slideflow / Git / [78ef36] /docs/custom

Models:
DanielG/
slideflow
Downloads: 1
[78ef36]: / docs / custom_extractors / index.html
History
Download this file
656 lines (468 with data), 42.8 kB

<!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>Custom Feature Extractors &mdash; 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="next" title="Strong Supervision with Tile Labels" href="../tile_labels/" />
    <link rel="prev" title="Dataloaders: Sampling and Augmentation" href="../dataloaders/" />




  <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 class="current">
<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 current"><a class="current reference internal" href="#">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>
<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 &amp; 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"><a class="reference internal" href="../tutorial8/">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> &gt;
      </li>


      <li>Custom Feature Extractors</li>


      <li class="pytorch-breadcrumbs-aside">


            <a href="../_sources/custom_extractors.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="custom-feature-extractors">
<span id="custom-extractors"></span><h1>Custom Feature Extractors<a class="headerlink" href="#custom-feature-extractors" title="Permalink to this heading">¶</a></h1>
<p>Slideflow includes several <a class="reference internal" href="../mil/#mil"><span class="std std-ref">pretrained feature extractors</span></a> for converting image tiles into feature vectors as well as tools to assist with building your own feature extractor. In this note, we’ll walk through the process of building a custom feature extractor from both a PyTorch and Tensorflow model.</p>
<section id="pytorch">
<h2>PyTorch<a class="headerlink" href="#pytorch" title="Permalink to this heading">¶</a></h2>
<p>Feature extractors are implemented as a subclass of <code class="xref py py-class docutils literal notranslate"><span class="pre">slideflow.model.extractors._factory_torch.TorchFeatureExtractor</span></code>. The base class provides core functionality and helper methods for generating features from image tiles (dtype uint8) or whole-slide images (type <code class="xref py py-class docutils literal notranslate"><span class="pre">slideflow.WSI</span></code>).</p>
<p>The initializer should create the feature extraction model and move it to the appropriate device (<em>i.e.</em> GPU). The model should be a <code class="xref py py-class docutils literal notranslate"><span class="pre">torch.nn.Module</span></code> that accepts an image tensor as input and returns a feature tensor as output.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># Import your custom torch.nn.Module,</span>
<span class="c1"># which generates features from an image.</span>
<span class="kn">from</span> <span class="nn">my_module</span> <span class="kn">import</span> <span class="n">MyModel</span>

<span class="kn">from</span> <span class="nn">slideflow.model.extractors._factory_torch</span> <span class="kn">import</span> <span class="n">TorchFeatureExtractor</span>

<span class="k">class</span> <span class="nc">MyFeatureExtractor</span><span class="p">(</span><span class="n">TorchFeatureExtractor</span><span class="p">):</span>

    <span class="n">tag</span> <span class="o">=</span> <span class="s1">&#39;my_feature_extractor&#39;</span>  <span class="c1"># Human-readable identifier</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>

        <span class="c1"># Create the device, move to GPU, and set in evaluation mode.</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">MyModel</span><span class="p">()</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="s1">&#39;cuda&#39;</span><span class="p">)</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
</pre></div>
</div>
<p>Next, the initializer should set the number of features expected to be returned by the model.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="o">...</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="o">...</span>

        <span class="bp">self</span><span class="o">.</span><span class="n">num_features</span> <span class="o">=</span> <span class="mi">1024</span>
</pre></div>
</div>
<p>The initializer is also responsible for registering image preprocessing. The image preprocessing transformation, a function which converts a raw <code class="docutils literal notranslate"><span class="pre">uint8</span></code> image to a <code class="docutils literal notranslate"><span class="pre">float32</span></code> tensor for model input, should be stored in <code class="docutils literal notranslate"><span class="pre">self.transform</span></code>. If the transformation standardizes the images, then the parameter <code class="docutils literal notranslate"><span class="pre">self.preprocess_kwargs</span></code> should be set to <code class="docutils literal notranslate"><span class="pre">{'standardize':</span> <span class="pre">False}</span></code>, indicating that Slideflow should not perform any additional standardization. You can use the class method <code class="docutils literal notranslate"><span class="pre">.build_transform()</span></code> to use the standard preprocessing pipeline.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">transforms</span>

<span class="o">...</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="o">...</span>

        <span class="c1"># Image preprocessing.</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">build_transform</span><span class="p">(</span><span class="n">img_size</span><span class="o">=</span><span class="mi">256</span><span class="p">)</span>
        <span class="c1"># Disable Slideflow standardization,</span>
        <span class="c1"># as we are standardizing with transforms.Normalize</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">preprocess_kwargs</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;standardize&#39;</span><span class="p">:</span> <span class="kc">False</span><span class="p">}</span>
</pre></div>
</div>
<p>The final required method is <code class="docutils literal notranslate"><span class="pre">.dump_config()</span></code>, which returns a dictionary of configuration parameters needed to regenerate this class. It should return a dictionary with <code class="docutils literal notranslate"><span class="pre">&quot;class&quot;</span></code> and <code class="docutils literal notranslate"><span class="pre">&quot;kwargs&quot;</span></code> attributes. This configuration is saved to a JSON configuration file when generating bags for MIL training.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="o">...</span>

