||
"panel-placement": "end",
|
||||
"type": "overlay",
|
||||
"limit": 50,
|
||||
"keyboard-shortcut": [
|
||||
"f",
|
||||
"/",
|
||||
"s"
|
||||
],
|
||||
"show-item-context": false,
|
||||
"language": {
|
||||
"search-no-results-text": "No results",
|
||||
"search-matching-documents-text": "matching documents",
|
||||
"search-copy-link-title": "Copy link to search",
|
||||
"search-hide-matches-text": "Hide additional matches",
|
||||
"search-more-match-text": "more match in this document",
|
||||
"search-more-matches-text": "more matches in this document",
|
||||
"search-clear-button-title": "Clear",
|
||||
"search-text-placeholder": "",
|
||||
"search-detached-cancel-button-title": "Cancel",
|
||||
"search-submit-button-title": "Submit",
|
||||
"search-label": "Search"
|
||||
}
|
||||
}</script>
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../styles.css">
|
||||
</head>
|
||||
|
||||
<body class="nav-sidebar docked nav-fixed quarto-light">
|
||||
|
||||
<div id="quarto-search-results"></div>
|
||||
<header id="quarto-header" class="headroom fixed-top">
|
||||
<nav class="navbar navbar-expand " data-bs-theme="dark">
|
||||
<div class="navbar-container container-fluid">
|
||||
<div class="navbar-brand-container mx-auto">
|
||||
<a href="../../index.html" class="navbar-brand navbar-brand-logo">
|
||||
<img src="../../image/axolotl_logo_digital_white.svg" alt="" class="navbar-logo">
|
||||
</a>
|
||||
</div>
|
||||
<div class="quarto-navbar-tools tools-wide tools-end">
|
||||
<a href="https://twitter.com/axolotl_ai" title="" class="quarto-navigation-tool px-1" aria-label=""><i class="bi bi-twitter"></i></a>
|
||||
<a href="https://github.com/axolotl-ai-cloud/axolotl/" title="" class="quarto-navigation-tool px-1" aria-label=""><i class="bi bi-github"></i></a>
|
||||
<a href="https://discord.gg/7m9sfhzaf3" title="" class="quarto-navigation-tool px-1" aria-label=""><i class="bi bi-discord"></i></a>
|
||||
</div>
|
||||
<div id="quarto-search" class="" title="Search"></div>
|
||||
</div> <!-- /container-fluid -->
|
||||
</nav>
|
||||
<nav class="quarto-secondary-nav">
|
||||
<div class="container-fluid d-flex">
|
||||
<button type="button" class="quarto-btn-toggle btn" data-bs-toggle="collapse" role="button" data-bs-target=".quarto-sidebar-collapse-item" aria-controls="quarto-sidebar" aria-expanded="false" aria-label="Toggle sidebar navigation" onclick="if (window.quartoToggleHeadroom) { window.quartoToggleHeadroom(); }">
|
||||
<i class="bi bi-layout-text-sidebar-reverse"></i>
|
||||
</button>
|
||||
<nav class="quarto-page-breadcrumbs" aria-label="breadcrumb"><ol class="breadcrumb"></ol></nav>
|
||||
<a class="flex-grow-1" role="navigation" data-bs-toggle="collapse" data-bs-target=".quarto-sidebar-collapse-item" aria-controls="quarto-sidebar" aria-expanded="false" aria-label="Toggle sidebar navigation" onclick="if (window.quartoToggleHeadroom) { window.quartoToggleHeadroom(); }">
|
||||
</a>
|
||||
</div>
|
||||
</nav>
|
||||
</header>
|
||||
<!-- content -->
|
||||
<div id="quarto-content" class="quarto-container page-columns page-rows-contents page-layout-article page-navbar">
|
||||
<!-- sidebar -->
|
||||
<nav id="quarto-sidebar" class="sidebar collapse collapse-horizontal quarto-sidebar-collapse-item sidebar-navigation docked overflow-auto">
|
||||
<div class="sidebar-menu-container">
|
||||
<ul class="list-unstyled mt-1">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../index.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Home</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-1" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Getting Started</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-1" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-1" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/getting-started.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Quickstart</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/installation.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Installation</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/inference.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Inference and Merging</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/cli.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Command Line Interface (CLI)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/config.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Config Reference</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/api" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">API Reference</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/index.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Dataset Formats</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-2" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-2" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/pretraining.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Pre-training</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/inst_tune.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Instruction Tuning</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/conversation.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Conversation</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/stepwise_supervised.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Stepwise Supervised Format</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/template_free.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Template-Free</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/tokenized.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Custom Pre-Tokenized Dataset</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-3" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Deployments</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-3" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-3" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/docker.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Docker</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multi-gpu.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Multi-GPU</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multi-node.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Multi Node</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/ray-integration.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Ray Train</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/amd_hpc.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">AMD GPUs on HPC Systems</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/mac.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Mac M-series</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-4" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">How To Guides</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-4" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-4" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multimodal.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">MultiModal / Vision Language Models (BETA)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/rlhf.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">RLHF (Beta)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/reward_modelling.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Reward Modelling</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/lr_groups.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Learning Rate Groups</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/lora_optims.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">LoRA Optimizations</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset_loading.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Dataset Loading</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-5" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Core Concepts</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-5" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-5" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/batch_vs_grad.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Batch size vs Gradient accumulation</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset_preprocessing.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Dataset Preprocessing</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multipack.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Multipack (Sample Packing)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-6" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Advanced Features</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-6" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-6" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/fsdp_qlora.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">FDSP + QLoRA</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/unsloth.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Unsloth</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/torchao.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">PyTorch ao</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/custom_integrations.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Custom Integrations</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/sequence_parallelism.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Sequence Parallelism</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-7" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Troubleshooting</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-7" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-7" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/faq.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">FAQ</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/debugging.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Debugging</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/nccl.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">NCCL</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
</ul>
|
||||
</div>
|
||||
</nav>
|
||||
<div id="quarto-sidebar-glass" class="quarto-sidebar-collapse-item" data-bs-toggle="collapse" data-bs-target=".quarto-sidebar-collapse-item"></div>
|
||||
<!-- margin-sidebar -->
|
||||
<div id="quarto-margin-sidebar" class="sidebar margin-sidebar">
|
||||
<nav id="TOC" role="doc-toc" class="toc-active">
|
||||
<h2 id="toc-title">On this page</h2>
|
||||
|
||||
<ul>
|
||||
<li><a href="#axolotl.loaders.constants" id="toc-axolotl.loaders.constants" class="nav-link active" data-scroll-target="#axolotl.loaders.constants">loaders.constants</a></li>
|
||||
</ul>
|
||||
</nav>
|
||||
</div>
|
||||
<!-- main -->
|
||||
<main class="content" id="quarto-document-content"><header id="title-block-header" class="quarto-title-block"></header>
|
||||
|
||||
|
||||
|
||||
|
||||
<section id="axolotl.loaders.constants" class="level1">
|
||||
<h1>loaders.constants</h1>
|
||||
<p><code>loaders.constants</code></p>
|
||||
<p>Shared constants for axolotl.loaders module</p>
|
||||
|
||||
|
||||
</section>
|
||||
|
||||
</main> <!-- /main -->
|
||||
<script id="quarto-html-after-body" type="application/javascript">
|
||||
window.document.addEventListener("DOMContentLoaded", function (event) {
|
||||
const icon = "";
|
||||
const anchorJS = new window.AnchorJS();
|
||||
anchorJS.options = {
|
||||
placement: 'right',
|
||||
icon: icon
|
||||
};
|
||||
anchorJS.add('.anchored');
|
||||
const isCodeAnnotation = (el) => {
|
||||
for (const clz of el.classList) {
|
||||
if (clz.startsWith('code-annotation-')) {
|
||||
return true;
|
||||
}
|
||||
}
|
||||
return false;
|
||||
}
|
||||
const onCopySuccess = function(e) {
|
||||
// button target
|
||||
const button = e.trigger;
|
||||
// don't keep focus
|
||||
button.blur();
|
||||
// flash "checked"
|
||||
button.classList.add('code-copy-button-checked');
|
||||
var currentTitle = button.getAttribute("title");
|
||||
button.setAttribute("title", "Copied!");
|
||||
let tooltip;
|
||||
if (window.bootstrap) {
|
||||
button.setAttribute("data-bs-toggle", "tooltip");
|
||||
button.setAttribute("data-bs-placement", "left");
|
||||
button.setAttribute("data-bs-title", "Copied!");
|
||||
tooltip = new bootstrap.Tooltip(button,
|
||||
{ trigger: "manual",
|
||||
customClass: "code-copy-button-tooltip",
|
||||
offset: [0, -8]});
|
||||
tooltip.show();
|
||||
}
|
||||
setTimeout(function() {
|
||||
if (tooltip) {
|
||||
tooltip.hide();
|
||||
button.removeAttribute("data-bs-title");
|
||||
button.removeAttribute("data-bs-toggle");
|
||||
button.removeAttribute("data-bs-placement");
|
||||
}
|
||||
button.setAttribute("title", currentTitle);
|
||||
button.classList.remove('code-copy-button-checked');
|
||||
}, 1000);
|
||||
// clear code selection
|
||||
e.clearSelection();
|
||||
}
|
||||
const getTextToCopy = function(trigger) {
|
||||
const codeEl = trigger.previousElementSibling.cloneNode(true);
|
||||
for (const childEl of codeEl.children) {
|
||||
if (isCodeAnnotation(childEl)) {
|
||||
childEl.remove();
|
||||
}
|
||||
}
|
||||
return codeEl.innerText;
|
||||
}
|
||||
const clipboard = new window.ClipboardJS('.code-copy-button:not([data-in-quarto-modal])', {
|
||||
text: getTextToCopy
|
||||
});
|
||||
clipboard.on('success', onCopySuccess);
|
||||
if (window.document.getElementById('quarto-embedded-source-code-modal')) {
|
||||
const clipboardModal = new window.ClipboardJS('.code-copy-button[data-in-quarto-modal]', {
|
||||
text: getTextToCopy,
|
||||
container: window.document.getElementById('quarto-embedded-source-code-modal')
|
||||
});
|
||||
clipboardModal.on('success', onCopySuccess);
|
||||
}
|
||||
var localhostRegex = new RegExp(/^(?:http|https):\/\/localhost\:?[0-9]*\//);
|
||||
var mailtoRegex = new RegExp(/^mailto:/);
|
||||
var filterRegex = new RegExp("https:\/\/docs\.axolotl\.ai");
|
||||
var isInternal = (href) => {
|
||||
return filterRegex.test(href) || localhostRegex.test(href) || mailtoRegex.test(href);
|
||||
}
|
||||
// Inspect non-navigation links and adorn them if external
|
||||
var links = window.document.querySelectorAll('a[href]:not(.nav-link):not(.navbar-brand):not(.toc-action):not(.sidebar-link):not(.sidebar-item-toggle):not(.pagination-link):not(.no-external):not([aria-hidden]):not(.dropdown-item):not(.quarto-navigation-tool):not(.about-link)');
|
||||
for (var i=0; i<links.length; i++) {
|
||||
const link = links[i];
|
||||
if (!isInternal(link.href)) {
|
||||
// undo the damage that might have been done by quarto-nav.js in the case of
|
||||
// links that we want to consider external
|
||||
if (link.dataset.originalHref !== undefined) {
|
||||
link.href = link.dataset.originalHref;
|
||||
}
|
||||
}
|
||||
}
|
||||
function tippyHover(el, contentFn, onTriggerFn, onUntriggerFn) {
|
||||
const config = {
|
||||
allowHTML: true,
|
||||
maxWidth: 500,
|
||||
delay: 100,
|
||||
arrow: false,
|
||||
appendTo: function(el) {
|
||||
return el.parentElement;
|
||||
},
|
||||
interactive: true,
|
||||
interactiveBorder: 10,
|
||||
theme: 'quarto',
|
||||
placement: 'bottom-start',
|
||||
};
|
||||
if (contentFn) {
|
||||
config.content = contentFn;
|
||||
}
|
||||
if (onTriggerFn) {
|
||||
config.onTrigger = onTriggerFn;
|
||||
}
|
||||
if (onUntriggerFn) {
|
||||
config.onUntrigger = onUntriggerFn;
|
||||
}
|
||||
window.tippy(el, config);
|
||||
}
|
||||
const noterefs = window.document.querySelectorAll('a[role="doc-noteref"]');
|
||||
for (var i=0; i<noterefs.length; i++) {
|
||||
const ref = noterefs[i];
|
||||
tippyHover(ref, function() {
|
||||
// use id or data attribute instead here
|
||||
let href = ref.getAttribute('data-footnote-href') || ref.getAttribute('href');
|
||||
try { href = new URL(href).hash; } catch {}
|
||||
const id = href.replace(/^#\/?/, "");
|
||||
const note = window.document.getElementById(id);
|
||||
if (note) {
|
||||
return note.innerHTML;
|
||||
} else {
|
||||
return "";
|
||||
}
|
||||
});
|
||||
}
|
||||
const xrefs = window.document.querySelectorAll('a.quarto-xref');
|
||||
const processXRef = (id, note) => {
|
||||
// Strip column container classes
|
||||
const stripColumnClz = (el) => {
|
||||
el.classList.remove("page-full", "page-columns");
|
||||
if (el.children) {
|
||||
for (const child of el.children) {
|
||||
stripColumnClz(child);
|
||||
}
|
||||
}
|
||||
}
|
||||
stripColumnClz(note)
|
||||
if (id === null || id.startsWith('sec-')) {
|
||||
// Special case sections, only their first couple elements
|
||||
const container = document.createElement("div");
|
||||
if (note.children && note.children.length > 2) {
|
||||
container.appendChild(note.children[0].cloneNode(true));
|
||||
for (let i = 1; i < note.children.length; i++) {
|
||||
const child = note.children[i];
|
||||
if (child.tagName === "P" && child.innerText === "") {
|
||||
continue;
|
||||
} else {
|
||||
container.appendChild(child.cloneNode(true));
|
||||
break;
|
||||
}
|
||||
}
|
||||
if (window.Quarto?.typesetMath) {
|
||||
window.Quarto.typesetMath(container);
|
||||
}
|
||||
return container.innerHTML
|
||||
} else {
|
||||
if (window.Quarto?.typesetMath) {
|
||||
window.Quarto.typesetMath(note);
|
||||
}
|
||||
return note.innerHTML;
|
||||
}
|
||||
} else {
|
||||
// Remove any anchor links if they are present
|
||||
const anchorLink = note.querySelector('a.anchorjs-link');
|
||||
if (anchorLink) {
|
||||
anchorLink.remove();
|
||||
}
|
||||
if (window.Quarto?.typesetMath) {
|
||||
window.Quarto.typesetMath(note);
|
||||
}
|
||||
if (note.classList.contains("callout")) {
|
||||
return note.outerHTML;
|
||||
} else {
|
||||
return note.innerHTML;
|
||||
}
|
||||
}
|
||||
}
|
||||
for (var i=0; i<xrefs.length; i++) {
|
||||
const xref = xrefs[i];
|
||||
tippyHover(xref, undefined, function(instance) {
|
||||
instance.disable();
|
||||
let url = xref.getAttribute('href');
|
||||
let hash = undefined;
|
||||
if (url.startsWith('#')) {
|
||||
hash = url;
|
||||
} else {
|
||||
try { hash = new URL(url).hash; } catch {}
|
||||
}
|
||||
if (hash) {
|
||||
const id = hash.replace(/^#\/?/, "");
|
||||
const note = window.document.getElementById(id);
|
||||
if (note !== null) {
|
||||
try {
|
||||
const html = processXRef(id, note.cloneNode(true));
|
||||
instance.setContent(html);
|
||||
} finally {
|
||||
instance.enable();
|
||||
