<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
<meta charset="utf-8">
<meta name="generator" content="Docutils 0.18.1: http://docutils.sourceforge.net/" />
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>Tutorial 8: Multiple-Instance Learning — slideflow 3.0.0 documentation</title>
<link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
<!-- <link rel="stylesheet" href="../_static/pygments.css" type="text/css" /> -->
<link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
<link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
<link rel="index" title="Index" href="../genindex/" />
<link rel="search" title="Search" href="../search/" />
<link rel="prev" title="Tutorial 7: Training with custom augmentations" href="../tutorial7/" />
<script src="../_static/js/modernizr.min.js"></script>
<!-- Preload the theme fonts -->
<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-book.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-Medium.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="../_static/fonts/FreightSans/freight-sans-medium-italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="../_static/fonts/IBMPlexMono/IBMPlexMono-SemiBold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<!-- Preload the katex fonts -->
<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Math-Italic.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Main-Bold.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size1-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size4-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size2-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Size3-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="preload" href="https://cdn.jsdelivr.net/npm/katex@0.10.0/dist/fonts/KaTeX_Caligraphic-Regular.woff2" as="font" type="font/woff2" crossorigin="anonymous">
<link rel="stylesheet" href="https://use.fontawesome.com/releases/v5.15.2/css/all.css" integrity="sha384-vSIIfh2YWi9wW0r9iZe7RJPrKwp6bG+s9QZMoITbCckVJqGCCRhc+ccxNcdpHuYu" crossorigin="anonymous">
<script defer data-domain="slideflow.dev" src="https://plausible.io/js/script.js"></script>
</head>
<div class="container-fluid header-holder tutorials-header" id="header-holder">
<div class="container">
<div class="header-container">
<a class="header-logo" href="https://slideflow.dev" aria-label="Slideflow"></a>
<div class="main-menu">
<ul>
<li class="active">
<a href="https://slideflow.dev">Docs</a>
</li>
<li>
<a href="https://slideflow.dev/tutorial1/">Tutorials</a>
</li>
<li>
<a href="https://github.com/slideflow/slideflow">GitHub</a>
</li>
</ul>
</div>
<a class="main-menu-open-button" href="#" data-behavior="open-mobile-menu"></a>
</div>
</div>
</div>
<body class="pytorch-body">
<div class="table-of-contents-link-wrapper">
<span>Table of Contents</span>
<a href="#" class="toggle-table-of-contents" data-behavior="toggle-table-of-contents"></a>
</div>
<nav data-toggle="wy-nav-shift" class="pytorch-left-menu" id="pytorch-left-menu">
<div class="pytorch-side-scroll">
<div class="pytorch-menu pytorch-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
<div class="pytorch-left-menu-search">
<div class="version">
3.0
</div>
<div role="search">
<form id="rtd-search-form" class="wy-form" action="../search/" method="get">
<input type="text" name="q" placeholder="Search Docs" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
</div>
</div>
<p class="caption" role="heading"><span class="caption-text">Introduction</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../installation/">Installation</a></li>
<li class="toctree-l1"><a class="reference internal" href="../overview/">Overview</a></li>
<li class="toctree-l1"><a class="reference internal" href="../quickstart/">Quickstart</a></li>
<li class="toctree-l1"><a class="reference internal" href="../project_setup/">Setting up a Project</a></li>
<li class="toctree-l1"><a class="reference internal" href="../datasets_and_val/">Datasets</a></li>
<li class="toctree-l1"><a class="reference internal" href="../slide_processing/">Slide Processing</a></li>
<li class="toctree-l1"><a class="reference internal" href="../training/">Training</a></li>
<li class="toctree-l1"><a class="reference internal" href="../evaluation/">Evaluation</a></li>