    <span class="k">def</span> <span class="nf">dump_config</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_dump_config</span><span class="p">(</span>
            <span class="n">class_name</span><span class="o">=</span><span class="s1">&#39;my_module.MyFeatureExtractor&#39;</span>
        <span class="p">)</span>
</pre></div>
</div>
<p>The final class should look like this:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">my_module</span> <span class="kn">import</span> <span class="n">MyModel</span>
<span class="kn">from</span> <span class="nn">slideflow.model.extractors._factory_torch</span> <span class="kn">import</span> <span class="n">TorchFeatureExtractor</span>
<span class="kn">from</span> <span class="nn">torchvision</span> <span class="kn">import</span> <span class="n">transforms</span>

<span class="k">class</span> <span class="nc">MyFeatureExtractor</span><span class="p">(</span><span class="n">TorchFeatureExtractor</span><span class="p">):</span>

    <span class="n">tag</span> <span class="o">=</span> <span class="s1">&#39;my_feature_extractor&#39;</span>  <span class="c1"># Human-readable identifier</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>

        <span class="c1"># Create the device, move to GPU, and set in evaluation mode.</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">MyModel</span><span class="p">()</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">to</span><span class="p">(</span><span class="s1">&#39;cuda&#39;</span><span class="p">)</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">model</span><span class="o">.</span><span class="n">eval</span><span class="p">()</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">num_features</span> <span class="o">=</span> <span class="mi">1024</span>

        <span class="c1"># Image preprocessing.</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">transform</span> <span class="o">=</span> <span class="bp">self</span><span class="o">.</span><span class="n">build_transform</span><span class="p">(</span><span class="n">img_size</span><span class="o">=</span><span class="mi">256</span><span class="p">)</span>
        <span class="c1"># Disable Slideflow standardization,</span>
        <span class="c1"># as we are standardizing with transforms.Normalize</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">preprocess_kwargs</span> <span class="o">=</span> <span class="p">{</span><span class="s1">&#39;standardize&#39;</span><span class="p">:</span> <span class="kc">False</span><span class="p">}</span>

    <span class="k">def</span> <span class="nf">dump_config</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_dump_config</span><span class="p">(</span>
            <span class="n">class_name</span><span class="o">=</span><span class="s1">&#39;my_module.MyFeatureExtractor&#39;</span>
        <span class="p">)</span>
</pre></div>
</div>
<p>You can then use the feature extractor for generating bags for MIL training, as described in <a class="reference internal" href="../mil/#mil"><span class="std std-ref">Multiple-Instance Learning (MIL)</span></a>.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="c1"># Build the feature extractor.</span>
<span class="n">myfeatures</span> <span class="o">=</span> <span class="n">MyFeatureExtractor</span><span class="p">()</span>

<span class="c1"># Load a dataset.</span>
<span class="n">project</span> <span class="o">=</span> <span class="n">slideflow</span><span class="o">.</span><span class="n">load_project</span><span class="p">(</span><span class="o">...</span><span class="p">)</span>
<span class="n">dataset</span> <span class="o">=</span> <span class="n">project</span><span class="o">.</span><span class="n">dataset</span><span class="p">(</span><span class="o">...</span><span class="p">)</span>

<span class="c1"># Generate bags.</span>
<span class="n">project</span><span class="o">.</span><span class="n">generate_feature_bags</span><span class="p">(</span><span class="n">myfeatures</span><span class="p">,</span> <span class="n">dataset</span><span class="p">)</span>
</pre></div>
</div>
<p>You can also generate features across whole-slide images, returning a grid of features for each slide. The size of the returned grid reflects the slide’s tile grid. For example, for a slide with 24 columns and 33 rows of tiles, the returned grid will have shape <code class="docutils literal notranslate"><span class="pre">(24,</span> <span class="pre">33,</span> <span class="pre">n_features)</span></code>.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">myfeatures</span> <span class="o">=</span> <span class="n">MyFeatureExtractor</span><span class="p">()</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">wsi</span> <span class="o">=</span> <span class="n">sf</span><span class="o">.</span><span class="n">WSI</span><span class="p">(</span><span class="s1">&#39;path/to/wsi&#39;</span><span class="p">,</span> <span class="n">tile_px</span><span class="o">=</span><span class="mi">256</span><span class="p">,</span> <span class="n">tile_um</span><span class="o">=</span><span class="mi">302</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">features</span> <span class="o">=</span> <span class="n">myfeatures</span><span class="p">(</span><span class="n">wsi</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">features</span><span class="o">.</span><span class="n">shape</span>
<span class="go">(24, 33, 1024)</span>
</pre></div>
</div>
<p>Finally, the feature extractor can also be used to perform latent space analysis and generate mosaic maps, as described in <a class="reference internal" href="../posthoc/#activations"><span class="std std-ref">Layer Activations</span></a>.</p>
<p>Slideflow includes a registration system for keeping track of all available feature extractors. To register your feature extractor, use the <code class="xref py py-func docutils literal notranslate"><span class="pre">slideflow.model.extractors.register_torch()</span></code> decorator.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">slideflow.model.extractors</span> <span class="kn">import</span> <span class="n">register_torch</span>