instance.show();
|
||||
}
|
||||
} else {
|
||||
// See if we can fetch this
|
||||
fetch(url.split('#')[0])
|
||||
.then(res => res.text())
|
||||
.then(html => {
|
||||
const parser = new DOMParser();
|
||||
const htmlDoc = parser.parseFromString(html, "text/html");
|
||||
const note = htmlDoc.getElementById(id);
|
||||
if (note !== null) {
|
||||
const html = processXRef(id, note);
|
||||
instance.setContent(html);
|
||||
}
|
||||
}).finally(() => {
|
||||
instance.enable();
|
||||
instance.show();
|
||||
});
|
||||
}
|
||||
} else {
|
||||
// See if we can fetch a full url (with no hash to target)
|
||||
// This is a special case and we should probably do some content thinning / targeting
|
||||
fetch(url)
|
||||
.then(res => res.text())
|
||||
.then(html => {
|
||||
const parser = new DOMParser();
|
||||
const htmlDoc = parser.parseFromString(html, "text/html");
|
||||
const note = htmlDoc.querySelector('main.content');
|
||||
if (note !== null) {
|
||||
// This should only happen for chapter cross references
|
||||
// (since there is no id in the URL)
|
||||
// remove the first header
|
||||
if (note.children.length > 0 && note.children[0].tagName === "HEADER") {
|
||||
note.children[0].remove();
|
||||
}
|
||||
const html = processXRef(null, note);
|
||||
instance.setContent(html);
|
||||
}
|
||||
}).finally(() => {
|
||||
instance.enable();
|
||||
instance.show();
|
||||
});
|
||||
}
|
||||
}, function(instance) {
|
||||
});
|
||||
}
|
||||
let selectedAnnoteEl;
|
||||
const selectorForAnnotation = ( cell, annotation) => {
|
||||
let cellAttr = 'data-code-cell="' + cell + '"';
|
||||
let lineAttr = 'data-code-annotation="' + annotation + '"';
|
||||
const selector = 'span[' + cellAttr + '][' + lineAttr + ']';
|
||||
return selector;
|
||||
}
|
||||
const selectCodeLines = (annoteEl) => {
|
||||
const doc = window.document;
|
||||
const targetCell = annoteEl.getAttribute("data-target-cell");
|
||||
const targetAnnotation = annoteEl.getAttribute("data-target-annotation");
|
||||
const annoteSpan = window.document.querySelector(selectorForAnnotation(targetCell, targetAnnotation));
|
||||
const lines = annoteSpan.getAttribute("data-code-lines").split(",");
|
||||
const lineIds = lines.map((line) => {
|
||||
return targetCell + "-" + line;
|
||||
})
|
||||
let top = null;
|
||||
let height = null;
|
||||
let parent = null;
|
||||
if (lineIds.length > 0) {
|
||||
//compute the position of the single el (top and bottom and make a div)
|
||||
const el = window.document.getElementById(lineIds[0]);
|
||||
top = el.offsetTop;
|
||||
height = el.offsetHeight;
|
||||
parent = el.parentElement.parentElement;
|
||||
if (lineIds.length > 1) {
|
||||
const lastEl = window.document.getElementById(lineIds[lineIds.length - 1]);
|
||||
const bottom = lastEl.offsetTop + lastEl.offsetHeight;
|
||||
height = bottom - top;
|
||||
}
|
||||
if (top !== null && height !== null && parent !== null) {
|
||||
// cook up a div (if necessary) and position it
|
||||
let div = window.document.getElementById("code-annotation-line-highlight");
|
||||
if (div === null) {
|
||||
div = window.document.createElement("div");
|
||||
div.setAttribute("id", "code-annotation-line-highlight");
|
||||
div.style.position = 'absolute';
|
||||
parent.appendChild(div);
|
||||
}
|
||||
div.style.top = top - 2 + "px";
|
||||
div.style.height = height + 4 + "px";
|
||||
div.style.left = 0;
|
||||
let gutterDiv = window.document.getElementById("code-annotation-line-highlight-gutter");
|
||||
if (gutterDiv === null) {
|
||||
gutterDiv = window.document.createElement("div");
|
||||
gutterDiv.setAttribute("id", "code-annotation-line-highlight-gutter");
|
||||
gutterDiv.style.position = 'absolute';
|
||||
const codeCell = window.document.getElementById(targetCell);
|
||||
const gutter = codeCell.querySelector('.code-annotation-gutter');
|
||||
gutter.appendChild(gutterDiv);
|
||||
}
|
||||
gutterDiv.style.top = top - 2 + "px";
|
||||
gutterDiv.style.height = height + 4 + "px";
|
||||
}
|
||||
selectedAnnoteEl = annoteEl;
|
||||
}
|
||||
};
|
||||
const unselectCodeLines = () => {
|
||||
const elementsIds = ["code-annotation-line-highlight", "code-annotation-line-highlight-gutter"];
|
||||
elementsIds.forEach((elId) => {
|
||||
const div = window.document.getElementById(elId);
|
||||
if (div) {
|
||||
div.remove();
|
||||
}
|
||||
});
|
||||
selectedAnnoteEl = undefined;
|
||||
};
|
||||
// Handle positioning of the toggle
|
||||
window.addEventListener(
|
||||
"resize",
|
||||
throttle(() => {
|
||||
elRect = undefined;
|
||||
if (selectedAnnoteEl) {
|
||||
selectCodeLines(selectedAnnoteEl);
|
||||
}
|
||||
}, 10)
|
||||
);
|
||||
function throttle(fn, ms) {
|
||||
let throttle = false;
|
||||
let timer;
|
||||
return (...args) => {
|
||||
if(!throttle) { // first call gets through
|
||||
fn.apply(this, args);
|
||||
throttle = true;
|
||||
} else { // all the others get throttled
|
||||
if(timer) clearTimeout(timer); // cancel #2
|
||||
timer = setTimeout(() => {
|
||||
fn.apply(this, args);
|
||||
timer = throttle = false;
|
||||
}, ms);
|
||||
}
|
||||
};
|
||||
}
|
||||
// Attach click handler to the DT
|
||||
const annoteDls = window.document.querySelectorAll('dt[data-target-cell]');
|
||||
for (const annoteDlNode of annoteDls) {
|
||||
annoteDlNode.addEventListener('click', (event) => {
|
||||
const clickedEl = event.target;
|
||||
if (clickedEl !== selectedAnnoteEl) {
|
||||
unselectCodeLines();
|
||||
const activeEl = window.document.querySelector('dt[data-target-cell].code-annotation-active');
|
||||
if (activeEl) {
|
||||
activeEl.classList.remove('code-annotation-active');
|
||||
}
|
||||
selectCodeLines(clickedEl);
|
||||
clickedEl.classList.add('code-annotation-active');
|
||||
} else {
|
||||
// Unselect the line
|
||||
unselectCodeLines();
|
||||
clickedEl.classList.remove('code-annotation-active');
|
||||
}
|
||||
});
|
||||
}
|
||||
const findCites = (el) => {
|
||||
const parentEl = el.parentElement;
|
||||
if (parentEl) {
|
||||
const cites = parentEl.dataset.cites;
|
||||
if (cites) {
|
||||
return {
|
||||
el,
|
||||
cites: cites.split(' ')
|
||||
};
|
||||
} else {
|
||||
return findCites(el.parentElement)
|
||||
}
|
||||
} else {
|
||||
return undefined;
|
||||
}
|
||||
};
|
||||
var bibliorefs = window.document.querySelectorAll('a[role="doc-biblioref"]');
|
||||
for (var i=0; i<bibliorefs.length; i++) {
|
||||
const ref = bibliorefs[i];
|
||||
const citeInfo = findCites(ref);
|
||||
if (citeInfo) {
|
||||
tippyHover(citeInfo.el, function() {
|
||||
var popup = window.document.createElement('div');
|
||||
citeInfo.cites.forEach(function(cite) {
|
||||
var citeDiv = window.document.createElement('div');
|
||||
citeDiv.classList.add('hanging-indent');
|
||||
citeDiv.classList.add('csl-entry');
|
||||
var biblioDiv = window.document.getElementById('ref-' + cite);
|
||||
if (biblioDiv) {
|
||||
citeDiv.innerHTML = biblioDiv.innerHTML;
|
||||
}
|
||||
popup.appendChild(citeDiv);
|
||||
});
|
||||
return popup.innerHTML;
|
||||
});
|
||||
}
|
||||
}
|
||||
});
|
||||
</script>
|
||||
</div> <!-- /content -->
|
||||
|
||||
|
||||
|
||||
|
||||
</body></html>
|
||||
1020
docs/api/loaders.model.html
Normal file
1020
docs/api/loaders.model.html
Normal file
File diff suppressed because it is too large
Load Diff
949
docs/api/loaders.patch_manager.html
Normal file
949
docs/api/loaders.patch_manager.html
Normal file
@@ -0,0 +1,949 @@
|
||||
<!DOCTYPE html>
|
||||
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en"><head>
|
||||
|
||||
<meta charset="utf-8">
|
||||
<meta name="generator" content="quarto-1.7.31">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
|
||||
|
||||
|
||||
<title>loaders.patch_manager – Axolotl</title>
|
||||
<style>
|
||||
code{white-space: pre-wrap;}
|
||||
span.smallcaps{font-variant: small-caps;}
|
||||
div.columns{display: flex; gap: min(4vw, 1.5em);}
|
||||
div.column{flex: auto; overflow-x: auto;}
|
||||
div.hanging-indent{margin-left: 1.5em; text-indent: -1.5em;}
|
||||
ul.task-list{list-style: none;}
|
||||
ul.task-list li input[type="checkbox"] {
|
||||
width: 0.8em;
|
||||
margin: 0 0.8em 0.2em -1em; /* quarto-specific, see https://github.com/quarto-dev/quarto-cli/issues/4556 */
|
||||
vertical-align: middle;
|
||||
}
|
||||
/* CSS for syntax highlighting */
|
||||
html { -webkit-text-size-adjust: 100%; }
|
||||
pre > code.sourceCode { white-space: pre; position: relative; }
|
||||
pre > code.sourceCode > span { display: inline-block; line-height: 1.25; }
|
||||
pre > code.sourceCode > span:empty { height: 1.2em; }
|
||||
.sourceCode { overflow: visible; }
|
||||
code.sourceCode > span { color: inherit; text-decoration: inherit; }
|
||||
div.sourceCode { margin: 1em 0; }
|
||||
pre.sourceCode { margin: 0; }
|
||||
@media screen {
|
||||
div.sourceCode { overflow: auto; }
|
||||
}
|
||||
@media print {
|
||||
pre > code.sourceCode { white-space: pre-wrap; }
|
||||
pre > code.sourceCode > span { text-indent: -5em; padding-left: 5em; }
|
||||
}
|
||||
pre.numberSource code
|
||||
{ counter-reset: source-line 0; }
|
||||
pre.numberSource code > span
|
||||
{ position: relative; left: -4em; counter-increment: source-line; }
|
||||
pre.numberSource code > span > a:first-child::before
|
||||
{ content: counter(source-line);
|
||||
position: relative; left: -1em; text-align: right; vertical-align: baseline;
|
||||
border: none; display: inline-block;
|
||||
-webkit-touch-callout: none; -webkit-user-select: none;
|
||||
-khtml-user-select: none; -moz-user-select: none;
|
||||
-ms-user-select: none; user-select: none;
|
||||
padding: 0 4px; width: 4em;
|
||||
}
|
||||
pre.numberSource { margin-left: 3em; padding-left: 4px; }
|
||||
div.sourceCode
|
||||
{ }
|
||||
@media screen {
|
||||
pre > code.sourceCode > span > a:first-child::before { text-decoration: underline; }
|
||||
}
|
||||
</style>
|
||||
|
||||
|
||||
<script src="../../site_libs/quarto-nav/quarto-nav.js"></script>
|
||||
<script src="../../site_libs/clipboard/clipboard.min.js"></script>
|
||||
<script src="../../site_libs/quarto-search/autocomplete.umd.js"></script>
|
||||
<script src="../../site_libs/quarto-search/fuse.min.js"></script>
|
||||
<script src="../../site_libs/quarto-search/quarto-search.js"></script>
|
||||
<meta name="quarto:offset" content="../../">
|
||||
<link href="../../favicon.jpg" rel="icon" type="image/jpeg">
|
||||
<script src="../../site_libs/quarto-html/quarto.js" type="module"></script>
|
||||
<script src="../../site_libs/quarto-html/tabsets/tabsets.js" type="module"></script>
|
||||
<script src="../../site_libs/quarto-html/popper.min.js"></script>
|
||||
<script src="../../site_libs/quarto-html/tippy.umd.min.js"></script>
|
||||
<script src="../../site_libs/quarto-html/anchor.min.js"></script>
|
||||
<link href="../../site_libs/quarto-html/tippy.css" rel="stylesheet">
|
||||
<link href="../../site_libs/quarto-html/quarto-syntax-highlighting-dark-8ef56b68f8fa1e9d2ba328e99e439f80.css" rel="stylesheet" id="quarto-text-highlighting-styles">
|
||||
<script src="../../site_libs/bootstrap/bootstrap.min.js"></script>
|
||||
<link href="../../site_libs/bootstrap/bootstrap-icons.css" rel="stylesheet">
|
||||
<link href="../../site_libs/bootstrap/bootstrap-2288ecdcbf81d2ab6432743cedd71d9a.min.css" rel="stylesheet" append-hash="true" id="quarto-bootstrap" data-mode="dark">
|
||||
<script id="quarto-search-options" type="application/json">{
|
||||
"location": "navbar",
|
||||
"copy-button": false,
|
||||
"collapse-after": 3,
|
||||
"panel-placement": "end",
|
||||
"type": "overlay",
|
||||
"limit": 50,
|
||||
"keyboard-shortcut": [
|
||||
"f",
|
||||
"/",
|
||||
"s"
|
||||
],
|
||||
"show-item-context": false,
|
||||
"language": {
|
||||
"search-no-results-text": "No results",
|
||||
"search-matching-documents-text": "matching documents",
|
||||
"search-copy-link-title": "Copy link to search",
|
||||
"search-hide-matches-text": "Hide additional matches",
|
||||
"search-more-match-text": "more match in this document",
|
||||
"search-more-matches-text": "more matches in this document",
|
||||
"search-clear-button-title": "Clear",
|
||||
"search-text-placeholder": "",
|
||||
"search-detached-cancel-button-title": "Cancel",
|
||||
"search-submit-button-title": "Submit",
|
||||
"search-label": "Search"
|
||||
}
|
||||
}</script>
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../styles.css">
|
||||
</head>
|
||||
|
||||
<body class="nav-sidebar docked nav-fixed quarto-light">
|
||||
|
||||
<div id="quarto-search-results"></div>
|
||||
<header id="quarto-header" class="headroom fixed-top">
|
||||
<nav class="navbar navbar-expand " data-bs-theme="dark">
|
||||
<div class="navbar-container container-fluid">
|
||||
<div class="navbar-brand-container mx-auto">
|
||||
<a href="../../index.html" class="navbar-brand navbar-brand-logo">
|
||||
<img src="../../image/axolotl_logo_digital_white.svg" alt="" class="navbar-logo">
|
||||
</a>
|
||||
</div>
|
||||
<div class="quarto-navbar-tools tools-wide tools-end">
|
||||
<a href="https://twitter.com/axolotl_ai" title="" class="quarto-navigation-tool px-1" aria-label=""><i class="bi bi-twitter"></i></a>
|
||||
<a href="https://github.com/axolotl-ai-cloud/axolotl/" title="" class="quarto-navigation-tool px-1" aria-label=""><i class="bi bi-github"></i></a>
|
||||
<a href="https://discord.gg/7m9sfhzaf3" title="" class="quarto-navigation-tool px-1" aria-label=""><i class="bi bi-discord"></i></a>
|
||||
</div>
|
||||
<div id="quarto-search" class="" title="Search"></div>
|
||||
</div> <!-- /container-fluid -->
|
||||
</nav>
|
||||
<nav class="quarto-secondary-nav">
|
||||
<div class="container-fluid d-flex">
|
||||
<button type="button" class="quarto-btn-toggle btn" data-bs-toggle="collapse" role="button" data-bs-target=".quarto-sidebar-collapse-item" aria-controls="quarto-sidebar" aria-expanded="false" aria-label="Toggle sidebar navigation" onclick="if (window.quartoToggleHeadroom) { window.quartoToggleHeadroom(); }">
|
||||
<i class="bi bi-layout-text-sidebar-reverse"></i>
|
||||
</button>
|
||||
<nav class="quarto-page-breadcrumbs" aria-label="breadcrumb"><ol class="breadcrumb"></ol></nav>
|
||||
<a class="flex-grow-1" role="navigation" data-bs-toggle="collapse" data-bs-target=".quarto-sidebar-collapse-item" aria-controls="quarto-sidebar" aria-expanded="false" aria-label="Toggle sidebar navigation" onclick="if (window.quartoToggleHeadroom) { window.quartoToggleHeadroom(); }">
|
||||
</a>
|
||||
</div>
|
||||
</nav>
|
||||
</header>
|
||||
<!-- content -->
|
||||
<div id="quarto-content" class="quarto-container page-columns page-rows-contents page-layout-article page-navbar">
|
||||
<!-- sidebar -->
|
||||
<nav id="quarto-sidebar" class="sidebar collapse collapse-horizontal quarto-sidebar-collapse-item sidebar-navigation docked overflow-auto">
|
||||
<div class="sidebar-menu-container">
|
||||
<ul class="list-unstyled mt-1">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../index.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Home</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-1" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Getting Started</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-1" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-1" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/getting-started.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Quickstart</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/installation.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Installation</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/inference.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Inference and Merging</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/cli.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Command Line Interface (CLI)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/config.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Config Reference</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/api" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">API Reference</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/index.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Dataset Formats</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-2" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-2" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/pretraining.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Pre-training</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/inst_tune.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Instruction Tuning</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/conversation.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Conversation</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/stepwise_supervised.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Stepwise Supervised Format</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/template_free.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Template-Free</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/tokenized.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Custom Pre-Tokenized Dataset</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-3" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Deployments</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-3" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-3" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/docker.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Docker</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multi-gpu.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Multi-GPU</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multi-node.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Multi Node</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/ray-integration.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Ray Train</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/amd_hpc.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">AMD GPUs on HPC Systems</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/mac.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Mac M-series</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-4" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">How To Guides</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-4" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-4" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multimodal.