<li class="toctree-l1"><a class="reference internal" href="../posthoc/">Layer Activations</a></li>
<li class="toctree-l1"><a class="reference internal" href="../uq/">Uncertainty Quantification</a></li>
<li class="toctree-l1"><a class="reference internal" href="../features/">Generating Features</a></li>
<li class="toctree-l1"><a class="reference internal" href="../mil/">Multiple-Instance Learning (MIL)</a></li>
<li class="toctree-l1"><a class="reference internal" href="../ssl/">Self-Supervised Learning (SSL)</a></li>
<li class="toctree-l1"><a class="reference internal" href="../stylegan/">Generative Networks (GANs)</a></li>
<li class="toctree-l1"><a class="reference internal" href="../saliency/">Saliency Maps</a></li>
<li class="toctree-l1"><a class="reference internal" href="../segmentation/">Tissue Segmentation</a></li>
<li class="toctree-l1"><a class="reference internal" href="../cellseg/">Cell Segmentation</a></li>
<li class="toctree-l1"><a class="reference internal" href="../custom_loops/">Custom Training Loops</a></li>
<li class="toctree-l1"><a class="reference internal" href="../studio/">Slideflow Studio: Live Visualization</a></li>
<li class="toctree-l1"><a class="reference internal" href="../troubleshooting/">Troubleshooting</a></li>
</ul>
<p class="caption" role="heading"><span class="caption-text">Developer Notes</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../tfrecords/">TFRecords: Reading and Writing</a></li>
<li class="toctree-l1"><a class="reference internal" href="../dataloaders/">Dataloaders: Sampling and Augmentation</a></li>
<li class="toctree-l1"><a class="reference internal" href="../custom_extractors/">Custom Feature Extractors</a></li>
<li class="toctree-l1"><a class="reference internal" href="../tile_labels/">Strong Supervision with Tile Labels</a></li>
<li class="toctree-l1"><a class="reference internal" href="../plugins/">Creating a Slideflow Plugin</a></li>
</ul>
<p class="caption" role="heading"><span class="caption-text">API</span></p>
<ul>
<li class="toctree-l1"><a class="reference internal" href="../slideflow/">slideflow</a></li>
<li class="toctree-l1"><a class="reference internal" href="../project/">slideflow.Project</a></li>
<li class="toctree-l1"><a class="reference internal" href="../dataset/">slideflow.Dataset</a></li>
<li class="toctree-l1"><a class="reference internal" href="../dataset_features/">slideflow.DatasetFeatures</a></li>
<li class="toctree-l1"><a class="reference internal" href="../heatmap/">slideflow.Heatmap</a></li>
<li class="toctree-l1"><a class="reference internal" href="../model_params/">slideflow.ModelParams</a></li>
<li class="toctree-l1"><a class="reference internal" href="../mosaic/">slideflow.Mosaic</a></li>
<li class="toctree-l1"><a class="reference internal" href="../slidemap/">slideflow.SlideMap</a></li>
<li class="toctree-l1"><a class="reference internal" href="../biscuit/">slideflow.biscuit</a></li>
<li class="toctree-l1"><a class="reference internal" href="../slideflow_cellseg/">slideflow.cellseg</a></li>
<li class="toctree-l1"><a class="reference internal" href="../io/">slideflow.io</a></li>
<li class="toctree-l1"><a class="reference internal" href="../io_tensorflow/">slideflow.io.tensorflow</a></li>
<li class="toctree-l1"><a class="reference internal" href="../io_torch/">slideflow.io.torch</a></li>
<li class="toctree-l1"><a class="reference internal" href="../gan/">slideflow.gan</a></li>
<li class="toctree-l1"><a class="reference internal" href="../grad/">slideflow.grad</a></li>
<li class="toctree-l1"><a class="reference internal" href="../mil_module/">slideflow.mil</a></li>
<li class="toctree-l1"><a class="reference internal" href="../model/">slideflow.model</a></li>
<li class="toctree-l1"><a class="reference internal" href="../model_tensorflow/">slideflow.model.tensorflow</a></li>
<li class="toctree-l1"><a class="reference internal" href="../model_torch/">slideflow.model.torch</a></li>
<li class="toctree-l1"><a class="reference internal" href="../norm/">slideflow.norm</a></li>
<li class="toctree-l1"><a class="reference internal" href="../simclr/">slideflow.simclr</a></li>
<li class="toctree-l1"><a class="reference internal" href="../slide/">slideflow.slide</a></li>
<li class="toctree-l1"><a class="reference internal" href="../slide_qc/">slideflow.slide.qc</a></li>