<span class="nd">@register_torch</span>
<span class="k">def</span> <span class="nf">my_feature_extractor</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
    <span class="k">return</span> <span class="n">MyFeatureExtractor</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
</pre></div>
</div>
<p>Once registered, a feature extractor can be built by name:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">slideflow</span> <span class="k">as</span> <span class="nn">sf</span>
<span class="n">extractor</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">&#39;my_feature_extractor&#39;</span><span class="p">)</span>
</pre></div>
</div>
</section>
<section id="tensorflow">
<h2>Tensorflow<a class="headerlink" href="#tensorflow" title="Permalink to this heading">¶</a></h2>
<p>Tensorflow feature extractors are implemented very similarly to PyTorch feature extractors, extended from <code class="xref py py-class docutils literal notranslate"><span class="pre">slideflow.model.extractors._tensorflow_base.TensorflowFeatureExtractor</span></code>.</p>
<p>The initializer should create the model and set the expected number of features.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">my_module</span> <span class="kn">import</span> <span class="n">MyModel</span>
<span class="kn">from</span> <span class="nn">slideflow.model.extractors._tensorflow_base</span> <span class="kn">import</span> <span class="n">TensorflowFeatureExtractor</span>

<span class="k">class</span> <span class="nc">MyFeatureExtractor</span><span class="p">(</span><span class="n">TensorflowFeatureExtractor</span><span class="p">):</span>

    <span class="n">tag</span> <span class="o">=</span> <span class="s1">&#39;my_feature_extractor&#39;</span>  <span class="c1"># Unique identifier</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>

        <span class="c1"># Create the model.</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">MyModel</span><span class="p">()</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">num_features</span> <span class="o">=</span> <span class="mi">1024</span>
</pre></div>
</div>
<p>The initializer is also responsible for registering image preprocessing and transformations. Preprocessing steps are stored in the <code class="docutils literal notranslate"><span class="pre">.preprocess_kwargs</span></code> dictionary, which should have the keys <code class="docutils literal notranslate"><span class="pre">standardize</span></code> and <code class="docutils literal notranslate"><span class="pre">transform</span></code>. If <code class="docutils literal notranslate"><span class="pre">standardize=True</span></code>, images will be standardized using <a class="reference external" href="https://www.tensorflow.org/api_docs/python/tf/image/per_image_standardization"><code class="docutils literal notranslate"><span class="pre">tf.image.per_image_standardization</span></code></a>. If <code class="docutils literal notranslate"><span class="pre">transform</span></code> is not None, it should be a callable that accepts a single image tensor and returns a transformed image tensor.</p>
<p>For example, to only perform standardization and no further preprocessing:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="o">...</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="o">...</span>

        <span class="c1"># Image preprocessing.</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">preprocess_kwargs</span> <span class="o">=</span> <span class="p">{</span>
            <span class="s1">&#39;standardize&#39;</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
            <span class="s1">&#39;transform&#39;</span><span class="p">:</span> <span class="kc">None</span>
        <span class="p">}</span>
</pre></div>
</div>
<p>To perform standardization and resize images to 256x256:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">tensorflow</span> <span class="k">as</span> <span class="nn">tf</span>

<span class="nd">@tf</span><span class="o">.</span><span class="n">function</span>
<span class="k">def</span> <span class="nf">resize_256</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
    <span class="k">return</span> <span class="o">=</span> <span class="n">tf</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">resize</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="p">(</span><span class="n">resize_px</span><span class="p">,</span> <span class="n">resize_px</span><span class="p">))</span>

<span class="o">...</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="o">...</span>