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">MultiModal / Vision Language Models (BETA)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/rlhf.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">RLHF (Beta)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/reward_modelling.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Reward Modelling</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/lr_groups.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Learning Rate Groups</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/lora_optims.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">LoRA Optimizations</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset_loading.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Dataset Loading</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-5" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Core Concepts</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-5" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-5" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/batch_vs_grad.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Batch size vs Gradient accumulation</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset_preprocessing.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Dataset Preprocessing</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multipack.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Multipack (Sample Packing)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-6" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Advanced Features</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-6" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-6" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/fsdp_qlora.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">FDSP + QLoRA</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/unsloth.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Unsloth</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/torchao.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">PyTorch ao</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/custom_integrations.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Custom Integrations</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/sequence_parallelism.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Sequence Parallelism</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-7" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Troubleshooting</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-7" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-7" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/faq.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">FAQ</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/debugging.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Debugging</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/nccl.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">NCCL</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
</ul>
|
||||
</div>
|
||||
</nav>
|
||||
<div id="quarto-sidebar-glass" class="quarto-sidebar-collapse-item" data-bs-toggle="collapse" data-bs-target=".quarto-sidebar-collapse-item"></div>
|
||||
<!-- margin-sidebar -->
|
||||
<div id="quarto-margin-sidebar" class="sidebar margin-sidebar">
|
||||
<nav id="TOC" role="doc-toc" class="toc-active">
|
||||
<h2 id="toc-title">On this page</h2>
|
||||
|
||||
<ul>
|
||||
<li><a href="#axolotl.loaders.patch_manager" id="toc-axolotl.loaders.patch_manager" class="nav-link active" data-scroll-target="#axolotl.loaders.patch_manager">loaders.patch_manager</a>
|
||||
<ul class="collapse">
|
||||
<li><a href="#classes" id="toc-classes" class="nav-link" data-scroll-target="#classes">Classes</a>
|
||||
<ul class="collapse">
|
||||
<li><a href="#axolotl.loaders.patch_manager.PatchManager" id="toc-axolotl.loaders.patch_manager.PatchManager" class="nav-link" data-scroll-target="#axolotl.loaders.patch_manager.PatchManager">PatchManager</a></li>
|
||||
</ul></li>
|
||||
</ul></li>
|
||||
</ul>
|
||||
</nav>
|
||||
</div>
|
||||
<!-- main -->
|
||||
<main class="content" id="quarto-document-content"><header id="title-block-header" class="quarto-title-block"></header>
|
||||
|
||||
|
||||
|
||||
|
||||
<section id="axolotl.loaders.patch_manager" class="level1">
|
||||
<h1>loaders.patch_manager</h1>
|
||||
<p><code>loaders.patch_manager</code></p>
|
||||
<p>Patch manager class implementation to complement <code>axolotl.loaders.ModelLoader</code>.</p>
|
||||
<p>Applies pre- and post-model load patches for various fixes and optimizations.</p>
|
||||
<section id="classes" class="level2">
|
||||
<h2 class="anchored" data-anchor-id="classes">Classes</h2>
|
||||
<table class="caption-top table">
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Description</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.loaders.patch_manager.PatchManager">PatchManager</a></td>
|
||||
<td>Manages the application of patches during the model loading process.</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
<section id="axolotl.loaders.patch_manager.PatchManager" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.loaders.patch_manager.PatchManager">PatchManager</h3>
|
||||
<div class="sourceCode" id="cb1"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a>loaders.patch_manager.PatchManager(<span class="va">self</span>, cfg, model_config, inference<span class="op">=</span><span class="va">False</span>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Manages the application of patches during the model loading process.</p>
|
||||
<section id="attributes" class="level4">
|
||||
<h4 class="anchored" data-anchor-id="attributes">Attributes</h4>
|
||||
<table class="caption-top table">
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Description</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.loaders.patch_manager.PatchManager.has_flash_attn">has_flash_attn</a></td>
|
||||
<td>Check if flash attention is installed.</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
<section id="methods" class="level4">
|
||||
<h4 class="anchored" data-anchor-id="methods">Methods</h4>
|
||||
<table class="caption-top table">
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Description</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.loaders.patch_manager.PatchManager.apply_post_model_load_patches">apply_post_model_load_patches</a></td>
|
||||
<td>Apply patches that require the model instance.</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.loaders.patch_manager.PatchManager.apply_pre_model_load_patches">apply_pre_model_load_patches</a></td>
|
||||
<td>Apply pre-model load patches based on config.</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
<section id="axolotl.loaders.patch_manager.PatchManager.apply_post_model_load_patches" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.loaders.patch_manager.PatchManager.apply_post_model_load_patches">apply_post_model_load_patches</h5>
|
||||
<div class="sourceCode" id="cb2"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a>loaders.patch_manager.PatchManager.apply_post_model_load_patches(model)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Apply patches that require the model instance.</p>
|
||||
</section>
|
||||
<section id="axolotl.loaders.patch_manager.PatchManager.apply_pre_model_load_patches" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.loaders.patch_manager.PatchManager.apply_pre_model_load_patches">apply_pre_model_load_patches</h5>
|
||||
<div class="sourceCode" id="cb3"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a>loaders.patch_manager.PatchManager.apply_pre_model_load_patches()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Apply pre-model load patches based on config.</p>
|
||||
|
||||
|
||||
</section>
|
||||
</section>
|
||||
</section>
|
||||
</section>
|
||||
</section>
|
||||
|
||||
</main> <!-- /main -->
|
||||
<script id="quarto-html-after-body" type="application/javascript">
|
||||
window.document.addEventListener("DOMContentLoaded", function (event) {
|
||||
const icon = "";
|
||||
const anchorJS = new window.AnchorJS();
|
||||
anchorJS.options = {
|
||||
placement: 'right',
|
||||
icon: icon
|
||||
};
|
||||
anchorJS.add('.anchored');
|
||||
const isCodeAnnotation = (el) => {
|
||||
for (const clz of el.classList) {
|
||||
if (clz.startsWith('code-annotation-')) {
|
||||
return true;
|
||||
}
|
||||
}
|
||||
return false;
|
||||
}
|
||||
const onCopySuccess = function(e) {
|
||||
// button target
|
||||
const button = e.trigger;
|
||||
// don't keep focus
|
||||
button.blur();
|
||||
// flash "checked"
|
||||
button.classList.add('code-copy-button-checked');
|
||||
var currentTitle = button.getAttribute("title");
|
||||
button.setAttribute("title", "Copied!");
|
||||
let tooltip;
|
||||
if (window.bootstrap) {
|
||||
button.setAttribute("data-bs-toggle", "tooltip");
|
||||
button.setAttribute("data-bs-placement", "left");
|
||||
button.setAttribute("data-bs-title", "Copied!");
|
||||
tooltip = new bootstrap.Tooltip(button,
|
||||
{ trigger: "manual",
|
||||
customClass: "code-copy-button-tooltip",
|
||||
offset: [0, -8]});
|
||||
tooltip.show();
|
||||
}
|
||||
setTimeout(function() {
|
||||
if (tooltip) {
|
||||
tooltip.hide();
|
||||
button.removeAttribute("data-bs-title");
|
||||
button.removeAttribute("data-bs-toggle");
|
||||
button.removeAttribute("data-bs-placement");
|
||||
}
|
||||
button.setAttribute("title", currentTitle);
|
||||
button.classList.remove('code-copy-button-checked');
|
||||
}, 1000);
|
||||
// clear code selection
|
||||
e.clearSelection();
|
||||
}
|
||||
const getTextToCopy = function(trigger) {
|
||||
const codeEl = trigger.previousElementSibling.cloneNode(true);
|
||||
for (const childEl of codeEl.children) {
|
||||
if (isCodeAnnotation(childEl)) {
|
||||
childEl.remove();
|
||||
}
|
||||
}
|
||||
return codeEl.innerText;
|
||||
}
|
||||
const clipboard = new window.ClipboardJS('.code-copy-button:not([data-in-quarto-modal])', {
|
||||
text: getTextToCopy
|
||||
});
|
||||
clipboard.on('success', onCopySuccess);
|
||||
if (window.document.getElementById('quarto-embedded-source-code-modal')) {
|
||||
const clipboardModal = new window.ClipboardJS('.code-copy-button[data-in-quarto-modal]', {
|
||||
text: getTextToCopy,
|
||||
container: window.document.getElementById('quarto-embedded-source-code-modal')
|
||||
});
|
||||
clipboardModal.on('success', onCopySuccess);
|
||||
}
|
||||
var localhostRegex = new RegExp(/^(?:http|https):\/\/localhost\:?[0-9]*\//);
|
||||
var mailtoRegex = new RegExp(/^mailto:/);
|
||||
var filterRegex = new RegExp("https:\/\/docs\.axolotl\.ai");
|
||||
var isInternal = (href) => {
|
||||
return filterRegex.test(href) || localhostRegex.test(href) || mailtoRegex.test(href);
|
||||
}
|
||||
// Inspect non-navigation links and adorn them if external
|
||||
var links = window.document.querySelectorAll('a[href]:not(.nav-link):not(.navbar-brand):not(.toc-action):not(.sidebar-link):not(.sidebar-item-toggle):not(.pagination-link):not(.no-external):not([aria-hidden]):not(.dropdown-item):not(.quarto-navigation-tool):not(.about-link)');
|
||||
for (var i=0; i<links.length; i++) {
|
||||
const link = links[i];
|
||||
if (!isInternal(link.href)) {
|
||||
// undo the damage that might have been done by quarto-nav.js in the case of
|
||||
// links that we want to consider external
|
||||
if (link.dataset.originalHref !== undefined) {
|
||||
link.href = link.dataset.originalHref;
|
||||
}
|
||||
}
|
||||
}
|
||||
function tippyHover(el, contentFn, onTriggerFn, onUntriggerFn) {
|
||||
const config = {
|
||||
allowHTML: true,
|
||||
maxWidth: 500,
|
||||
delay: 100,
|
||||
arrow: false,
|
||||
appendTo: function(el) {
|
||||
return el.parentElement;
|
||||
},
|
||||
interactive: true,
|
||||
interactiveBorder: 10,
|
||||
theme: 'quarto',
|
||||
placement: 'bottom-start',
|
||||
};
|
||||
if (contentFn) {
|
||||
config.content = contentFn;
|
||||
}
|
||||
if (onTriggerFn) {
|
||||
config.onTrigger = onTriggerFn;
|
||||
}
|
||||
if (onUntriggerFn) {
|
||||
config.onUntrigger = onUntriggerFn;
|
||||
}
|
||||
window.tippy(el, config);
|
||||
}
|
||||
const noterefs = window.document.querySelectorAll('a[role="doc-noteref"]');
|
||||
for (var i=0; i<noterefs.length; i++) {
|
||||
const ref = noterefs[i];
|
||||
tippyHover(ref, function() {
|
||||
// use id or data attribute instead here
|
||||
let href = ref.getAttribute('data-footnote-href') || ref.getAttribute('href');
|
||||
try { href = new URL(href).hash; } catch {}
|
||||
const id = href.replace(/^#\/?/, "");
|
||||
const note = window.document.getElementById(id);
|
||||
if (note) {
|
||||
return note.innerHTML;
|
||||
} else {
|
||||
return "";
|
||||
}
|
||||
});
|
||||
}
|
||||
const xrefs = window.document.querySelectorAll('a.quarto-xref');
|
||||
const processXRef = (id, note) => {
|
||||
// Strip column container classes
|
||||
const stripColumnClz = (el) => {
|
||||
el.classList.remove("page-full", "page-columns");
|
||||
if (el.children) {
|
||||
for (const child of el.children) {
|
||||
stripColumnClz(child);
|
||||
}
|
||||
}
|
||||
}
|
||||
stripColumnClz(note)
|
||||
if (id === null || id.startsWith('sec-')) {
|
||||
// Special case sections, only their first couple elements
|
||||
const container = document.createElement("div");
|
||||
if (note.children && note.children.length > 2) {
|
||||
container.appendChild(note.children[0].cloneNode(true));
|
||||
for (let i = 1; i < note.children.length; i++) {
|
||||
const child = note.children[i];
|
||||
if (child.tagName === "P" && child.innerText === "") {
|
||||
continue;
|
||||
} else {
|
||||
container.appendChild(child.cloneNode(true));
|
||||
break;
|
||||
}
|
||||
}
|
||||
if (window.Quarto?.typesetMath) {
|
||||
window.Quarto.typesetMath(container);
|
||||
}
|
||||
return container.innerHTML
|
||||
} else {
|
||||
if (window.Quarto?.typesetMath) {
|
||||
window.Quarto.typesetMath(note);
|
||||
}
|
||||
return note.innerHTML;
|
||||
}
|
||||
} else {
|
||||
// Remove any anchor links if they are present
|
||||
const anchorLink = note.querySelector('a.anchorjs-link');
|
||||
if (anchorLink) {
|
||||
anchorLink.remove();
|
||||
}
|
||||
if (window.Quarto?.typesetMath) {
|
||||
window.Quarto.typesetMath(note);
|
||||
}
|
||||
if (note.classList.contains("callout")) {
|
||||
return note.outerHTML;
|
||||
} else {
|
||||
return note.innerHTML;
|
||||
}
|
||||
}
|
||||
}
|
||||
for (var i=0; i<xrefs.length; i++) {
|
||||
const xref = xrefs[i];
|
||||
tippyHover(xref, undefined, function(instance) {
|
||||
instance.disable();
|
||||
let url = xref.getAttribute('href');
|
||||
let hash = undefined;
|
||||
if (url.startsWith('#')) {
|
||||
hash = url;
|
||||
} else {
|
||||
try { hash = new URL(url).hash; } catch {}
|
||||
}
|
||||
if (hash) {
|
||||
const id = hash.replace(/^#\/?/, "");
|
||||
const note = window.document.getElementById(id);
|
||||
if (note !== null) {
|
||||
try {
|
||||
const html = processXRef(id, note.cloneNode(true));
|
||||
instance.setContent(html);
|
||||
} finally {
|
||||
instance.enable();
|
||||
instance.show();
|
||||
}
|
||||
} else {
|
||||
// See if we can fetch this
|
||||
fetch(url.split('#')[0])
|
||||
.then(res => res.text())
|
||||
.then(html => {
|
||||
const parser = new DOMParser();
|
||||
const htmlDoc = parser.parseFromString(html, "text/html");
|
||||
const note = htmlDoc.getElementById(id);
|
||||
if (note !== null) {
|
||||
const html = processXRef(id, note);
|
||||
instance.setContent(html);
|
||||
}
|
||||
}).finally(() => {
|
||||
instance.enable();
|
||||
instance.show();
|
||||
});
|
||||
}
|
||||
} else {
|
||||
// See if we can fetch a full url (with no hash to target)
|
||||
// This is a special case and we should probably do some content thinning / targeting
|
||||
fetch(url)
|
||||
.then(res => res.text())
|
||||
.then(html => {
|
||||
const parser = new DOMParser();
|
||||
const htmlDoc = parser.parseFromString(html, "text/html");
|
||||
const note = htmlDoc.querySelector('main.content');
|
||||
if (note !== null) {
|
||||
// This should only happen for chapter cross references
|
||||
// (since there is no id in the URL)
|
||||
// remove the first header
|
||||
if (note.children.length > 0 && note.children[0].tagName === "HEADER") {
|
||||
note.children[0].remove();
|
||||
}
|
||||
const html = processXRef(null, note);
|
||||
instance.setContent(html);
|
||||
}
|
||||
}).finally(() => {
|
||||
instance.enable();
|
||||
instance.show();
|
||||
});
|
||||
}
|
||||
}, function(instance) {
|
||||
});
|
||||
}
|
||||
let selectedAnnoteEl;
|
||||
const selectorForAnnotation = ( cell, annotation) => {
|
||||
let cellAttr = 'data-code-cell="' + cell + '"';
|
||||
let lineAttr = 'data-code-annotation="' + annotation + '"';
|
||||
const selector = 'span[' + cellAttr + '][' + lineAttr + ']';
|
||||
return selector;
|
||||
}
|
||||
const selectCodeLines = (annoteEl) => {
|
||||
const doc = window.document;
|
||||
const targetCell = annoteEl.getAttribute("data-target-cell");
|
||||
const targetAnnotation = annoteEl.getAttribute("data-target-annotation");
|
||||
const annoteSpan = window.document.querySelector(selectorForAnnotation(targetCell, targetAnnotation));
|
||||
const lines = annoteSpan.getAttribute("data-code-lines").split(",");
|
||||
const lineIds = lines.map((line) => {
|
||||
return targetCell + "-" + line;
|
||||
})
|
||||
let top = null;
|
||||
let height = null;
|
||||
let parent = null;
|
||||
if (lineIds.length > 0) {
|
||||
//compute the position of the single el (top and bottom and make a div)
|
||||
const el = window.document.getElementById(lineIds[0]);
|
||||
top = el.offsetTop;
|
||||
height = el.offsetHeight;
|
||||
parent = el.parentElement.parentElement;
|
||||
if (lineIds.length > 1) {
|
||||
const lastEl = window.document.getElementById(lineIds[lineIds.length - 1]);
|
||||
const bottom = lastEl.offsetTop + lastEl.offsetHeight;
|
||||
height = bottom - top;
|
||||
}
|
||||
if (top !== null && height !== null && parent !== null) {
|
||||