<li class="toctree-l1"><a class="reference internal" href="../stats/">slideflow.stats</a></li>
<li class="toctree-l1"><a class="reference internal" href="../util/">slideflow.util</a></li>
<li class="toctree-l1"><a class="reference internal" href="../studio_module/">slideflow.studio</a></li>
</ul>
<p class="caption" role="heading"><span class="caption-text">Tutorials</span></p>
<ul class="current">
<li class="toctree-l1"><a class="reference internal" href="../tutorial1/">Tutorial 1: Model training (simple)</a></li>
<li class="toctree-l1"><a class="reference internal" href="../tutorial2/">Tutorial 2: Model training (advanced)</a></li>
<li class="toctree-l1"><a class="reference internal" href="../tutorial3/">Tutorial 3: Using a custom architecture</a></li>
<li class="toctree-l1"><a class="reference internal" href="../tutorial4/">Tutorial 4: Model evaluation & heatmaps</a></li>
<li class="toctree-l1"><a class="reference internal" href="../tutorial5/">Tutorial 5: Creating a mosaic map</a></li>
<li class="toctree-l1"><a class="reference internal" href="../tutorial6/">Tutorial 6: Custom slide filtering</a></li>
<li class="toctree-l1"><a class="reference internal" href="../tutorial7/">Tutorial 7: Training with custom augmentations</a></li>
<li class="toctree-l1 current"><a class="current reference internal" href="#">Tutorial 8: Multiple-Instance Learning</a></li>
</ul>
</div>
</div>
</nav>
<div class="pytorch-container">
<div class="pytorch-page-level-bar" id="pytorch-page-level-bar">
<div class="pytorch-breadcrumbs-wrapper">
<div role="navigation" aria-label="breadcrumbs navigation">
<ul class="pytorch-breadcrumbs">
<li>
<a href="../">
Docs
</a> >
</li>
<li>Tutorial 8: Multiple-Instance Learning</li>
<li class="pytorch-breadcrumbs-aside">
<a href="../_sources/tutorial8.rst.txt" rel="nofollow"><img src="../_static/images/view-page-source-icon.svg"></a>
</li>
</ul>
</div>
</div>
<div class="pytorch-shortcuts-wrapper" id="pytorch-shortcuts-wrapper">
Shortcuts
</div>
</div>
<section data-toggle="wy-nav-shift" id="pytorch-content-wrap" class="pytorch-content-wrap">
<div class="pytorch-content-left">
<div class="rst-content">
<div role="main" class="main-content" itemscope="itemscope" itemtype="http://schema.org/Article">
<article itemprop="articleBody" id="pytorch-article" class="pytorch-article">
<section id="tutorial-8-multiple-instance-learning">
<span id="tutorial8"></span><h1>Tutorial 8: Multiple-Instance Learning<a class="headerlink" href="#tutorial-8-multiple-instance-learning" title="Permalink to this heading">¶</a></h1>
<p>In contrast with tutorials 1-4, which focused on training and evaluating traditional tile-based models, this tutorial provides an example of training a multiple-instance learning (MIL) model. MIL models are particularly useful for heterogeneous tumors, when only parts of a whole-slide image may carry a distinctive histological signature. In this tutorial, we’ll train a MIL model to predict the ER status of breast cancer patients from whole slide images. Note: MIL models require PyTorch.</p>
<p>We’ll start the same way as <a class="reference internal" href="../tutorial1/#tutorial1"><span class="std std-ref">Tutorial 1: Model training (simple)</span></a>, loading a project and preparing a dataset.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">import</span> <span class="nn">slideflow</span> <span class="k">as</span> <span class="nn">sf</span>
<span class="gp">>>> </span><span class="n">P</span> <span class="o">=</span> <span class="n">sf</span><span class="o">.</span><span class="n">load_project</span><span class="p">(</span><span class="s1">'/home/er_project'</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">dataset</span> <span class="o">=</span> <span class="n">P</span><span class="o">.</span><span class="n">dataset</span><span class="p">(</span>
<span class="gp">... </span> <span class="n">tile_px</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">tile_um</span><span class="o">=</span><span class="mi">128</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">filters</span><span class="o">=</span><span class="p">{</span>