        <span class="c1"># Image preprocessing.</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">preprocess_kwargs</span> <span class="o">=</span> <span class="p">{</span>
            <span class="s1">&#39;standardize&#39;</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
            <span class="s1">&#39;transform&#39;</span><span class="p">:</span> <span class="n">resize_256</span>
        <span class="p">}</span>
</pre></div>
</div>
<p>The <code class="docutils literal notranslate"><span class="pre">.dump_config()</span></code> method should then be set, which is expected to return a dictionary of configuration parameters needed to regenerate this class. It should return a dictionary with <code class="docutils literal notranslate"><span class="pre">&quot;class&quot;</span></code> and <code class="docutils literal notranslate"><span class="pre">&quot;kwargs&quot;</span></code> attributes. This configuration is saved to a JSON configuration file when generating bags for MIL training.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="o">...</span>

    <span class="k">def</span> <span class="nf">dump_config</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">return</span> <span class="p">{</span>
            <span class="s1">&#39;class&#39;</span><span class="p">:</span> <span class="s1">&#39;MyFeatureExtractor&#39;</span><span class="p">,</span>
            <span class="s1">&#39;kwargs&#39;</span><span class="p">:</span> <span class="p">{}</span>
        <span class="p">}</span>
</pre></div>
</div>
<p>The final class should look like this:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">my_module</span> <span class="kn">import</span> <span class="n">MyModel</span>
<span class="kn">from</span> <span class="nn">slideflow.model.extractors._tensorflow_base</span> <span class="kn">import</span> <span class="n">TensorflowFeatureExtractor</span>

<span class="k">class</span> <span class="nc">MyFeatureExtractor</span><span class="p">(</span><span class="n">TensorflowFeatureExtractor</span><span class="p">):</span>

    <span class="n">tag</span> <span class="o">=</span> <span class="s1">&#39;my_feature_extractor&#39;</span>  <span class="c1"># Unique identifier</span>

    <span class="k">def</span> <span class="fm">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">()</span>

        <span class="c1"># Create the model.</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">model</span> <span class="o">=</span> <span class="n">MyModel</span><span class="p">()</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">num_features</span> <span class="o">=</span> <span class="mi">1024</span>

        <span class="c1"># Image preprocessing.</span>
        <span class="bp">self</span><span class="o">.</span><span class="n">preprocess_kwargs</span> <span class="o">=</span> <span class="p">{</span>
            <span class="s1">&#39;standardize&#39;</span><span class="p">:</span> <span class="kc">True</span><span class="p">,</span>
            <span class="s1">&#39;transform&#39;</span><span class="p">:</span> <span class="kc">None</span>
        <span class="p">}</span>

    <span class="k">def</span> <span class="nf">dump_config</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
        <span class="k">return</span> <span class="p">{</span>
            <span class="s1">&#39;class&#39;</span><span class="p">:</span> <span class="s1">&#39;MyFeatureExtractor&#39;</span><span class="p">,</span>
            <span class="s1">&#39;kwargs&#39;</span><span class="p">:</span> <span class="p">{}</span>
        <span class="p">}</span>
</pre></div>
</div>
<p>As described above, this feature extractor can then be used to create bags for MIL training, generate features across whole-slide images, or perform feature space analysis across a dataset.</p>
<p>To register your feature extractor, use the <code class="xref py py-func docutils literal notranslate"><span class="pre">slideflow.model.extractors.register_tensorflow()</span></code> decorator.</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">from</span> <span class="nn">slideflow.model.extractors</span> <span class="kn">import</span> <span class="n">register_tf</span>

<span class="nd">@register_tf</span>
<span class="k">def</span> <span class="nf">my_feature_extractor</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">):</span>
    <span class="k">return</span> <span class="n">MyFeatureExtractor</span><span class="p">(</span><span class="o">**</span><span class="n">kwargs</span><span class="p">)</span>
</pre></div>
</div>
<p>…which will allow the feature extractor to be built by name:</p>
<div class="highlight-python notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">slideflow</span> <span class="k">as</span> <span class="nn">sf</span>
<span class="n">extractor</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">&#39;my_feature_extractor&#39;</span><span class="p">)</span>
</pre></div>
</div>
</section>
</section>


             </article>

            </div>
            <footer>

    <div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">

        <a href="../tile_labels/" class="btn btn-neutral float-right" title="Strong Supervision with Tile Labels" accesskey="n" rel="next">Next <img src="../_static/images/chevron-right-orange.svg" class="next-page"></a>


        <a href="../dataloaders/" class="btn btn-neutral" title="Dataloaders: Sampling and Augmentation" accesskey="p" rel="prev"><img src="../_static/images/chevron-right-orange.svg" class="previous-page"> Previous</a>

    </div>




    <hr>



  <div role="contentinfo">
    <p>
        &copy; 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="#">Custom Feature Extractors</a><ul>
<li><a class="reference internal" href="#pytorch">PyTorch</a></li>
<li><a class="reference internal" href="#tensorflow">Tensorflow</a></li>
</ul>
</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>