// cook up a div (if necessary) and position it
|
||||
let div = window.document.getElementById("code-annotation-line-highlight");
|
||||
if (div === null) {
|
||||
div = window.document.createElement("div");
|
||||
div.setAttribute("id", "code-annotation-line-highlight");
|
||||
div.style.position = 'absolute';
|
||||
parent.appendChild(div);
|
||||
}
|
||||
div.style.top = top - 2 + "px";
|
||||
div.style.height = height + 4 + "px";
|
||||
div.style.left = 0;
|
||||
let gutterDiv = window.document.getElementById("code-annotation-line-highlight-gutter");
|
||||
if (gutterDiv === null) {
|
||||
gutterDiv = window.document.createElement("div");
|
||||
gutterDiv.setAttribute("id", "code-annotation-line-highlight-gutter");
|
||||
gutterDiv.style.position = 'absolute';
|
||||
const codeCell = window.document.getElementById(targetCell);
|
||||
const gutter = codeCell.querySelector('.code-annotation-gutter');
|
||||
gutter.appendChild(gutterDiv);
|
||||
}
|
||||
gutterDiv.style.top = top - 2 + "px";
|
||||
gutterDiv.style.height = height + 4 + "px";
|
||||
}
|
||||
selectedAnnoteEl = annoteEl;
|
||||
}
|
||||
};
|
||||
const unselectCodeLines = () => {
|
||||
const elementsIds = ["code-annotation-line-highlight", "code-annotation-line-highlight-gutter"];
|
||||
elementsIds.forEach((elId) => {
|
||||
const div = window.document.getElementById(elId);
|
||||
if (div) {
|
||||
div.remove();
|
||||
}
|
||||
});
|
||||
selectedAnnoteEl = undefined;
|
||||
};
|
||||
// Handle positioning of the toggle
|
||||
window.addEventListener(
|
||||
"resize",
|
||||
throttle(() => {
|
||||
elRect = undefined;
|
||||
if (selectedAnnoteEl) {
|
||||
selectCodeLines(selectedAnnoteEl);
|
||||
}
|
||||
}, 10)
|
||||
);
|
||||
function throttle(fn, ms) {
|
||||
let throttle = false;
|
||||
let timer;
|
||||
return (...args) => {
|
||||
if(!throttle) { // first call gets through
|
||||
fn.apply(this, args);
|
||||
throttle = true;
|
||||
} else { // all the others get throttled
|
||||
if(timer) clearTimeout(timer); // cancel #2
|
||||
timer = setTimeout(() => {
|
||||
fn.apply(this, args);
|
||||
timer = throttle = false;
|
||||
}, ms);
|
||||
}
|
||||
};
|
||||
}
|
||||
// Attach click handler to the DT
|
||||
const annoteDls = window.document.querySelectorAll('dt[data-target-cell]');
|
||||
for (const annoteDlNode of annoteDls) {
|
||||
annoteDlNode.addEventListener('click', (event) => {
|
||||
const clickedEl = event.target;
|
||||
if (clickedEl !== selectedAnnoteEl) {
|
||||
unselectCodeLines();
|
||||
const activeEl = window.document.querySelector('dt[data-target-cell].code-annotation-active');
|
||||
if (activeEl) {
|
||||
activeEl.classList.remove('code-annotation-active');
|
||||
}
|
||||
selectCodeLines(clickedEl);
|
||||
clickedEl.classList.add('code-annotation-active');
|
||||
} else {
|
||||
// Unselect the line
|
||||
unselectCodeLines();
|
||||
clickedEl.classList.remove('code-annotation-active');
|
||||
}
|
||||
});
|
||||
}
|
||||
const findCites = (el) => {
|
||||
const parentEl = el.parentElement;
|
||||
if (parentEl) {
|
||||
const cites = parentEl.dataset.cites;
|
||||
if (cites) {
|
||||
return {
|
||||
el,
|
||||
cites: cites.split(' ')
|
||||
};
|
||||
} else {
|
||||
return findCites(el.parentElement)
|
||||
}
|
||||
} else {
|
||||
return undefined;
|
||||
}
|
||||
};
|
||||
var bibliorefs = window.document.querySelectorAll('a[role="doc-biblioref"]');
|
||||
for (var i=0; i<bibliorefs.length; i++) {
|
||||
const ref = bibliorefs[i];
|
||||
const citeInfo = findCites(ref);
|
||||
if (citeInfo) {
|
||||
tippyHover(citeInfo.el, function() {
|
||||
var popup = window.document.createElement('div');
|
||||
citeInfo.cites.forEach(function(cite) {
|
||||
var citeDiv = window.document.createElement('div');
|
||||
citeDiv.classList.add('hanging-indent');
|
||||
citeDiv.classList.add('csl-entry');
|
||||
var biblioDiv = window.document.getElementById('ref-' + cite);
|
||||
if (biblioDiv) {
|
||||
citeDiv.innerHTML = biblioDiv.innerHTML;
|
||||
}
|
||||
popup.appendChild(citeDiv);
|
||||
});
|
||||
return popup.innerHTML;
|
||||
});
|
||||
}
|
||||
}
|
||||
});
|
||||
</script>
|
||||
</div> <!-- /content -->
|
||||
|
||||
|
||||
|
||||
|
||||
</body></html>
|
||||
837
docs/api/loaders.processor.html
Normal file
837
docs/api/loaders.processor.html
Normal file
@@ -0,0 +1,837 @@
|
||||
<!DOCTYPE html>
|
||||
<html xmlns="http://www.w3.org/1999/xhtml" lang="en" xml:lang="en"><head>
|
||||
|
||||
<meta charset="utf-8">
|
||||
<meta name="generator" content="quarto-1.7.31">
|
||||
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
|
||||
|
||||
|
||||
<title>loaders.processor – Axolotl</title>
|
||||
<style>
|
||||
code{white-space: pre-wrap;}
|
||||
span.smallcaps{font-variant: small-caps;}
|
||||
div.columns{display: flex; gap: min(4vw, 1.5em);}
|
||||
div.column{flex: auto; overflow-x: auto;}
|
||||
div.hanging-indent{margin-left: 1.5em; text-indent: -1.5em;}
|
||||
ul.task-list{list-style: none;}
|
||||
ul.task-list li input[type="checkbox"] {
|
||||
width: 0.8em;
|
||||
margin: 0 0.8em 0.2em -1em; /* quarto-specific, see https://github.com/quarto-dev/quarto-cli/issues/4556 */
|
||||
vertical-align: middle;
|
||||
}
|
||||
</style>
|
||||
|
||||
|
||||
<script src="../../site_libs/quarto-nav/quarto-nav.js"></script>
|
||||
<script src="../../site_libs/clipboard/clipboard.min.js"></script>
|
||||
<script src="../../site_libs/quarto-search/autocomplete.umd.js"></script>
|
||||
<script src="../../site_libs/quarto-search/fuse.min.js"></script>
|
||||
<script src="../../site_libs/quarto-search/quarto-search.js"></script>
|
||||
<meta name="quarto:offset" content="../../">
|
||||
<link href="../../favicon.jpg" rel="icon" type="image/jpeg">
|
||||
<script src="../../site_libs/quarto-html/quarto.js" type="module"></script>
|
||||
<script src="../../site_libs/quarto-html/tabsets/tabsets.js" type="module"></script>
|
||||
<script src="../../site_libs/quarto-html/popper.min.js"></script>
|
||||
<script src="../../site_libs/quarto-html/tippy.umd.min.js"></script>
|
||||
<script src="../../site_libs/quarto-html/anchor.min.js"></script>
|
||||
<link href="../../site_libs/quarto-html/tippy.css" rel="stylesheet">
|
||||
<link href="../../site_libs/quarto-html/quarto-syntax-highlighting-dark-8ef56b68f8fa1e9d2ba328e99e439f80.css" rel="stylesheet" id="quarto-text-highlighting-styles">
|
||||
<script src="../../site_libs/bootstrap/bootstrap.min.js"></script>
|
||||
<link href="../../site_libs/bootstrap/bootstrap-icons.css" rel="stylesheet">
|
||||
<link href="../../site_libs/bootstrap/bootstrap-2288ecdcbf81d2ab6432743cedd71d9a.min.css" rel="stylesheet" append-hash="true" id="quarto-bootstrap" data-mode="dark">
|
||||
<script id="quarto-search-options" type="application/json">{
|
||||
"location": "navbar",
|
||||
"copy-button": false,
|
||||
"collapse-after": 3,
|
||||
"panel-placement": "end",
|
||||
"type": "overlay",
|
||||
"limit": 50,
|
||||
"keyboard-shortcut": [
|
||||
"f",
|
||||
"/",
|
||||
"s"
|
||||
],
|
||||
"show-item-context": false,
|
||||
"language": {
|
||||
"search-no-results-text": "No results",
|
||||
"search-matching-documents-text": "matching documents",
|
||||
"search-copy-link-title": "Copy link to search",
|
||||
"search-hide-matches-text": "Hide additional matches",
|
||||
"search-more-match-text": "more match in this document",
|
||||
"search-more-matches-text": "more matches in this document",
|
||||
"search-clear-button-title": "Clear",
|
||||
"search-text-placeholder": "",
|
||||
"search-detached-cancel-button-title": "Cancel",
|
||||
"search-submit-button-title": "Submit",
|
||||
"search-label": "Search"
|
||||
}
|
||||
}</script>
|
||||
|
||||
|
||||
<link rel="stylesheet" href="../../styles.css">
|
||||
</head>
|
||||
|
||||
<body class="nav-sidebar docked nav-fixed quarto-light">
|
||||
|
||||
<div id="quarto-search-results"></div>
|
||||
<header id="quarto-header" class="headroom fixed-top">
|
||||
<nav class="navbar navbar-expand " data-bs-theme="dark">
|
||||
<div class="navbar-container container-fluid">
|
||||
<div class="navbar-brand-container mx-auto">
|
||||
<a href="../../index.html" class="navbar-brand navbar-brand-logo">
|
||||
<img src="../../image/axolotl_logo_digital_white.svg" alt="" class="navbar-logo">
|
||||
</a>
|
||||
</div>
|
||||
<div class="quarto-navbar-tools tools-wide tools-end">
|
||||
<a href="https://twitter.com/axolotl_ai" title="" class="quarto-navigation-tool px-1" aria-label=""><i class="bi bi-twitter"></i></a>
|
||||
<a href="https://github.com/axolotl-ai-cloud/axolotl/" title="" class="quarto-navigation-tool px-1" aria-label=""><i class="bi bi-github"></i></a>
|
||||
<a href="https://discord.gg/7m9sfhzaf3" title="" class="quarto-navigation-tool px-1" aria-label=""><i class="bi bi-discord"></i></a>
|
||||
</div>
|
||||
<div id="quarto-search" class="" title="Search"></div>
|
||||
</div> <!-- /container-fluid -->
|
||||
</nav>
|
||||
<nav class="quarto-secondary-nav">
|
||||
<div class="container-fluid d-flex">
|
||||
<button type="button" class="quarto-btn-toggle btn" data-bs-toggle="collapse" role="button" data-bs-target=".quarto-sidebar-collapse-item" aria-controls="quarto-sidebar" aria-expanded="false" aria-label="Toggle sidebar navigation" onclick="if (window.quartoToggleHeadroom) { window.quartoToggleHeadroom(); }">
|
||||
<i class="bi bi-layout-text-sidebar-reverse"></i>
|
||||
</button>
|
||||
<nav class="quarto-page-breadcrumbs" aria-label="breadcrumb"><ol class="breadcrumb"></ol></nav>
|
||||
<a class="flex-grow-1" role="navigation" data-bs-toggle="collapse" data-bs-target=".quarto-sidebar-collapse-item" aria-controls="quarto-sidebar" aria-expanded="false" aria-label="Toggle sidebar navigation" onclick="if (window.quartoToggleHeadroom) { window.quartoToggleHeadroom(); }">
|
||||
</a>
|
||||
</div>
|
||||
</nav>
|
||||
</header>
|
||||
<!-- content -->
|
||||
<div id="quarto-content" class="quarto-container page-columns page-rows-contents page-layout-article page-navbar">
|
||||
<!-- sidebar -->
|
||||
<nav id="quarto-sidebar" class="sidebar collapse collapse-horizontal quarto-sidebar-collapse-item sidebar-navigation docked overflow-auto">
|
||||
<div class="sidebar-menu-container">
|
||||
<ul class="list-unstyled mt-1">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../index.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Home</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-1" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Getting Started</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-1" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-1" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/getting-started.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Quickstart</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/installation.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Installation</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/inference.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Inference and Merging</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/cli.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Command Line Interface (CLI)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/config.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Config Reference</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/api" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">API Reference</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/index.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Dataset Formats</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-2" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-2" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/pretraining.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Pre-training</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/inst_tune.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Instruction Tuning</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/conversation.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Conversation</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/stepwise_supervised.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Stepwise Supervised Format</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/template_free.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Template-Free</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset-formats/tokenized.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Custom Pre-Tokenized Dataset</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-3" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Deployments</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-3" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-3" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/docker.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Docker</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multi-gpu.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Multi-GPU</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multi-node.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Multi Node</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/ray-integration.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Ray Train</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/amd_hpc.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">AMD GPUs on HPC Systems</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/mac.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Mac M-series</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-4" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">How To Guides</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-4" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-4" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multimodal.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">MultiModal / Vision Language Models (BETA)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/rlhf.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">RLHF (Beta)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/reward_modelling.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Reward Modelling</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/lr_groups.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Learning Rate Groups</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/lora_optims.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">LoRA Optimizations</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset_loading.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Dataset Loading</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-5" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Core Concepts</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-5" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-5" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/batch_vs_grad.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Batch size vs Gradient accumulation</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/dataset_preprocessing.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Dataset Preprocessing</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/multipack.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Multipack (Sample Packing)</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-6" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Advanced Features</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-6" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-6" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/fsdp_qlora.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">FDSP + QLoRA</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/unsloth.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Unsloth</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/torchao.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">PyTorch ao</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/custom_integrations.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Custom Integrations</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/sequence_parallelism.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Sequence Parallelism</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
<li class="sidebar-item sidebar-item-section">
|
||||
<div class="sidebar-item-container">
|
||||
<a class="sidebar-item-text sidebar-link text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-7" role="navigation" aria-expanded="true">
|
||||
<span class="menu-text">Troubleshooting</span></a>
|
||||
<a class="sidebar-item-toggle text-start" data-bs-toggle="collapse" data-bs-target="#quarto-sidebar-section-7" role="navigation" aria-expanded="true" aria-label="Toggle section">
|
||||
<i class="bi bi-chevron-right ms-2"></i>
|
||||
</a>
|
||||
</div>
|
||||
<ul id="quarto-sidebar-section-7" class="collapse list-unstyled sidebar-section depth1 show">
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/faq.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">FAQ</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/debugging.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">Debugging</span></a>
|
||||
</div>
|
||||
</li>
|
||||
<li class="sidebar-item">
|
||||
<div class="sidebar-item-container">
|
||||
<a href="../../docs/nccl.html" class="sidebar-item-text sidebar-link">
|
||||
<span class="menu-text">NCCL</span></a>
|
||||
</div>
|
||||
</li>
|
||||
</ul>
|
||||
</li>
|
||||
</ul>
|
||||
</div>
|
||||
</nav>
|
||||
<div id="quarto-sidebar-glass" class="quarto-sidebar-collapse-item" data-bs-toggle="collapse" data-bs-target=".quarto-sidebar-collapse-item"></div>
|
||||
<!-- margin-sidebar -->
|
||||
<div id="quarto-margin-sidebar" class="sidebar margin-sidebar">
|
||||
<nav id="TOC" role="doc-toc" class="toc-active">
|
||||
<h2 id="toc-title">On this page</h2>
|
||||
|
||||
<ul>
|
||||
<li><a href="#axolotl.loaders.processor" id="toc-axolotl.loaders.processor" class="nav-link active" data-scroll-target="#axolotl.loaders.processor">loaders.processor</a></li>
|
||||
</ul>
|
||||
</nav>
|
||||
</div>
|
||||
<!-- main -->
|
||||
<main class="content" id="quarto-document-content"><header id="title-block-header" class="quarto-title-block"></header>
|
||||
|
||||
|
||||
|