<span class="gp">... </span> <span class="s1">'er_status_by_ihc'</span><span class="p">:</span> <span class="p">[</span><span class="s1">'Positive'</span><span class="p">,</span> <span class="s1">'Negative'</span><span class="p">]</span>
<span class="gp">... </span><span class="p">})</span>
</pre></div>
</div>
<p>If tiles have not yet been <a class="reference internal" href="../slide_processing/#filtering"><span class="std std-ref">extracted</span></a> for this dataset, do that now.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">dataset</span><span class="o">.</span><span class="n">extract_tiles</span><span class="p">(</span><span class="n">qc</span><span class="o">=</span><span class="s1">'otsu'</span><span class="p">)</span>
</pre></div>
</div>
<p>Once a dataset has been prepared, the next step in training an MIL model is <a class="reference internal" href="../mil/#mil"><span class="std std-ref">converting images into features</span></a>. For this example, we’ll use the pretrained <a class="reference external" href="https://huggingface.co/paige-ai/Virchow">Virchow</a> feature extractor, a vision transformer pretrained on 1.5M whole-slide images. Virchow has an input size of 224x224, so our images will be resized to match.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">virchow</span> <span class="o">=</span> <span class="n">sf</span><span class="o">.</span><span class="n">build_feature_extractor</span><span class="p">(</span><span class="s1">'virchow'</span><span class="p">,</span> <span class="n">center_crop</span><span class="o">=</span><span class="kc">True</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">virchow</span><span class="o">.</span><span class="n">cite</span><span class="p">()</span>
<span class="go">@misc{vorontsov2024virchowmillionslidedigitalpathology,</span>
<span class="go"> title={Virchow: A Million-Slide Digital Pathology Foundation Model},</span>
<span class="go"> author={Eugene Vorontsov and Alican Bozkurt and Adam Casson and George Shaikovski and Michal Zelechowski and Siqi Liu and Kristen Severson and Eric Zimmermann and James Hall and Neil Tenenholtz and Nicolo Fusi and Philippe Mathieu and Alexander van Eck and Donghun Lee and Julian Viret and Eric Robert and Yi Kan Wang and Jeremy D. Kunz and Matthew C. H. Lee and Jan Bernhard and Ran A. Godrich and Gerard Oakley and Ewan Millar and Matthew Hanna and Juan Retamero and William A. Moye and Razik Yousfi and Christopher Kanan and David Klimstra and Brandon Rothrock and Thomas J. Fuchs},</span>
<span class="go"> year={2024},</span>
<span class="go"> eprint={2309.07778},</span>
<span class="go"> archivePrefix={arXiv},</span>
<span class="go"> primaryClass={eess.IV},</span>
<span class="go"> url={https://arxiv.org/abs/2309.07778},</span>
<span class="go">}</span>
<span class="gp">>>> </span><span class="n">virchow</span><span class="o">.</span><span class="n">num_features</span>
<span class="go">2560</span>
</pre></div>
</div>
<p>The Virchow feature extractor produces a 2560-dimensional vector for each tile. We can generate and export <a class="reference internal" href="../features/#bags"><span class="std std-ref">bags</span></a> of these features for all slides in our dataset using <a class="reference internal" href="../project/#slideflow.Project.generate_feature_bags" title="slideflow.Project.generate_feature_bags"><code class="xref py py-func docutils literal notranslate"><span class="pre">slideflow.Project.generate_feature_bags()</span></code></a>.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">P</span><span class="o">.</span><span class="n">generate_feature_bags</span><span class="p">(</span>
<span class="gp">... </span> <span class="n">virchow</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">dataset</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">outdir</span><span class="o">=</span><span class="s1">'/bags/path'</span>
<span class="gp">... </span><span class="p">)</span>
</pre></div>
</div>
<p>The output directory, <code class="docutils literal notranslate"><span class="pre">/bags/path</span></code>, should look like:</p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>/bags/path
├──<span class="w"> </span>slide1.pt
├──<span class="w"> </span>slide1.indez.npz
├──<span class="w"> </span>slide2.pt
├──<span class="w"> </span>slide2.index.npz
├──<span class="w"> </span>...