||||
|
||||
<section id="axolotl.loaders.processor" class="level1">
|
||||
<h1>loaders.processor</h1>
|
||||
<p><code>loaders.processor</code></p>
|
||||
<p>Processor loading functionality for multi-modal models</p>
|
||||
|
||||
|
||||
</section>
|
||||
|
||||
</main> <!-- /main -->
|
||||
<script id="quarto-html-after-body" type="application/javascript">
|
||||
window.document.addEventListener("DOMContentLoaded", function (event) {
|
||||
const icon = "";
|
||||
const anchorJS = new window.AnchorJS();
|
||||
anchorJS.options = {
|
||||
placement: 'right',
|
||||
icon: icon
|
||||
};
|
||||
anchorJS.add('.anchored');
|
||||
const isCodeAnnotation = (el) => {
|
||||
for (const clz of el.classList) {
|
||||
if (clz.startsWith('code-annotation-')) {
|
||||
return true;
|
||||
}
|
||||
}
|
||||
return false;
|
||||
}
|
||||
const onCopySuccess = function(e) {
|
||||
// button target
|
||||
const button = e.trigger;
|
||||
// don't keep focus
|
||||
button.blur();
|
||||
// flash "checked"
|
||||
button.classList.add('code-copy-button-checked');
|
||||
var currentTitle = button.getAttribute("title");
|
||||
button.setAttribute("title", "Copied!");
|
||||
let tooltip;
|
||||
if (window.bootstrap) {
|
||||
button.setAttribute("data-bs-toggle", "tooltip");
|
||||
button.setAttribute("data-bs-placement", "left");
|
||||
button.setAttribute("data-bs-title", "Copied!");
|
||||
tooltip = new bootstrap.Tooltip(button,
|
||||
{ trigger: "manual",
|
||||
customClass: "code-copy-button-tooltip",
|
||||
offset: [0, -8]});
|
||||
tooltip.show();
|
||||
}
|
||||
setTimeout(function() {
|
||||
if (tooltip) {
|
||||
tooltip.hide();
|
||||
button.removeAttribute("data-bs-title");
|
||||
button.removeAttribute("data-bs-toggle");
|
||||
button.removeAttribute("data-bs-placement");
|
||||
}
|
||||
button.setAttribute("title", currentTitle);
|
||||
button.classList.remove('code-copy-button-checked');
|
||||
}, 1000);
|
||||
// clear code selection
|
||||
e.clearSelection();
|
||||
}
|
||||
const getTextToCopy = function(trigger) {
|
||||
const codeEl = trigger.previousElementSibling.cloneNode(true);
|
||||
for (const childEl of codeEl.children) {
|
||||
if (isCodeAnnotation(childEl)) {
|
||||
childEl.remove();
|
||||
}
|
||||
}
|
||||
return codeEl.innerText;
|
||||
}
|
||||
const clipboard = new window.ClipboardJS('.code-copy-button:not([data-in-quarto-modal])', {
|
||||
text: getTextToCopy
|
||||
});
|
||||
clipboard.on('success', onCopySuccess);
|
||||
if (window.document.getElementById('quarto-embedded-source-code-modal')) {
|
||||
const clipboardModal = new window.ClipboardJS('.code-copy-button[data-in-quarto-modal]', {
|
||||
text: getTextToCopy,
|
||||
container: window.document.getElementById('quarto-embedded-source-code-modal')
|
||||
});
|
||||
clipboardModal.on('success', onCopySuccess);
|
||||
}
|
||||
var localhostRegex = new RegExp(/^(?:http|https):\/\/localhost\:?[0-9]*\//);
|
||||
var mailtoRegex = new RegExp(/^mailto:/);
|
||||
var filterRegex = new RegExp("https:\/\/docs\.axolotl\.ai");
|
||||
var isInternal = (href) => {
|
||||
return filterRegex.test(href) || localhostRegex.test(href) || mailtoRegex.test(href);
|
||||
}
|
||||
// Inspect non-navigation links and adorn them if external
|
||||
var links = window.document.querySelectorAll('a[href]:not(.nav-link):not(.navbar-brand):not(.toc-action):not(.sidebar-link):not(.sidebar-item-toggle):not(.pagination-link):not(.no-external):not([aria-hidden]):not(.dropdown-item):not(.quarto-navigation-tool):not(.about-link)');
|
||||
for (var i=0; i<links.length; i++) {
|
||||
const link = links[i];
|
||||
if (!isInternal(link.href)) {
|
||||
// undo the damage that might have been done by quarto-nav.js in the case of
|
||||
// links that we want to consider external
|
||||
if (link.dataset.originalHref !== undefined) {
|
||||
link.href = link.dataset.originalHref;
|
||||
}
|
||||
}
|
||||
}
|
||||
function tippyHover(el, contentFn, onTriggerFn, onUntriggerFn) {
|
||||
const config = {
|
||||
allowHTML: true,
|
||||
maxWidth: 500,
|
||||
delay: 100,
|
||||
arrow: false,
|
||||
appendTo: function(el) {
|
||||
return el.parentElement;
|
||||
},
|
||||
interactive: true,
|
||||
interactiveBorder: 10,
|
||||
theme: 'quarto',
|
||||
placement: 'bottom-start',
|
||||
};
|
||||
if (contentFn) {
|
||||
config.content = contentFn;
|
||||
}
|
||||
if (onTriggerFn) {
|
||||
config.onTrigger = onTriggerFn;
|
||||
}
|
||||
if (onUntriggerFn) {
|
||||
config.onUntrigger = onUntriggerFn;
|
||||
}
|
||||
window.tippy(el, config);
|
||||
}
|
||||
const noterefs = window.document.querySelectorAll('a[role="doc-noteref"]');
|
||||
for (var i=0; i<noterefs.length; i++) {
|
||||
const ref = noterefs[i];
|
||||
tippyHover(ref, function() {
|
||||
// use id or data attribute instead here
|
||||
let href = ref.getAttribute('data-footnote-href') || ref.getAttribute('href');
|
||||
try { href = new URL(href).hash; } catch {}
|
||||
const id = href.replace(/^#\/?/, "");
|
||||
const note = window.document.getElementById(id);
|
||||
if (note) {
|
||||
return note.innerHTML;
|
||||
} else {
|
||||
return "";
|
||||
}
|
||||
});
|
||||
}
|
||||
const xrefs = window.document.querySelectorAll('a.quarto-xref');
|
||||
const processXRef = (id, note) => {
|
||||
// Strip column container classes
|
||||
const stripColumnClz = (el) => {
|
||||
el.classList.remove("page-full", "page-columns");
|
||||
if (el.children) {
|
||||
for (const child of el.children) {
|
||||
stripColumnClz(child);
|
||||
}
|
||||
}
|
||||
}
|
||||
stripColumnClz(note)
|
||||
if (id === null || id.startsWith('sec-')) {
|
||||
// Special case sections, only their first couple elements
|
||||
const container = document.createElement("div");
|
||||
if (note.children && note.children.length > 2) {
|
||||
container.appendChild(note.children[0].cloneNode(true));
|
||||
for (let i = 1; i < note.children.length; i++) {
|
||||
const child = note.children[i];
|
||||
if (child.tagName === "P" && child.innerText === "") {
|
||||
continue;
|
||||
} else {
|
||||
container.appendChild(child.cloneNode(true));
|
||||
break;
|
||||
}
|
||||
}
|
||||
if (window.Quarto?.typesetMath) {
|
||||
window.Quarto.typesetMath(container);
|
||||
}
|
||||
return container.innerHTML
|
||||
} else {
|
||||
if (window.Quarto?.typesetMath) {
|
||||
window.Quarto.typesetMath(note);
|
||||
}
|
||||
return note.innerHTML;
|
||||
}
|
||||
} else {
|
||||
// Remove any anchor links if they are present
|
||||
const anchorLink = note.querySelector('a.anchorjs-link');
|
||||
if (anchorLink) {
|
||||
anchorLink.remove();
|
||||
}
|
||||
if (window.Quarto?.typesetMath) {
|
||||
window.Quarto.typesetMath(note);
|
||||
}
|
||||
if (note.classList.contains("callout")) {
|
||||
return note.outerHTML;
|
||||
} else {
|
||||
return note.innerHTML;
|
||||
}
|
||||
}
|
||||
}
|
||||
for (var i=0; i<xrefs.length; i++) {
|
||||
const xref = xrefs[i];
|
||||
tippyHover(xref, undefined, function(instance) {
|
||||
instance.disable();
|
||||
let url = xref.getAttribute('href');
|
||||
let hash = undefined;
|
||||
if (url.startsWith('#')) {
|
||||
hash = url;
|
||||
} else {
|
||||
try { hash = new URL(url).hash; } catch {}
|
||||
}
|
||||
if (hash) {
|
||||
const id = hash.replace(/^#\/?/, "");
|
||||
const note = window.document.getElementById(id);
|
||||
if (note !== null) {
|
||||
try {
|
||||
const html = processXRef(id, note.cloneNode(true));
|
||||
instance.setContent(html);
|
||||
} finally {
|
||||
instance.enable();
|
||||
instance.show();
|
||||
}
|
||||
} else {
|
||||
// See if we can fetch this
|
||||
fetch(url.split('#')[0])
|
||||
.then(res => res.text())
|
||||
.then(html => {
|
||||
const parser = new DOMParser();
|
||||
const htmlDoc = parser.parseFromString(html, "text/html");
|
||||
const note = htmlDoc.getElementById(id);
|
||||
if (note !== null) {
|
||||
const html = processXRef(id, note);
|
||||
instance.setContent(html);
|
||||
}
|
||||
}).finally(() => {
|
||||
instance.enable();
|
||||
instance.show();
|
||||
});
|
||||
}
|
||||
} else {
|
||||
// See if we can fetch a full url (with no hash to target)
|
||||
// This is a special case and we should probably do some content thinning / targeting
|
||||
fetch(url)
|
||||
.then(res => res.text())
|
||||
.then(html => {
|
||||
const parser = new DOMParser();
|
||||
const htmlDoc = parser.parseFromString(html, "text/html");
|
||||
const note = htmlDoc.querySelector('main.content');
|
||||
if (note !== null) {
|
||||
// This should only happen for chapter cross references
|
||||
// (since there is no id in the URL)
|
||||
// remove the first header
|
||||
if (note.children.length > 0 && note.children[0].tagName === "HEADER") {
|
||||
note.children[0].remove();
|
||||
}
|
||||
const html = processXRef(null, note);
|
||||
instance.setContent(html);
|
||||
}
|
||||
}).finally(() => {
|
||||
instance.enable();
|
||||
instance.show();
|
||||
});
|
||||
}
|
||||
}, function(instance) {
|
||||
});
|
||||
}
|
||||
let selectedAnnoteEl;
|
||||
const selectorForAnnotation = ( cell, annotation) => {
|
||||
let cellAttr = 'data-code-cell="' + cell + '"';
|
||||
let lineAttr = 'data-code-annotation="' + annotation + '"';
|
||||
const selector = 'span[' + cellAttr + '][' + lineAttr + ']';
|
||||
return selector;
|
||||
}
|
||||
const selectCodeLines = (annoteEl) => {
|
||||
const doc = window.document;
|
||||
const targetCell = annoteEl.getAttribute("data-target-cell");
|
||||
const targetAnnotation = annoteEl.getAttribute("data-target-annotation");
|
||||
const annoteSpan = window.document.querySelector(selectorForAnnotation(targetCell, targetAnnotation));
|
||||
const lines = annoteSpan.getAttribute("data-code-lines").split(",");
|
||||
const lineIds = lines.map((line) => {
|
||||
return targetCell + "-" + line;
|
||||
})
|
||||
let top = null;
|
||||
let height = null;
|
||||
let parent = null;
|
||||
if (lineIds.length > 0) {
|
||||
//compute the position of the single el (top and bottom and make a div)
|
||||
const el = window.document.getElementById(lineIds[0]);
|
||||
top = el.offsetTop;
|
||||
height = el.offsetHeight;
|
||||
parent = el.parentElement.parentElement;
|
||||
if (lineIds.length > 1) {
|
||||
const lastEl = window.document.getElementById(lineIds[lineIds.length - 1]);
|
||||
const bottom = lastEl.offsetTop + lastEl.offsetHeight;
|
||||
height = bottom - top;
|
||||
}
|
||||
if (top !== null && height !== null && parent !== null) {
|
||||
// cook up a div (if necessary) and position it
|
||||
let div = window.document.getElementById("code-annotation-line-highlight");
|
||||
if (div === null) {
|
||||
div = window.document.createElement("div");
|
||||
div.setAttribute("id", "code-annotation-line-highlight");
|
||||
div.style.position = 'absolute';
|
||||
parent.appendChild(div);
|
||||
}
|
||||
div.style.top = top - 2 + "px";
|
||||
div.style.height = height + 4 + "px";
|
||||
div.style.left = 0;
|
||||
let gutterDiv = window.document.getElementById("code-annotation-line-highlight-gutter");
|
||||
if (gutterDiv === null) {
|
||||
gutterDiv = window.document.createElement("div");
|
||||
gutterDiv.setAttribute("id", "code-annotation-line-highlight-gutter");
|
||||
gutterDiv.style.position = 'absolute';
|
||||
const codeCell = window.document.getElementById(targetCell);
|
||||
const gutter = codeCell.querySelector('.code-annotation-gutter');
|
||||
gutter.appendChild(gutterDiv);
|
||||
}
|
||||
gutterDiv.style.top = top - 2 + "px";
|
||||
gutterDiv.style.height = height + 4 + "px";
|
||||
}
|
||||
selectedAnnoteEl = annoteEl;
|
||||
}
|
||||
};
|
||||
const unselectCodeLines = () => {
|
||||
const elementsIds = ["code-annotation-line-highlight", "code-annotation-line-highlight-gutter"];
|
||||
elementsIds.forEach((elId) => {
|
||||
const div = window.document.getElementById(elId);
|
||||
if (div) {
|
||||
div.remove();
|
||||
}
|
||||
});
|
||||
selectedAnnoteEl = undefined;
|
||||
};
|
||||
// Handle positioning of the toggle
|
||||
window.addEventListener(
|
||||
"resize",
|
||||
throttle(() => {
|
||||
elRect = undefined;
|
||||
if (selectedAnnoteEl) {
|
||||
selectCodeLines(selectedAnnoteEl);
|
||||
}
|
||||
}, 10)
|
||||
);
|
||||
function throttle(fn, ms) {
|
||||
let throttle = false;
|
||||
let timer;
|
||||
return (...args) => {
|
||||
if(!throttle) { // first call gets through
|
||||
fn.apply(this, args);
|
||||
throttle = true;
|
||||
} else { // all the others get throttled
|
||||
if(timer) clearTimeout(timer); // cancel #2
|
||||
timer = setTimeout(() => {
|
||||
fn.apply(this, args);
|
||||
timer = throttle = false;
|
||||
}, ms);
|
||||
}
|
||||
};
|
||||
}
|
||||
// Attach click handler to the DT
|
||||
const annoteDls = window.document.querySelectorAll('dt[data-target-cell]');
|
||||
for (const annoteDlNode of annoteDls) {
|
||||
annoteDlNode.addEventListener('click', (event) => {
|
||||
const clickedEl = event.target;
|
||||
if (clickedEl !== selectedAnnoteEl) {
|
||||
unselectCodeLines();
|
||||
const activeEl = window.document.querySelector('dt[data-target-cell].code-annotation-active');
|
||||
if (activeEl) {
|
||||
activeEl.classList.remove('code-annotation-active');
|
||||
}
|
||||
selectCodeLines(clickedEl);
|
||||
clickedEl.classList.add('code-annotation-active');
|
||||
} else {
|
||||
// Unselect the line
|
||||
unselectCodeLines();
|
||||
clickedEl.classList.remove('code-annotation-active');
|
||||
}
|
||||
});
|
||||
}
|
||||
const findCites = (el) => {
|
||||
const parentEl = el.parentElement;
|
||||
if (parentEl) {
|
||||
const cites = parentEl.dataset.cites;
|
||||
if (cites) {
|
||||
return {
|
||||
el,
|
||||
cites: cites.split(' ')
|
||||
};
|
||||
} else {
|
||||
return findCites(el.parentElement)
|
||||
}
|
||||
} else {
|
||||
return undefined;
|
||||
}
|
||||
};
|
||||
var bibliorefs = window.document.querySelectorAll('a[role="doc-biblioref"]');
|
||||
for (var i=0; i<bibliorefs.length; i++) {
|
||||
const ref = bibliorefs[i];
|
||||
const citeInfo = findCites(ref);
|
||||
if (citeInfo) {
|
||||
tippyHover(citeInfo.el, function() {
|
||||
var popup = window.document.createElement('div');
|
||||
citeInfo.cites.forEach(function(cite) {
|
||||
var citeDiv = window.document.createElement('div');
|
||||
citeDiv.classList.add('hanging-indent');
|
||||
citeDiv.classList.add('csl-entry');
|
||||
var biblioDiv = window.document.getElementById('ref-' + cite);
|
||||
if (biblioDiv) {
|
||||
citeDiv.innerHTML = biblioDiv.innerHTML;
|
||||
}
|
||||
popup.appendChild(citeDiv);
|
||||
});
|
||||
return popup.innerHTML;
|
||||
});
|
||||
}
|
||||
}
|
||||
});
|
||||
</script>
|
||||
</div> <!-- /content -->
|
||||
|
||||
|
||||
|
||||
|
||||
</body></html>
|
||||
@@ -7,7 +7,7 @@
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
|
||||
|
||||
|
||||
<title>utils.models – Axolotl</title>
|
||||
<title>loaders.tokenizer – Axolotl</title>
|
||||
<style>
|
||||
code{white-space: pre-wrap;}
|
||||
span.smallcaps{font-variant: small-caps;}
|
||||
@@ -446,20 +446,12 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
|
||||
<h2 id="toc-title">On this page</h2>
|
||||
|
||||
<ul>
|
||||
<li><a href="#axolotl.utils.models" id="toc-axolotl.utils.models" class="nav-link active" data-scroll-target="#axolotl.utils.models">utils.models</a>
|
||||
<li><a href="#axolotl.loaders.tokenizer" id="toc-axolotl.loaders.tokenizer" class="nav-link active" data-scroll-target="#axolotl.loaders.tokenizer">loaders.tokenizer</a>
|
||||
<ul class="collapse">
|
||||
<li><a href="#classes" id="toc-classes" class="nav-link" data-scroll-target="#classes">Classes</a>
|
||||
<ul class="collapse">
|
||||
<li><a href="#axolotl.utils.models.ModelLoader" id="toc-axolotl.utils.models.ModelLoader" class="nav-link" data-scroll-target="#axolotl.utils.models.ModelLoader">ModelLoader</a></li>
|
||||
</ul></li>
|
||||
<li><a href="#functions" id="toc-functions" class="nav-link" data-scroll-target="#functions">Functions</a>
|
||||
<ul class="collapse">
|
||||
<li><a href="#axolotl.utils.models.get_module_class_from_name" id="toc-axolotl.utils.models.get_module_class_from_name" class="nav-link" data-scroll-target="#axolotl.utils.models.get_module_class_from_name">get_module_class_from_name</a></li>
|
||||
<li><a href="#axolotl.utils.models.load_model" id="toc-axolotl.utils.models.load_model" class="nav-link" data-scroll-target="#axolotl.utils.models.load_model">load_model</a></li>
|
||||
<li><a href="#axolotl.utils.models.load_tokenizer" id="toc-axolotl.utils.models.load_tokenizer" class="nav-link" data-scroll-target="#axolotl.utils.models.load_tokenizer">load_tokenizer</a></li>
|
||||
<li><a href="#axolotl.utils.models.modify_tokenizer_files" id="toc-axolotl.utils.models.modify_tokenizer_files" class="nav-link" data-scroll-target="#axolotl.utils.models.modify_tokenizer_files">modify_tokenizer_files</a></li>
|
||||
<li><a href="#axolotl.utils.models.setup_quantized_meta_for_peft" id="toc-axolotl.utils.models.setup_quantized_meta_for_peft" class="nav-link" data-scroll-target="#axolotl.utils.models.setup_quantized_meta_for_peft">setup_quantized_meta_for_peft</a></li>
|
||||
<li><a href="#axolotl.utils.models.setup_quantized_peft_meta_for_training" id="toc-axolotl.utils.models.setup_quantized_peft_meta_for_training" class="nav-link" data-scroll-target="#axolotl.utils.models.setup_quantized_peft_meta_for_training">setup_quantized_peft_meta_for_training</a></li>
|
||||
<li><a href="#axolotl.loaders.tokenizer.load_tokenizer" id="toc-axolotl.loaders.tokenizer.load_tokenizer" class="nav-link" data-scroll-target="#axolotl.loaders.tokenizer.load_tokenizer">load_tokenizer</a></li>
|
||||
<li><a href="#axolotl.loaders.tokenizer.modify_tokenizer_files" id="toc-axolotl.loaders.tokenizer.modify_tokenizer_files" class="nav-link" data-scroll-target="#axolotl.loaders.tokenizer.modify_tokenizer_files">modify_tokenizer_files</a></li>
|
||||
</ul></li>
|
||||
</ul></li>
|
||||
</ul>
|
||||
@@ -471,109 +463,10 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
|
||||
|
||||
|
||||
|
||||
<section id="axolotl.utils.models" class="level1">
|
||||
<h1>utils.models</h1>
|
||||
<p><code>utils.models</code></p>
|
||||
<p>Module for models and model loading</p>
|
||||
<section id="classes" class="level2">
|
||||
<h2 class="anchored" data-anchor-id="classes">Classes</h2>
|
||||
<table class="caption-top table">
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Description</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.models.ModelLoader">ModelLoader</a></td>
|
||||
<td>ModelLoader: managing all the config and monkey patches while loading model</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
<section id="axolotl.utils.models.ModelLoader" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.models.ModelLoader">ModelLoader</h3>
|
||||
<div class="sourceCode" id="cb1"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a>utils.models.ModelLoader(</span>
|
||||
<span id="cb1-2"><a href="#cb1-2" aria-hidden="true" tabindex="-1"></a> <span class="va">self</span>,</span>
|
||||
<span id="cb1-3"><a href="#cb1-3" aria-hidden="true" tabindex="-1"></a> cfg,</span>
|
||||
<span id="cb1-4"><a href="#cb1-4" aria-hidden="true" tabindex="-1"></a> tokenizer,</span>
|
||||
<span id="cb1-5"><a href="#cb1-5" aria-hidden="true" tabindex="-1"></a> <span class="op">*</span>,</span>
|
||||
<span id="cb1-6"><a href="#cb1-6" aria-hidden="true" tabindex="-1"></a> processor<span class="op">=</span><span class="va">None</span>,</span>
|
||||
<span id="cb1-7"><a href="#cb1-7" aria-hidden="true" tabindex="-1"></a> inference<span class="op">=</span><span class="va">False</span>,</span>
|
||||
<span id="cb1-8"><a href="#cb1-8" aria-hidden="true" tabindex="-1"></a> reference_model<span class="op">=</span><span class="va">False</span>,</span>
|
||||
<span id="cb1-9"><a href="#cb1-9" aria-hidden="true" tabindex="-1"></a> <span class="op">**</span>kwargs,</span>
|
||||
<span id="cb1-10"><a href="#cb1-10" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>ModelLoader: managing all the config and monkey patches while loading model</p>
|
||||
<section id="attributes" class="level4">
|
||||
<h4 class="anchored" data-anchor-id="attributes">Attributes</h4>
|
||||
<table class="caption-top table">
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Description</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.models.ModelLoader.has_flash_attn">has_flash_attn</a></td>
|
||||
<td>Check if flash attention is installed</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
<section id="methods" class="level4">
|
||||
<h4 class="anchored" data-anchor-id="methods">Methods</h4>
|
||||
<table class="caption-top table">
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Description</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.models.ModelLoader.patch_llama_derived_model">patch_llama_derived_model</a></td>
|
||||
<td>Modify all llama derived models in one block</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.models.ModelLoader.patch_loss_llama">patch_loss_llama</a></td>
|
||||
<td>Patch loss functions and other optimizations</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.models.ModelLoader.set_attention_config">set_attention_config</a></td>
|
||||
<td>sample packing uses custom FA2 patch</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.models.ModelLoader.set_auto_model_loader">set_auto_model_loader</a></td>
|
||||
<td>Set self.auto_model_loader. Defaults to <code>transformers.AutoModelForCausalLM</code></td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
<section id="axolotl.utils.models.ModelLoader.patch_llama_derived_model" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.models.ModelLoader.patch_llama_derived_model">patch_llama_derived_model</h5>
|
||||
<div class="sourceCode" id="cb2"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a>utils.models.ModelLoader.patch_llama_derived_model()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Modify all llama derived models in one block</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.models.ModelLoader.patch_loss_llama" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.models.ModelLoader.patch_loss_llama">patch_loss_llama</h5>
|
||||
<div class="sourceCode" id="cb3"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a>utils.models.ModelLoader.patch_loss_llama()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Patch loss functions and other optimizations</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.models.ModelLoader.set_attention_config" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.models.ModelLoader.set_attention_config">set_attention_config</h5>
|
||||
<div class="sourceCode" id="cb4"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a>utils.models.ModelLoader.set_attention_config()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>sample packing uses custom FA2 patch</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.models.ModelLoader.set_auto_model_loader" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.models.ModelLoader.set_auto_model_loader">set_auto_model_loader</h5>
|
||||
<div class="sourceCode" id="cb5"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a>utils.models.ModelLoader.set_auto_model_loader()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Set self.auto_model_loader. Defaults to <code>transformers.AutoModelForCausalLM</code>
|
||||
(set at <code>__init__</code>). When using a multimodal model, <code>self.auto_model_loader</code>
|
||||
should be set according to the type of the model.</p>
|
||||
</section>
|
||||
</section>
|
||||
</section>
|
||||
</section>
|
||||
<section id="axolotl.loaders.tokenizer" class="level1">
|
||||
<h1>loaders.tokenizer</h1>
|
||||
<p><code>loaders.tokenizer</code></p>
|
||||
<p>Tokenizer loading functionality and associated utils</p>
|
||||
<section id="functions" class="level2">
|
||||
<h2 class="anchored" data-anchor-id="functions">Functions</h2>
|
||||
<table class="caption-top table">
|
||||
@@ -585,94 +478,33 @@ should be set according to the type of the model.</p>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.models.get_module_class_from_name">get_module_class_from_name</a></td>
|
||||
<td>Gets a class from a module by its name.</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.models.load_model">load_model</a></td>
|
||||
<td>Load a model for a given configuration and tokenizer.</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.models.load_tokenizer">load_tokenizer</a></td>
|
||||
<td><a href="#axolotl.loaders.tokenizer.load_tokenizer">load_tokenizer</a></td>
|
||||
<td>Load and configure the tokenizer based on the provided config.</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.models.modify_tokenizer_files">modify_tokenizer_files</a></td>
|
||||
<td>Modify tokenizer files to replace added_tokens strings, save to output directory, and return the path to the modified tokenizer.</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.models.setup_quantized_meta_for_peft">setup_quantized_meta_for_peft</a></td>
|
||||
<td>Replaces <code>quant_state.to</code> with a dummy function to prevent PEFT from moving <code>quant_state</code> to meta device</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.models.setup_quantized_peft_meta_for_training">setup_quantized_peft_meta_for_training</a></td>
|
||||
<td>Replaces dummy <code>quant_state.to</code> method with the original function to allow training to continue</td>
|
||||
<td><a href="#axolotl.loaders.tokenizer.modify_tokenizer_files">modify_tokenizer_files</a></td>
|
||||
<td>Modify tokenizer files to replace added_tokens strings, save to output directory,</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
<section id="axolotl.utils.models.get_module_class_from_name" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.models.get_module_class_from_name">get_module_class_from_name</h3>
|
||||
<div class="sourceCode" id="cb6"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb6-1"><a href="#cb6-1" aria-hidden="true" tabindex="-1"></a>utils.models.get_module_class_from_name(module, name)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Gets a class from a module by its name.</p>
|
||||
<section id="parameters" class="level4 doc-section doc-section-parameters">
|
||||
<h4 class="doc-section doc-section-parameters anchored" data-anchor-id="parameters">Parameters</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 10%">
|
||||
<col style="width: 25%">
|
||||
<col style="width: 47%">
|
||||
<col style="width: 16%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Type</th>
|
||||
<th>Description</th>
|
||||
<th>Default</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td>module</td>
|
||||
<td><code>torch.nn.Module</code></td>
|
||||
<td>The module to get the class from.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>name</td>
|
||||
<td><code>str</code></td>
|
||||
<td>The name of the class.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
</section>
|
||||
<section id="axolotl.utils.models.load_model" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.models.load_model">load_model</h3>
|
||||
<div class="sourceCode" id="cb7"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb7-1"><a href="#cb7-1" aria-hidden="true" tabindex="-1"></a>utils.models.load_model(</span>
|
||||
<span id="cb7-2"><a href="#cb7-2" aria-hidden="true" tabindex="-1"></a> cfg,</span>
|
||||
<span id="cb7-3"><a href="#cb7-3" aria-hidden="true" tabindex="-1"></a> tokenizer,</span>
|
||||
<span id="cb7-4"><a href="#cb7-4" aria-hidden="true" tabindex="-1"></a> <span class="op">*</span>,</span>
|
||||
<span id="cb7-5"><a href="#cb7-5" aria-hidden="true" tabindex="-1"></a> processor<span class="op">=</span><span class="va">None</span>,</span>
|
||||
<span id="cb7-6"><a href="#cb7-6" aria-hidden="true" tabindex="-1"></a> inference<span class="op">=</span><span class="va">False</span>,</span>
|
||||
<span id="cb7-7"><a href="#cb7-7" aria-hidden="true" tabindex="-1"></a> reference_model<span class="op">=</span><span class="va">False</span>,</span>
|
||||
<span id="cb7-8"><a href="#cb7-8" aria-hidden="true" tabindex="-1"></a> <span class="op">**</span>kwargs,</span>
|
||||
<span id="cb7-9"><a href="#cb7-9" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Load a model for a given configuration and tokenizer.</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.models.load_tokenizer" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.models.load_tokenizer">load_tokenizer</h3>
|
||||
<div class="sourceCode" id="cb8"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb8-1"><a href="#cb8-1" aria-hidden="true" tabindex="-1"></a>utils.models.load_tokenizer(cfg)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<section id="axolotl.loaders.tokenizer.load_tokenizer" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.loaders.tokenizer.load_tokenizer">load_tokenizer</h3>
|
||||
<div class="sourceCode" id="cb1"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a>loaders.tokenizer.load_tokenizer(cfg)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Load and configure the tokenizer based on the provided config.</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.models.modify_tokenizer_files" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.models.modify_tokenizer_files">modify_tokenizer_files</h3>
|
||||
<div class="sourceCode" id="cb9"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a>utils.models.modify_tokenizer_files(tokenizer_path, token_mappings, output_dir)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Modify tokenizer files to replace added_tokens strings, save to output directory, and return the path to the modified tokenizer.</p>
|
||||
<p>This only works with reserved tokens that were added to the tokenizer, not tokens already part of the vocab.</p>
|
||||
<section id="parameters-1" class="level4 doc-section doc-section-parameters">
|
||||
<h4 class="doc-section doc-section-parameters anchored" data-anchor-id="parameters-1">Parameters</h4>
|
||||
<section id="axolotl.loaders.tokenizer.modify_tokenizer_files" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.loaders.tokenizer.modify_tokenizer_files">modify_tokenizer_files</h3>
|
||||
<div class="sourceCode" id="cb2"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a>loaders.tokenizer.modify_tokenizer_files(</span>
|
||||
<span id="cb2-2"><a href="#cb2-2" aria-hidden="true" tabindex="-1"></a> tokenizer_path,</span>
|
||||
<span id="cb2-3"><a href="#cb2-3" aria-hidden="true" tabindex="-1"></a> token_mappings,</span>
|
||||
<span id="cb2-4"><a href="#cb2-4" aria-hidden="true" tabindex="-1"></a> output_dir,</span>
|
||||
<span id="cb2-5"><a href="#cb2-5" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Modify tokenizer files to replace added_tokens strings, save to output directory,
|
||||
and return the path to the modified tokenizer.</p>
|
||||
<p>This only works with reserved tokens that were added to the tokenizer, not tokens
|
||||
already part of the vocab.</p>
|
||||
<section id="parameters" class="level4 doc-section doc-section-parameters">
|
||||
<h4 class="doc-section doc-section-parameters anchored" data-anchor-id="parameters">Parameters</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 16%">
|
||||
@@ -697,7 +529,7 @@ should be set according to the type of the model.</p>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>token_mappings</td>
|
||||
<td>Dict[int, str]</td>
|
||||
<td>dict[int, str]</td>
|
||||
<td>Dict mapping {token_id (int): new_token_string}</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
@@ -729,22 +561,12 @@ should be set according to the type of the model.</p>
|
||||
</tbody>
|
||||
</table>
|
||||
<p>Ref: https://github.com/huggingface/transformers/issues/27974#issuecomment-1854188941</p>
|
||||
</section>
|
||||
</section>
|
||||
<section id="axolotl.utils.models.setup_quantized_meta_for_peft" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.models.setup_quantized_meta_for_peft">setup_quantized_meta_for_peft</h3>
|
||||
<div class="sourceCode" id="cb10"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb10-1"><a href="#cb10-1" aria-hidden="true" tabindex="-1"></a>utils.models.setup_quantized_meta_for_peft(model)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Replaces <code>quant_state.to</code> with a dummy function to prevent PEFT from moving <code>quant_state</code> to meta device</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.models.setup_quantized_peft_meta_for_training" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.models.setup_quantized_peft_meta_for_training">setup_quantized_peft_meta_for_training</h3>
|
||||
<div class="sourceCode" id="cb11"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb11-1"><a href="#cb11-1" aria-hidden="true" tabindex="-1"></a>utils.models.setup_quantized_peft_meta_for_training(model)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Replaces dummy <code>quant_state.to</code> method with the original function to allow training to continue</p>
|
||||
|
||||
|
||||
</section>
|
||||
</section>
|
||||
</section>
|
||||
</section>
|
||||
|
||||
</main> <!-- /main -->
|
||||
<script id="quarto-html-after-body" type="application/javascript">
|
||||
@@ -7,7 +7,7 @@
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
|
||||
|
||||
|
||||
<title>utils.gradient_checkpointing.offload_cpu – Axolotl</title>
|
||||
<title>monkeypatch.gradient_checkpointing.offload_cpu – Axolotl</title>
|
||||
<style>
|
||||
code{white-space: pre-wrap;}
|
||||
span.smallcaps{font-variant: small-caps;}
|
||||
@@ -446,11 +446,11 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
|
||||
<h2 id="toc-title">On this page</h2>
|
||||
|
||||
<ul>
|
||||
<li><a href="#axolotl.utils.gradient_checkpointing.offload_cpu" id="toc-axolotl.utils.gradient_checkpointing.offload_cpu" class="nav-link active" data-scroll-target="#axolotl.utils.gradient_checkpointing.offload_cpu">utils.gradient_checkpointing.offload_cpu</a>
|
||||
<li><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_cpu" id="toc-axolotl.monkeypatch.gradient_checkpointing.offload_cpu" class="nav-link active" data-scroll-target="#axolotl.monkeypatch.gradient_checkpointing.offload_cpu">monkeypatch.gradient_checkpointing.offload_cpu</a>
|
||||
<ul class="collapse">
|
||||
<li><a href="#classes" id="toc-classes" class="nav-link" data-scroll-target="#classes">Classes</a>
|
||||
<ul class="collapse">
|
||||
<li><a href="#axolotl.utils.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer" id="toc-axolotl.utils.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer" class="nav-link" data-scroll-target="#axolotl.utils.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer">CPU_Offloaded_Gradient_Checkpointer</a></li>
|
||||
<li><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer" id="toc-axolotl.monkeypatch.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer" class="nav-link" data-scroll-target="#axolotl.monkeypatch.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer">CPU_Offloaded_Gradient_Checkpointer</a></li>
|
||||
</ul></li>
|
||||
</ul></li>
|
||||
</ul>
|
||||
@@ -462,9 +462,9 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
|
||||
|
||||
|
||||
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_cpu" class="level1">
|
||||
<h1>utils.gradient_checkpointing.offload_cpu</h1>
|
||||
<p><code>utils.gradient_checkpointing.offload_cpu</code></p>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_cpu" class="level1">
|
||||
<h1>monkeypatch.gradient_checkpointing.offload_cpu</h1>
|
||||
<p><code>monkeypatch.gradient_checkpointing.offload_cpu</code></p>
|
||||
<p>CPU offloaded checkpointing</p>
|
||||
<section id="classes" class="level2">
|
||||
<h2 class="anchored" data-anchor-id="classes">Classes</h2>
|
||||
@@ -477,14 +477,15 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer">CPU_Offloaded_Gradient_Checkpointer</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer">CPU_Offloaded_Gradient_Checkpointer</a></td>
|
||||
<td>Saves VRAM by smartly offloading to RAM.</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer">CPU_Offloaded_Gradient_Checkpointer</h3>
|
||||
<div class="sourceCode" id="cb1"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer">CPU_Offloaded_Gradient_Checkpointer</h3>
|
||||
<div class="sourceCode" id="cb1"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_cpu.CPU_Offloaded_Gradient_Checkpointer(</span>
|
||||
<span id="cb1-2"><a href="#cb1-2" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Saves VRAM by smartly offloading to RAM.