└──<span class="w"> </span>bags_config.json
</pre></div>
</div>
<p>The <code class="docutils literal notranslate"><span class="pre">*.pt</span></code> files contain the feature vectors for tiles in each slide, and the <code class="docutils literal notranslate"><span class="pre">*.index.npz</span></code> files contain the corresponding X, Y coordinates for each tile. The <code class="docutils literal notranslate"><span class="pre">bags_config.json</span></code> file contains the feature extractor configuration.</p>
<p>The next step is to create an MIL model configuration using <a class="reference internal" href="../mil_module/#slideflow.mil.mil_config" title="slideflow.mil.mil_config"><code class="xref py py-func docutils literal notranslate"><span class="pre">slideflow.mil.mil_config()</span></code></a>, specifying the architecture and relevant hyperparameters. For the architecture, we’ll use <code class="xref py py-class docutils literal notranslate"><span class="pre">slideflow.mil.models.Attention_MIL</span></code>. For the hyperparameters, we’ll use a learning rate of 1e-4, a batch size of 32, 1cycle learning rate scheduling, and train for 10 epochs.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">slideflow.mil</span> <span class="kn">import</span> <span class="n">mil_config</span>
<span class="gp">>>> </span><span class="n">config</span> <span class="o">=</span> <span class="n">mil_config</span><span class="p">(</span>
<span class="gp">... </span> <span class="n">model</span><span class="o">=</span><span class="s1">'attention_mil'</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">lr</span><span class="o">=</span><span class="mf">1e-4</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">batch_size</span><span class="o">=</span><span class="mi">32</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">epochs</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">fit_one_cycle</span><span class="o">=</span><span class="kc">True</span>
<span class="gp">... </span><span class="p">)</span>
</pre></div>
</div>
<p>Finally, we can train the model using <a class="reference internal" href="../mil_module/#slideflow.mil.train_mil" title="slideflow.mil.train_mil"><code class="xref py py-func docutils literal notranslate"><span class="pre">slideflow.mil.train_mil()</span></code></a>. We’ll split our dataset into 70% training and 30% validation, training to the outcome “er_status_by_ihc” and saving the model to <code class="docutils literal notranslate"><span class="pre">/model/path</span></code>.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="kn">from</span> <span class="nn">slideflow.mil</span> <span class="kn">import</span> <span class="n">train_mil</span>
<span class="gp">>>> </span><span class="n">train</span><span class="p">,</span> <span class="n">val</span> <span class="o">=</span> <span class="n">dataset</span><span class="o">.</span><span class="n">split</span><span class="p">(</span><span class="n">labels</span><span class="o">=</span><span class="s1">'er_status_by_ihc'</span><span class="p">,</span> <span class="n">val_fraction</span><span class="o">=</span><span class="mf">0.3</span><span class="p">)</span>
<span class="gp">>>> </span><span class="n">train_mil</span><span class="p">(</span>