|
||||
Tiny hit to performance, since we mask the movement via non blocking calls.</p>
|
||||
|
||||
@@ -7,7 +7,7 @@
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes">
|
||||
|
||||
|
||||
<title>utils.gradient_checkpointing.offload_disk – Axolotl</title>
|
||||
<title>monkeypatch.gradient_checkpointing.offload_disk – Axolotl</title>
|
||||
<style>
|
||||
code{white-space: pre-wrap;}
|
||||
span.smallcaps{font-variant: small-caps;}
|
||||
@@ -446,12 +446,12 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
|
||||
<h2 id="toc-title">On this page</h2>
|
||||
|
||||
<ul>
|
||||
<li><a href="#axolotl.utils.gradient_checkpointing.offload_disk" id="toc-axolotl.utils.gradient_checkpointing.offload_disk" class="nav-link active" data-scroll-target="#axolotl.utils.gradient_checkpointing.offload_disk">utils.gradient_checkpointing.offload_disk</a>
|
||||
<li><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk" id="toc-axolotl.monkeypatch.gradient_checkpointing.offload_disk" class="nav-link active" data-scroll-target="#axolotl.monkeypatch.gradient_checkpointing.offload_disk">monkeypatch.gradient_checkpointing.offload_disk</a>
|
||||
<ul class="collapse">
|
||||
<li><a href="#classes" id="toc-classes" class="nav-link" data-scroll-target="#classes">Classes</a>
|
||||
<ul class="collapse">
|
||||
<li><a href="#axolotl.utils.gradient_checkpointing.offload_disk.Disco" id="toc-axolotl.utils.gradient_checkpointing.offload_disk.Disco" class="nav-link" data-scroll-target="#axolotl.utils.gradient_checkpointing.offload_disk.Disco">Disco</a></li>
|
||||
<li><a href="#axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager" id="toc-axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager" class="nav-link" data-scroll-target="#axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager">DiskOffloadManager</a></li>
|
||||
<li><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco" id="toc-axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco" class="nav-link" data-scroll-target="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco">Disco</a></li>
|
||||
<li><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager" id="toc-axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager" class="nav-link" data-scroll-target="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager">DiskOffloadManager</a></li>
|
||||
</ul></li>
|
||||
</ul></li>
|
||||
</ul>
|
||||
@@ -463,9 +463,9 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
|
||||
|
||||
|
||||
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk" class="level1">
|
||||
<h1>utils.gradient_checkpointing.offload_disk</h1>
|
||||
<p><code>utils.gradient_checkpointing.offload_disk</code></p>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk" class="level1">
|
||||
<h1>monkeypatch.gradient_checkpointing.offload_disk</h1>
|
||||
<p><code>monkeypatch.gradient_checkpointing.offload_disk</code></p>
|
||||
<p>DISCO - DIsk-based Storage and Checkpointing with Optimized prefetching</p>
|
||||
<section id="classes" class="level2">
|
||||
<h2 class="anchored" data-anchor-id="classes">Classes</h2>
|
||||
@@ -478,18 +478,18 @@ pre > code.sourceCode > span > a:first-child::before { text-decoration: underlin
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_disk.Disco">Disco</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco">Disco</a></td>
|
||||
<td>Disco: DIsk-based Storage and Checkpointing with Optimized prefetching</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager">DiskOffloadManager</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager">DiskOffloadManager</a></td>
|
||||
<td>Manages offloaded tensors and handles prefetching in a separate thread.</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk.Disco" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_disk.Disco">Disco</h3>
|
||||
<div class="sourceCode" id="cb1"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_disk.Disco()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco">Disco</h3>
|
||||
<div class="sourceCode" id="cb1"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb1-1"><a href="#cb1-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_disk.Disco()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Disco: DIsk-based Storage and Checkpointing with Optimized prefetching
|
||||
Advanced disk-based gradient checkpointer with prefetching.</p>
|
||||
<section id="methods" class="level4">
|
||||
@@ -503,27 +503,30 @@ Advanced disk-based gradient checkpointer with prefetching.</p>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_disk.Disco.backward">backward</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco.backward">backward</a></td>
|
||||
<td>Backward pass that loads activations from disk with prefetching</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_disk.Disco.forward">forward</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco.forward">forward</a></td>
|
||||
<td>Forward pass that offloads activations to disk asynchronously</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_disk.Disco.get_instance">get_instance</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco.get_instance">get_instance</a></td>
|
||||
<td>Get or create the offload manager</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk.Disco.backward" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_disk.Disco.backward">backward</h5>
|
||||
<div class="sourceCode" id="cb2"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_disk.Disco.backward(ctx, <span class="op">*</span>grad_outputs)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco.backward" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco.backward">backward</h5>
|
||||
<div class="sourceCode" id="cb2"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_disk.Disco.backward(</span>
|
||||
<span id="cb2-2"><a href="#cb2-2" aria-hidden="true" tabindex="-1"></a> ctx,</span>
|
||||
<span id="cb2-3"><a href="#cb2-3" aria-hidden="true" tabindex="-1"></a> <span class="op">*</span>grad_outputs,</span>
|
||||
<span id="cb2-4"><a href="#cb2-4" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Backward pass that loads activations from disk with prefetching</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk.Disco.forward" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_disk.Disco.forward">forward</h5>
|
||||
<div class="sourceCode" id="cb3"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_disk.Disco.forward(</span>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco.forward" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco.forward">forward</h5>
|
||||
<div class="sourceCode" id="cb3"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_disk.Disco.forward(</span>
|
||||
<span id="cb3-2"><a href="#cb3-2" aria-hidden="true" tabindex="-1"></a> ctx,</span>
|
||||
<span id="cb3-3"><a href="#cb3-3" aria-hidden="true" tabindex="-1"></a> forward_function,</span>
|
||||
<span id="cb3-4"><a href="#cb3-4" aria-hidden="true" tabindex="-1"></a> hidden_states,</span>
|
||||
@@ -534,9 +537,9 @@ Advanced disk-based gradient checkpointer with prefetching.</p>
|
||||
<span id="cb3-9"><a href="#cb3-9" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Forward pass that offloads activations to disk asynchronously</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk.Disco.get_instance" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_disk.Disco.get_instance">get_instance</h5>
|
||||
<div class="sourceCode" id="cb4"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_disk.Disco.get_instance(</span>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco.get_instance" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.Disco.get_instance">get_instance</h5>
|
||||
<div class="sourceCode" id="cb4"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_disk.Disco.get_instance(</span>
|
||||
<span id="cb4-2"><a href="#cb4-2" aria-hidden="true" tabindex="-1"></a> prefetch_size<span class="op">=</span><span class="dv">1</span>,</span>
|
||||
<span id="cb4-3"><a href="#cb4-3" aria-hidden="true" tabindex="-1"></a> prefetch_to_gpu<span class="op">=</span><span class="va">True</span>,</span>
|
||||
<span id="cb4-4"><a href="#cb4-4" aria-hidden="true" tabindex="-1"></a> save_workers<span class="op">=</span><span class="dv">4</span>,</span>
|
||||
@@ -545,9 +548,9 @@ Advanced disk-based gradient checkpointer with prefetching.</p>
|
||||
</section>
|
||||
</section>
|
||||
</section>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager">DiskOffloadManager</h3>
|
||||
<div class="sourceCode" id="cb5"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_disk.DiskOffloadManager(</span>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager">DiskOffloadManager</h3>
|
||||
<div class="sourceCode" id="cb5"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager(</span>
|
||||
<span id="cb5-2"><a href="#cb5-2" aria-hidden="true" tabindex="-1"></a> <span class="va">self</span>,</span>
|
||||
<span id="cb5-3"><a href="#cb5-3" aria-hidden="true" tabindex="-1"></a> prefetch_size<span class="op">=</span><span class="dv">3</span>,</span>
|
||||
<span id="cb5-4"><a href="#cb5-4" aria-hidden="true" tabindex="-1"></a> prefetch_to_gpu<span class="op">=</span><span class="va">True</span>,</span>
|
||||
@@ -566,66 +569,68 @@ Includes synchronization to prevent race conditions.</p>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup">cleanup</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup">cleanup</a></td>
|
||||
<td>Clean up all temp files and stop prefetch thread with proper synchronization</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup_tensor">cleanup_tensor</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup_tensor">cleanup_tensor</a></td>
|
||||
<td>Clean up a specific tensor file after it’s been used</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.load_tensor">load_tensor</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.load_tensor">load_tensor</a></td>
|
||||
<td>Load tensor from disk or prefetch cache with proper synchronization</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.save_tensor">save_tensor</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.save_tensor">save_tensor</a></td>
|
||||
<td>Save tensor to disk asynchronously and return file path with thread-safe operations</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.trigger_prefetch">trigger_prefetch</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.trigger_prefetch">trigger_prefetch</a></td>
|
||||
<td>Trigger prefetching of the next N tensors with proper synchronization</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.wait_for_save">wait_for_save</a></td>
|
||||
<td><a href="#axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.wait_for_save">wait_for_save</a></td>
|
||||
<td>Wait for a tensor to be saved to disk</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup">cleanup</h5>
|
||||
<div class="sourceCode" id="cb6"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb6-1"><a href="#cb6-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup">cleanup</h5>
|
||||
<div class="sourceCode" id="cb6"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb6-1"><a href="#cb6-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Clean up all temp files and stop prefetch thread with proper synchronization</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup_tensor" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup_tensor">cleanup_tensor</h5>
|
||||
<div class="sourceCode" id="cb7"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb7-1"><a href="#cb7-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup_tensor(</span>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup_tensor" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup_tensor">cleanup_tensor</h5>
|
||||
<div class="sourceCode" id="cb7"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb7-1"><a href="#cb7-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.cleanup_tensor(</span>
|
||||
<span id="cb7-2"><a href="#cb7-2" aria-hidden="true" tabindex="-1"></a> file_path,</span>
|
||||
<span id="cb7-3"><a href="#cb7-3" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Clean up a specific tensor file after it’s been used</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.load_tensor" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.load_tensor">load_tensor</h5>
|
||||
<div class="sourceCode" id="cb8"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb8-1"><a href="#cb8-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_disk.DiskOffloadManager.load_tensor(</span>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.load_tensor" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.load_tensor">load_tensor</h5>
|
||||
<div class="sourceCode" id="cb8"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb8-1"><a href="#cb8-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.load_tensor(</span>
|
||||
<span id="cb8-2"><a href="#cb8-2" aria-hidden="true" tabindex="-1"></a> file_path,</span>
|
||||
<span id="cb8-3"><a href="#cb8-3" aria-hidden="true" tabindex="-1"></a> target_device<span class="op">=</span><span class="st">'cuda'</span>,</span>
|
||||
<span id="cb8-4"><a href="#cb8-4" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Load tensor from disk or prefetch cache with proper synchronization</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.save_tensor" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.save_tensor">save_tensor</h5>
|
||||
<div class="sourceCode" id="cb9"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_disk.DiskOffloadManager.save_tensor(tensor)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.save_tensor" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.save_tensor">save_tensor</h5>
|
||||
<div class="sourceCode" id="cb9"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.save_tensor(</span>
|
||||
<span id="cb9-2"><a href="#cb9-2" aria-hidden="true" tabindex="-1"></a> tensor,</span>
|
||||
<span id="cb9-3"><a href="#cb9-3" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Save tensor to disk asynchronously and return file path with thread-safe operations</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.trigger_prefetch" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.trigger_prefetch">trigger_prefetch</h5>
|
||||
<div class="sourceCode" id="cb10"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb10-1"><a href="#cb10-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_disk.DiskOffloadManager.trigger_prefetch(</span>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.trigger_prefetch" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.trigger_prefetch">trigger_prefetch</h5>
|
||||
<div class="sourceCode" id="cb10"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb10-1"><a href="#cb10-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.trigger_prefetch(</span>
|
||||
<span id="cb10-2"><a href="#cb10-2" aria-hidden="true" tabindex="-1"></a> n<span class="op">=</span><span class="va">None</span>,</span>
|
||||
<span id="cb10-3"><a href="#cb10-3" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Trigger prefetching of the next N tensors with proper synchronization</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.wait_for_save" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.gradient_checkpointing.offload_disk.DiskOffloadManager.wait_for_save">wait_for_save</h5>
|
||||
<div class="sourceCode" id="cb11"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb11-1"><a href="#cb11-1" aria-hidden="true" tabindex="-1"></a>utils.gradient_checkpointing.offload_disk.DiskOffloadManager.wait_for_save(</span>
|
||||
<section id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.wait_for_save" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.wait_for_save">wait_for_save</h5>
|
||||
<div class="sourceCode" id="cb11"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb11-1"><a href="#cb11-1" aria-hidden="true" tabindex="-1"></a>monkeypatch.gradient_checkpointing.offload_disk.DiskOffloadManager.wait_for_save(</span>
|
||||
<span id="cb11-2"><a href="#cb11-2" aria-hidden="true" tabindex="-1"></a> file_path,</span>
|
||||
<span id="cb11-3"><a href="#cb11-3" aria-hidden="true" tabindex="-1"></a> timeout<span class="op">=</span><span class="va">None</span>,</span>
|
||||
<span id="cb11-4"><a href="#cb11-4" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
@@ -525,19 +525,19 @@ sequential packing (preserving original sequence order).</p>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.samplers.multipack.MultipackBatchSampler.efficiency">efficiency</a></td>
|
||||
<td>Calculate the packing efficiency (ratio of tokens used to total token slots)</td>
|
||||
<td>Calculate the packing efficiency (ratio of tokens used to total token slots).</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.samplers.multipack.MultipackBatchSampler.gather_efficiency">gather_efficiency</a></td>
|
||||
<td>Gather and synchronize packing efficiency estimates across all distributed ranks</td>
|
||||
<td>Gather and synchronize packing efficiency estimates across all distributed</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.samplers.multipack.MultipackBatchSampler.gather_len_batches">gather_len_batches</a></td>
|
||||
<td>Gather and synchronize batch counts across all distributed ranks</td>
|
||||
<td>Gather and synchronize batch counts across all distributed ranks. Returns</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.samplers.multipack.MultipackBatchSampler.generate_batches">generate_batches</a></td>
|
||||
<td>Generate packed batches for training</td>
|
||||
<td>Generate packed batches for training.</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.samplers.multipack.MultipackBatchSampler.set_epoch">set_epoch</a></td>
|
||||
@@ -548,32 +548,56 @@ sequential packing (preserving original sequence order).</p>
|
||||
<section id="axolotl.utils.samplers.multipack.MultipackBatchSampler.efficiency" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.samplers.multipack.MultipackBatchSampler.efficiency">efficiency</h5>
|
||||
<div class="sourceCode" id="cb2"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb2-1"><a href="#cb2-1" aria-hidden="true" tabindex="-1"></a>utils.samplers.multipack.MultipackBatchSampler.efficiency()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Calculate the packing efficiency (ratio of tokens used to total token slots)
|
||||
Higher is better - 1.0 would mean perfect packing with no wasted space</p>
|
||||
<p>Calculate the packing efficiency (ratio of tokens used to total token slots).