<span class="gp">... </span> <span class="n">config</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">train_dataset</span><span class="o">=</span><span class="n">train</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">val_dataset</span><span class="o">=</span><span class="n">val</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">outcomes</span><span class="o">=</span><span class="s1">'er_status_by_ihc'</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">bags</span><span class="o">=</span><span class="s1">'/bags/path'</span><span class="p">,</span>
<span class="gp">... </span> <span class="n">outdir</span><span class="o">=</span><span class="s1">'/model/path'</span>
<span class="gp">... </span><span class="p">)</span>
</pre></div>
</div>
<p>During training, you’ll see the training/validation loss and validation AUROC for each epoch. At the end of training, you’ll see the validation metrics for each outcome.</p>
<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[18:51:01] INFO Training FastAI MIL model with config:
INFO TrainerConfigFastAI(
aggregation_level='slide'
lr=0.0001
wd=1e-05
bag_size=512
fit_one_cycle=True
epochs=10
batch_size=32
model='attention_mil'
apply_softmax=True
model_kwargs=None
use_lens=True
)
[18:51:02] INFO Training dataset: 272 merged bags (from 272 possible slides)
INFO Validation dataset: 116 merged bags (from 116 possible slides)
[18:51:04] INFO Training model Attention_MIL (in=1024, out=2, loss=CrossEntropyLoss)
epoch train_loss valid_loss roc_auc_score time
0 0.328032 0.285096 0.580233 00:01
Better model found at epoch 0 with valid_loss value: 0.2850962281227112.
1 0.319219 0.266496 0.733721 00:01
Better model found at epoch 1 with valid_loss value: 0.266496479511261.
2 0.293969 0.230561 0.859690 00:01
Better model found at epoch 2 with valid_loss value: 0.23056122660636902.
3 0.266627 0.190546 0.927519 00:01
Better model found at epoch 3 with valid_loss value: 0.1905461698770523.
4 0.236985 0.165320 0.939147 00:01
Better model found at epoch 4 with valid_loss value: 0.16532012820243835.
5 0.215019 0.153572 0.946512 00:01
Better model found at epoch 5 with valid_loss value: 0.153572216629982.
6 0.199093 0.144464 0.948837 00:01
Better model found at epoch 6 with valid_loss value: 0.1444639265537262.
7 0.185597 0.141776 0.952326 00:01
Better model found at epoch 7 with valid_loss value: 0.14177580177783966.
8 0.173794 0.141409 0.951938 00:01
Better model found at epoch 8 with valid_loss value: 0.14140936732292175.
9 0.167547 0.140791 0.952713 00:01
Better model found at epoch 9 with valid_loss value: 0.14079126715660095.
[18:51:18] INFO Predictions saved to {...}/predictions.parquet
INFO Validation metrics for outcome brs_class:
[18:51:18] INFO slide-level AUC (cat # 0): 0.953 AP: 0.984 (opt. threshold: 0.544)
INFO slide-level AUC (cat # 1): 0.953 AP: 0.874 (opt. threshold: 0.458)
INFO Category 0 acc: 88.4% (76/86)
INFO Category 1 acc: 83.3% (25/30)
</pre></div>
</div>
<p>After training has completed, the output directory, <code class="docutils literal notranslate"><span class="pre">/model/path</span></code>, should look like:</p>
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>/model/path
├──<span class="w"> </span>attention
│<span class="w"> </span>├──<span class="w"> </span>slide1_att.npz
│<span class="w"> </span>└──<span class="w"> </span>...
├──<span class="w"> </span>models
│<span class="w"> </span>└──<span class="w"> </span>best_valid.pth
├──<span class="w"> </span>history.csv
├──<span class="w"> </span>mil_params.json
├──<span class="w"> </span>predictions.parquet
└──<span class="w"> </span>slide_manifest.csv
</pre></div>
</div>
<p>The final model weights are saved in <code class="docutils literal notranslate"><span class="pre">models/best_valid.pth</span></code>. Validation dataset predictions are saved in the “predictions.parquet” file. A manifest of training/validation data is saved in the “slide_manifest.csv” file, and training history is saved in the “history.csv” file. Attention values for all tiles in each slide are saved in the <code class="docutils literal notranslate"><span class="pre">attention/</span></code> directory.</p>
<p>The final saved model can be used for evaluation (<a class="reference internal" href="../mil_module/#slideflow.mil.eval_mil" title="slideflow.mil.eval_mil"><code class="xref py py-class docutils literal notranslate"><span class="pre">slideflow.mil.eval_mil</span></code></a>) or inference (<a class="reference internal" href="../mil_module/#slideflow.mil.predict_slide" title="slideflow.mil.predict_slide"><code class="xref py py-class docutils literal notranslate"><span class="pre">slideflow.mil.predict_slide</span></code></a> or <a class="reference internal" href="../studio/#studio-mil"><span class="std std-ref">Slideflow Studio</span></a>). The saved model path should be referenced by the parent directory (in this case, “/model/path”) rather than the model file itself. For more information on MIL models, see <a class="reference internal" href="../mil/#mil"><span class="std std-ref">Multiple-Instance Learning (MIL)</span></a>.</p>
</section>
</article>
</div>
<footer>
<div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
<a href="../tutorial7/" class="btn btn-neutral" title="Tutorial 7: Training with custom augmentations" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>
</div>
<hr>
<div role="contentinfo">
<p>
© Copyright 2023, James M Dolezal.
</p>
</div>
<div>
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
</div>
</footer>
</div>
</div>
<div class="pytorch-content-right" id="pytorch-content-right">
<div class="pytorch-right-menu" id="pytorch-right-menu">
<div class="pytorch-side-scroll" id="pytorch-side-scroll-right">
<ul>
<li><a class="reference internal" href="#">Tutorial 8: Multiple-Instance Learning</a></li>
</ul>
</div>
</div>
</div>
</section>
</div>
<script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
<script data-url_root="../" id="documentation_options" src="../_static/documentation_options.js"></script>
<script src="../_static/doctools.js"></script>
<script src="../_static/sphinx_highlight.js"></script>
<script type="text/javascript" src="../_static/js/vendor/jquery-3.6.3.min.js"></script>
<script type="text/javascript" src="../_static/js/vendor/popper.min.js"></script>
<script type="text/javascript" src="../_static/js/vendor/bootstrap.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/list.js/1.5.0/list.min.js"></script>
<script type="text/javascript" src="../_static/js/theme.js"></script>
<script type="text/javascript">
jQuery(function () {
SphinxRtdTheme.Navigation.enable(true);
});
</script>
<!-- Begin Footer -->
<!-- End Footer -->
<!-- Begin Mobile Menu -->
<div class="mobile-main-menu">
<div class="container-fluid">
<div class="container">
<div class="mobile-main-menu-header-container">
<a class="header-logo" href="https://pytorch.org/" aria-label="PyTorch"></a>
<a class="main-menu-close-button" href="#" data-behavior="close-mobile-menu"></a>
</div>
</div>
</div>
<div class="mobile-main-menu-links-container">
<div class="main-menu">
<ul>
<li>
<a href="https://slideflow.dev">Docs</a>
</li>
<li>
<a href="https://slideflow.dev/tutorial1/">Tutorials</a>
</li>
<li>
<a href="https://github.com/slideflow/slideflow">Github</a>
</li>
</ul>
</div>
</div>
</div>
<!-- End Mobile Menu -->
<script script type="text/javascript">
var collapsedSections = [];
</script>
<script type="text/javascript" src="../_static/js/vendor/anchor.min.js"></script>
<script type="text/javascript">
$(document).ready(function() {
mobileMenu.bind();
mobileTOC.bind();
pytorchAnchors.bind();
sideMenus.bind();
scrollToAnchor.bind();
highlightNavigation.bind();
mainMenuDropdown.bind();
filterTags.bind();
// Add class to links that have code blocks, since we cannot create links in code blocks
$("article.pytorch-article a span.pre").each(function(e) {
$(this).closest("a").addClass("has-code");
});
})
</script>
</body>
</html>
Datasets
Models