|
||||
Higher is better - 1.0 would mean perfect packing with no wasted space.</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.samplers.multipack.MultipackBatchSampler.gather_efficiency" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.samplers.multipack.MultipackBatchSampler.gather_efficiency">gather_efficiency</h5>
|
||||
<div class="sourceCode" id="cb3"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb3-1"><a href="#cb3-1" aria-hidden="true" tabindex="-1"></a>utils.samplers.multipack.MultipackBatchSampler.gather_efficiency()</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Gather and synchronize packing efficiency estimates across all distributed ranks
|
||||
Returns a conservative efficiency estimate based on the measurements</p>
|
||||
<p>Gather and synchronize packing efficiency estimates across all distributed
|
||||
ranks.</p>
|
||||
<section id="returns" class="level6 doc-section doc-section-returns">
|
||||
<h6 class="doc-section doc-section-returns anchored" data-anchor-id="returns">Returns</h6>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 10%">
|
||||
<col style="width: 10%">
|
||||
<col style="width: 79%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Type</th>
|
||||
<th>Description</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td></td>
|
||||
<td>float</td>
|
||||
<td>A conservative efficiency estimate based on the measurements.</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
</section>
|
||||
<section id="axolotl.utils.samplers.multipack.MultipackBatchSampler.gather_len_batches" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.samplers.multipack.MultipackBatchSampler.gather_len_batches">gather_len_batches</h5>
|
||||
<div class="sourceCode" id="cb4"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb4-1"><a href="#cb4-1" aria-hidden="true" tabindex="-1"></a>utils.samplers.multipack.MultipackBatchSampler.gather_len_batches(num)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Gather and synchronize batch counts across all distributed ranks
|
||||
Returns the minimum number of batches available on any rank</p>
|
||||
<p>Gather and synchronize batch counts across all distributed ranks. Returns
|
||||
the minimum number of batches available on any rank.</p>
|
||||
</section>
|
||||
<section id="axolotl.utils.samplers.multipack.MultipackBatchSampler.generate_batches" class="level5">
|
||||
<h5 class="anchored" data-anchor-id="axolotl.utils.samplers.multipack.MultipackBatchSampler.generate_batches">generate_batches</h5>
|
||||
<div class="sourceCode" id="cb5"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb5-1"><a href="#cb5-1" aria-hidden="true" tabindex="-1"></a>utils.samplers.multipack.MultipackBatchSampler.generate_batches(set_stats<span class="op">=</span><span class="va">False</span>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Generate packed batches for training</p>
|
||||
<p>Generate packed batches for training.</p>
|
||||
<section id="parameters" class="level6 doc-section doc-section-parameters">
|
||||
<h6 class="doc-section doc-section-parameters anchored" data-anchor-id="parameters">Parameters</h6>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 15%">
|
||||
<col style="width: 11%">
|
||||
<col style="width: 57%">
|
||||
<col style="width: 58%">
|
||||
<col style="width: 15%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
@@ -587,20 +611,20 @@ Returns the minimum number of batches available on any rank</p>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td>set_stats</td>
|
||||
<td></td>
|
||||
<td>Whether to update efficiency statistics</td>
|
||||
<td>bool</td>
|
||||
<td>Whether to update efficiency statistics.</td>
|
||||
<td><code>False</code></td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
<section id="returns" class="level6 doc-section doc-section-returns">
|
||||
<h6 class="doc-section doc-section-returns anchored" data-anchor-id="returns">Returns</h6>
|
||||
<section id="returns-1" class="level6 doc-section doc-section-returns">
|
||||
<h6 class="doc-section doc-section-returns anchored" data-anchor-id="returns-1">Returns</h6>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 10%">
|
||||
<col style="width: 10%">
|
||||
<col style="width: 78%">
|
||||
<col style="width: 5%">
|
||||
<col style="width: 20%">
|
||||
<col style="width: 74%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
@@ -612,13 +636,8 @@ Returns the minimum number of batches available on any rank</p>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td></td>
|
||||
<td></td>
|
||||
<td>List of batches, where each batch contains multiple bins,</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td></td>
|
||||
<td></td>
|
||||
<td>and each bin contains multiple sequence indices</td>
|
||||
<td>list[list[list[int]]]</td>
|
||||
<td>List of batches, where each batch contains multiple bins, and each bin contains multiple sequence indices.</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
@@ -644,19 +663,19 @@ Returns the minimum number of batches available on any rank</p>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.samplers.multipack.allocate_sequentially">allocate_sequentially</a></td>
|
||||
<td>Sequential allocator that preserves example order</td>
|
||||
<td>Sequential allocator that preserves example order.</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.samplers.multipack.ffd_check">ffd_check</a></td>
|
||||
<td>First-fit-decreasing bin packing algorithm check</td>
|
||||
<td>First-fit-decreasing bin packing algorithm check.</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td><a href="#axolotl.utils.samplers.multipack.pack_group">pack_group</a></td>
|
||||
<td>Pack a group of sequences into bins using First-Fit Decreasing algorithm</td>
|
||||
<td>Pack a group of sequences into bins using First-Fit Decreasing algorithm.</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td><a href="#axolotl.utils.samplers.multipack.pack_parallel">pack_parallel</a></td>
|
||||
<td>Pack sequences into bins using parallel processing</td>
|
||||
<td>Pack sequences into bins using parallel processing.</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
@@ -668,11 +687,172 @@ Returns the minimum number of batches available on any rank</p>
|
||||
<span id="cb7-4"><a href="#cb7-4" aria-hidden="true" tabindex="-1"></a> bin_capacity,</span>
|
||||
<span id="cb7-5"><a href="#cb7-5" aria-hidden="true" tabindex="-1"></a> num_ranks,</span>
|
||||
<span id="cb7-6"><a href="#cb7-6" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Sequential allocator that preserves example order</p>
|
||||
<p>Sequential allocator that preserves example order.</p>
|
||||
<section id="parameters-1" class="level4 doc-section doc-section-parameters">
|
||||
<h4 class="doc-section doc-section-parameters anchored" data-anchor-id="parameters-1">Parameters</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 18%">
|
||||
<col style="width: 12%">
|
||||
<col style="width: 55%">
|
||||
<col style="width: 12%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Type</th>
|
||||
<th>Description</th>
|
||||
<th>Default</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td>sequence_lengths</td>
|
||||
<td>np.ndarray</td>
|
||||
<td>The lengths of all examples.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>rank</td>
|
||||
<td>int</td>
|
||||
<td>The current rank (for distributed training).</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td>bin_capacity</td>
|
||||
<td>int</td>
|
||||
<td>The capacity of each bin (maximum sequence length).</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>num_ranks</td>
|
||||
<td>int</td>
|
||||
<td>Number of ranks (processes / GPUs).</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
<section id="returns-2" class="level4 doc-section doc-section-returns">
|
||||
<h4 class="doc-section doc-section-returns anchored" data-anchor-id="returns-2">Returns</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 15%">
|
||||
<col style="width: 17%">
|
||||
<col style="width: 66%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Type</th>
|
||||
<th>Description</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td>rank_batches</td>
|
||||
<td>list[list[int]]</td>
|
||||
<td>List of batches for the current rank.</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>total_tokens_used</td>
|
||||
<td>int</td>
|
||||
<td>Number of actual example tokens.</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td>total_token_slots</td>
|
||||
<td>int</td>
|
||||
<td>Maximum theoretical number of example tokens (number of bins * bin capacity).</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
</section>
|
||||
<section id="axolotl.utils.samplers.multipack.ffd_check" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.samplers.multipack.ffd_check">ffd_check</h3>
|
||||
<div class="sourceCode" id="cb8"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb8-1"><a href="#cb8-1" aria-hidden="true" tabindex="-1"></a>utils.samplers.multipack.ffd_check(sequence_lengths, bin_capacity, num_bins)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>First-fit-decreasing bin packing algorithm check.</p>
|
||||
<p>Checks if sequences with the given lengths could fit in the specified number of
|
||||
bins.</p>
|
||||
<section id="parameters-2" class="level4 doc-section doc-section-parameters">
|
||||
<h4 class="doc-section doc-section-parameters anchored" data-anchor-id="parameters-2">Parameters</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 24%">
|
||||
<col style="width: 16%">
|
||||
<col style="width: 42%">
|
||||
<col style="width: 16%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Type</th>
|
||||
<th>Description</th>
|
||||
<th>Default</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td>sequence_lengths</td>
|
||||
<td>np.ndarray</td>
|
||||
<td>Array of sequence lengths.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>bin_capacity</td>
|
||||
<td>int</td>
|
||||
<td>Maximum capacity of each bin.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td>num_bins</td>
|
||||
<td>int</td>
|
||||
<td>Number of bins available.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
<section id="returns-3" class="level4 doc-section doc-section-returns">
|
||||
<h4 class="doc-section doc-section-returns anchored" data-anchor-id="returns-3">Returns</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 10%">
|
||||
<col style="width: 10%">
|
||||
<col style="width: 78%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Type</th>
|
||||
<th>Description</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td></td>
|
||||
<td>bool</td>
|
||||
<td><code>True</code> if all sequences can be packed, <code>False</code> otherwise.</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
</section>
|
||||
<section id="axolotl.utils.samplers.multipack.pack_group" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.samplers.multipack.pack_group">pack_group</h3>
|
||||
<div class="sourceCode" id="cb9"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a>utils.samplers.multipack.pack_group(</span>
|
||||
<span id="cb9-2"><a href="#cb9-2" aria-hidden="true" tabindex="-1"></a> sequence_lengths,</span>
|
||||
<span id="cb9-3"><a href="#cb9-3" aria-hidden="true" tabindex="-1"></a> group_offset,</span>
|
||||
<span id="cb9-4"><a href="#cb9-4" aria-hidden="true" tabindex="-1"></a> bin_capacity,</span>
|
||||
<span id="cb9-5"><a href="#cb9-5" aria-hidden="true" tabindex="-1"></a> max_bins,</span>
|
||||
<span id="cb9-6"><a href="#cb9-6" aria-hidden="true" tabindex="-1"></a> bin_size,</span>
|
||||
<span id="cb9-7"><a href="#cb9-7" aria-hidden="true" tabindex="-1"></a> safe_mode<span class="op">=</span><span class="va">True</span>,</span>
|
||||
<span id="cb9-8"><a href="#cb9-8" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Pack a group of sequences into bins using First-Fit Decreasing algorithm.</p>
|
||||
<section id="parameters-3" class="level4 doc-section doc-section-parameters">
|
||||
<h4 class="doc-section doc-section-parameters anchored" data-anchor-id="parameters-3">Parameters</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 19%">
|
||||
<col style="width: 12%">
|
||||
<col style="width: 55%">
|
||||
@@ -690,209 +870,49 @@ Returns the minimum number of batches available on any rank</p>
|
||||
<tr class="odd">
|
||||
<td>sequence_lengths</td>
|
||||
<td>np.ndarray</td>
|
||||
<td>The lengths of all examples</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>rank</td>
|
||||
<td>int</td>
|
||||
<td>The current rank (for distributed training)</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td>bin_capacity</td>
|
||||
<td>int</td>
|
||||
<td>The capacity of each bin (maximum sequence length)</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>num_ranks</td>
|
||||
<td>int</td>
|
||||
<td>Number of ranks (processes/GPUs)</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
<section id="returns-1" class="level4 doc-section doc-section-returns">
|
||||
<h4 class="doc-section doc-section-returns anchored" data-anchor-id="returns-1">Returns</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 18%">
|
||||
<col style="width: 7%">
|
||||
<col style="width: 74%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Type</th>
|
||||
<th>Description</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td>rank_batches</td>
|
||||
<td></td>
|
||||
<td>List of batches for the current rank</td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>total_tokens_used</td>
|
||||
<td></td>
|
||||
<td>Number of actual example tokens</td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td>total_token_slots</td>
|
||||
<td></td>
|
||||
<td>Maximum theoretical number of example tokens (number of bins * bin capacity)</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
</section>
|
||||
<section id="axolotl.utils.samplers.multipack.ffd_check" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.samplers.multipack.ffd_check">ffd_check</h3>
|
||||
<div class="sourceCode" id="cb8"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb8-1"><a href="#cb8-1" aria-hidden="true" tabindex="-1"></a>utils.samplers.multipack.ffd_check(sequence_lengths, bin_capacity, num_bins)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>First-fit-decreasing bin packing algorithm check</p>
|
||||
<p>Checks if sequences with the given lengths could fit in the specified number of bins</p>
|
||||
<section id="parameters-2" class="level4 doc-section doc-section-parameters">
|
||||
<h4 class="doc-section doc-section-parameters anchored" data-anchor-id="parameters-2">Parameters</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 25%">
|
||||
<col style="width: 16%">
|
||||
<col style="width: 41%">
|
||||
<col style="width: 16%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Type</th>
|
||||
<th>Description</th>
|
||||
<th>Default</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td>sequence_lengths</td>
|
||||
<td>np.ndarray</td>
|
||||
<td>Array of sequence lengths</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>bin_capacity</td>
|
||||
<td>int</td>
|
||||
<td>Maximum capacity of each bin</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td>num_bins</td>
|
||||
<td>int</td>
|
||||
<td>Number of bins available</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
<section id="returns-2" class="level4 doc-section doc-section-returns">
|
||||
<h4 class="doc-section doc-section-returns anchored" data-anchor-id="returns-2">Returns</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 11%">
|
||||
<col style="width: 11%">
|
||||
<col style="width: 77%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Type</th>
|
||||
<th>Description</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td></td>
|
||||
<td></td>
|
||||
<td>True if all sequences can be packed, False otherwise</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
</section>
|
||||
<section id="axolotl.utils.samplers.multipack.pack_group" class="level3">
|
||||
<h3 class="anchored" data-anchor-id="axolotl.utils.samplers.multipack.pack_group">pack_group</h3>
|
||||
<div class="sourceCode" id="cb9"><pre class="sourceCode python code-with-copy"><code class="sourceCode python"><span id="cb9-1"><a href="#cb9-1" aria-hidden="true" tabindex="-1"></a>utils.samplers.multipack.pack_group(</span>
|
||||
<span id="cb9-2"><a href="#cb9-2" aria-hidden="true" tabindex="-1"></a> sequence_lengths,</span>
|
||||
<span id="cb9-3"><a href="#cb9-3" aria-hidden="true" tabindex="-1"></a> group_offset,</span>
|
||||
<span id="cb9-4"><a href="#cb9-4" aria-hidden="true" tabindex="-1"></a> bin_capacity,</span>
|
||||
<span id="cb9-5"><a href="#cb9-5" aria-hidden="true" tabindex="-1"></a> max_bins,</span>
|
||||
<span id="cb9-6"><a href="#cb9-6" aria-hidden="true" tabindex="-1"></a> bin_size,</span>
|
||||
<span id="cb9-7"><a href="#cb9-7" aria-hidden="true" tabindex="-1"></a> safe_mode<span class="op">=</span><span class="va">True</span>,</span>
|
||||
<span id="cb9-8"><a href="#cb9-8" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Pack a group of sequences into bins using First-Fit Decreasing algorithm</p>
|
||||
<section id="parameters-3" class="level4 doc-section doc-section-parameters">
|
||||
<h4 class="doc-section doc-section-parameters anchored" data-anchor-id="parameters-3">Parameters</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 19%">
|
||||
<col style="width: 12%">
|
||||
<col style="width: 54%">
|
||||
<col style="width: 12%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
<th>Name</th>
|
||||
<th>Type</th>
|
||||
<th>Description</th>
|
||||
<th>Default</th>
|
||||
</tr>
|
||||
</thead>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td>sequence_lengths</td>
|
||||
<td>np.ndarray</td>
|
||||
<td>Array of sequence lengths</td>
|
||||
<td>Array of sequence lengths.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>group_offset</td>
|
||||
<td>int</td>
|
||||
<td>Offset to apply to indices when returning results</td>
|
||||
<td>Offset to apply to indices when returning results.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td>bin_capacity</td>
|
||||
<td>int</td>
|
||||
<td>Maximum capacity of each bin</td>
|
||||
<td>Maximum capacity of each bin.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>max_bins</td>
|
||||
<td>int</td>
|
||||
<td>Maximum number of bins to use</td>
|
||||
<td>Maximum number of bins to use.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td>bin_size</td>
|
||||
<td>int</td>
|
||||
<td>Maximum number of sequences per bin</td>
|
||||
<td>Maximum number of sequences per bin.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>safe_mode</td>
|
||||
<td>bool</td>
|
||||
<td>If True, use a more conservative packing approach</td>
|
||||
<td>If True, use a more conservative packing approach.</td>
|
||||
<td><code>True</code></td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
</section>
|
||||
<section id="returns-3" class="level4 doc-section doc-section-returns">
|
||||
<h4 class="doc-section doc-section-returns anchored" data-anchor-id="returns-3">Returns</h4>
|
||||
<section id="returns-4" class="level4 doc-section doc-section-returns">
|
||||
<h4 class="doc-section doc-section-returns anchored" data-anchor-id="returns-4">Returns</h4>
|
||||
<table class="caption-top table">
|
||||
<colgroup>
|
||||
<col style="width: 8%">
|
||||
<col style="width: 8%">
|
||||
<col style="width: 82%">
|
||||
<col style="width: 7%">
|
||||
<col style="width: 20%">
|
||||
<col style="width: 72%">
|
||||
</colgroup>
|
||||
<thead>
|
||||
<tr class="header">
|
||||
@@ -904,8 +924,8 @@ Returns the minimum number of batches available on any rank</p>
|
||||
<tbody>
|
||||
<tr class="odd">
|
||||
<td></td>
|
||||
<td></td>
|
||||
<td>List of bins, where each bin contains indices of sequences assigned to it</td>
|
||||
<td>list[list[int]]</td>
|
||||
<td>List of bins, where each bin contains indices of sequences assigned to it.</td>
|
||||
</tr>
|
||||
</tbody>
|
||||
</table>
|
||||
@@ -922,7 +942,7 @@ Returns the minimum number of batches available on any rank</p>
|
||||
<span id="cb10-7"><a href="#cb10-7" aria-hidden="true" tabindex="-1"></a> safe_mode<span class="op">=</span><span class="va">True</span>,</span>
|
||||
<span id="cb10-8"><a href="#cb10-8" aria-hidden="true" tabindex="-1"></a> mp_start_method<span class="op">=</span><span class="st">'spawn'</span>,</span>
|
||||
<span id="cb10-9"><a href="#cb10-9" aria-hidden="true" tabindex="-1"></a>)</span></code><button title="Copy to Clipboard" class="code-copy-button"><i class="bi"></i></button></pre></div>
|
||||
<p>Pack sequences into bins using parallel processing</p>
|
||||
<p>Pack sequences into bins using parallel processing.</p>
|
||||
<section id="parameters-4" class="level4 doc-section doc-section-parameters">
|
||||
<h4 class="doc-section doc-section-parameters anchored" data-anchor-id="parameters-4">Parameters</h4>
|
||||
<table class="caption-top table">
|
||||
@@ -944,37 +964,37 @@ Returns the minimum number of batches available on any rank</p>
|
||||
<tr class="odd">
|
||||
<td>sequence_lengths</td>
|
||||
<td>np.ndarray</td>
|
||||
<td>Array of sequence lengths</td>
|
||||
<td>Array of sequence lengths.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>bin_capacity</td>
|
||||
<td>int</td>
|
||||
<td>Maximum capacity of each bin as total number of tokens</td>
|
||||
<td>Maximum capacity of each bin as total number of tokens.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td>group_size</td>
|
||||
<td>int</td>
|
||||
<td>Number of sequences to process in each group</td>
|
||||
<td>Number of sequences to process in each group.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>bin_size</td>
|
||||
<td>int</td>
|
||||
<td>Maximum number of bins to use</td>
|
||||
<td>Maximum number of bins to use.</td>
|
||||
<td><em>required</em></td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
<td>num_processes</td>
|
||||
<td>int | None</td>
|
||||
<td>Number of parallel processes to use</td>
|
||||
<td>Number of parallel processes to use.</td>
|
||||
<td><code>None</code></td>
|
||||
</tr>
|
||||
<tr class="even">
|
||||
<td>safe_mode</td>
|
||||
<td>bool</td>
|
||||
<td>If True, use a more conservative packing approach</td>
|
||||
<td>If True, use a more conservative packing approach.</td>
|
||||
<td><code>True</code></td>
|
||||
</tr>
|
||||
<tr class="odd">
|
||||
@@ -986,7 +1006,7 @@ Returns the minimum number of batches available on any rank</p>
|
||||
</tbody>
|
||||
</table>
|
||||
<p>Returns:
|
||||
List of bins, where each bin contains indices of sequences assigned to it</p>
|
||||
List of bins, where each bin contains indices of sequences assigned to it.</p>
|
||||
|
||||
|
||||
</section>
|
||||
|
||||
@@ -641,7 +641,7 @@ Tip
|
||||
<p>Check our other guides for details on these topics:</p>
|
||||
<ul>
|
||||
<li><a href="../docs/config.html">Configuration Guide</a> - Full configuration options</li>
|
||||
<li><a href="dataset-loading.qmd">Dataset Loading</a> - Loading datasets from various sources</li>
|
||||
<li><a href="../docs/dataset_loading.html">Dataset Loading</a> - Loading datasets from various sources</li>
|
||||
<li><a href="dataset-formats">Dataset Formats</a> - Working with different data formats</li>
|
||||
<li><a href="../docs/multi-gpu.html">Multi-GPU Training</a></li>
|
||||
<li><a href="../docs/multi-node.html">Multi-Node Training</a></li>
|
||||
|
||||
1778
search.json
1778
search.json
File diff suppressed because one or more lines are too long
1136
sitemap.xml
1136
sitemap.xml
File diff suppressed because it is too large
Load Diff
Reference in New Issue
Block a user