dev/modules/metrics.html


<!DOCTYPE html>


<html lang="en" data-content_root="../" >

  <head>
    <meta charset="utf-8" />
    <meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
<meta property="og:title" content="7.8. Pairwise metrics, Affinities and Kernels" />
<meta property="og:type" content="website" />
<meta property="og:url" content="https://scikit-learn/stable/modules/metrics.html" />
<meta property="og:site_name" content="scikit-learn" />
<meta property="og:description" content="The sklearn.metrics.pairwise submodule implements utilities to evaluate pairwise distances or affinity of sets of samples. This module contains both distance metrics and kernels. A brief summary is..." />
<meta property="og:image" content="https://scikit-learn.org/stable/_static/scikit-learn-logo-small.png" />
<meta property="og:image:alt" content="scikit-learn" />
<meta name="description" content="The sklearn.metrics.pairwise submodule implements utilities to evaluate pairwise distances or affinity of sets of samples. This module contains both distance metrics and kernels. A brief summary is..." />

    <title>7.8. Pairwise metrics, Affinities and Kernels &#8212; scikit-learn 1.7.dev0 documentation</title>
  
  
  <script data-cfasync="false">
    document.documentElement.dataset.mode = localStorage.getItem("mode") || "";
    document.documentElement.dataset.theme = localStorage.getItem("theme") || "";
  </script>
  <!--
    this give us a css class that will be invisible only if js is disabled
  -->
  <noscript>
    <style>
      .pst-js-only { display: none !important; }

    </style>
  </noscript>
  
  <!-- Loaded before other Sphinx assets -->
  <link href="../_static/styles/theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />
<link href="../_static/styles/pydata-sphinx-theme.css?digest=8878045cc6db502f8baf" rel="stylesheet" />

    <link rel="stylesheet" type="text/css" href="../_static/pygments.css?v=8f2a1f02" />
    <link rel="stylesheet" type="text/css" href="../_static/copybutton.css?v=76b2166b" />
    <link rel="stylesheet" type="text/css" href="../_static/plot_directive.css" />
    <link rel="stylesheet" type="text/css" href="https://fonts.googleapis.com/css?family=Vibur" />
    <link rel="stylesheet" type="text/css" href="../_static/jupyterlite_sphinx.css?v=2c9f8f05" />
    <link rel="stylesheet" type="text/css" href="../_static/sg_gallery.css?v=d2d258e8" />
    <link rel="stylesheet" type="text/css" href="../_static/sg_gallery-binder.css?v=f4aeca0c" />
    <link rel="stylesheet" type="text/css" href="../_static/sg_gallery-dataframe.css?v=2082cf3c" />
    <link rel="stylesheet" type="text/css" href="../_static/sg_gallery-rendered-html.css?v=1277b6f3" />
    <link rel="stylesheet" type="text/css" href="../_static/sphinx-design.min.css?v=95c83b7e" />
    <link rel="stylesheet" type="text/css" href="../_static/styles/colors.css?v=cc94ab7d" />
    <link rel="stylesheet" type="text/css" href="../_static/styles/custom.css?v=8f525996" />
  
  <!-- So that users can add custom icons -->
  <script src="../_static/scripts/fontawesome.js?digest=8878045cc6db502f8baf"></script>
  <!-- Pre-loaded scripts that we'll load fully later -->
  <link rel="preload" as="script" href="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf" />
<link rel="preload" as="script" href="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf" />

    <script src="../_static/documentation_options.js?v=473747f4"></script>
    <script src="../_static/doctools.js?v=9bcbadda"></script>
    <script src="../_static/sphinx_highlight.js?v=dc90522c"></script>
    <script src="../_static/clipboard.min.js?v=a7894cd8"></script>
    <script src="../_static/copybutton.js?v=97f0b27d"></script>
    <script src="../_static/jupyterlite_sphinx.js?v=96e329c5"></script>
    <script src="../_static/design-tabs.js?v=f930bc37"></script>
    <script data-domain="scikit-learn.org" defer="defer" src="https://views.scientific-python.org/js/script.js"></script>
    <script async="async" src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-chtml.js"></script>
    <script>DOCUMENTATION_OPTIONS.pagename = 'modules/metrics';</script>
    <script>
        DOCUMENTATION_OPTIONS.theme_version = '0.16.1';
        DOCUMENTATION_OPTIONS.theme_switcher_json_url = 'https://scikit-learn.org/dev/_static/versions.json';
        DOCUMENTATION_OPTIONS.theme_switcher_version_match = '1.7.dev0';
        DOCUMENTATION_OPTIONS.show_version_warning_banner =
            true;
        </script>
    <script src="../_static/scripts/dropdown.js?v=d6825577"></script>
    <script src="../_static/scripts/version-switcher.js?v=a6dd8357"></script>
    <script src="../_static/scripts/sg_plotly_resize.js?v=2167d4db"></script>
    <link rel="canonical" href="https://scikit-learn.org/stable/modules/metrics.html" />
    <link rel="icon" href="../_static/favicon.ico"/>
    <link rel="author" title="About these documents" href="../about.html" />
    <link rel="search" title="Search" href="../search.html" />
    <link rel="next" title="7.9. Transforming the prediction target (y)" href="preprocessing_targets.html" />
    <link rel="prev" title="7.7. Kernel Approximation" href="kernel_approximation.html" />
  <meta name="viewport" content="width=device-width, initial-scale=1"/>
  <meta name="docsearch:language" content="en"/>
  <meta name="docsearch:version" content="1.7" />
  </head>
  
  
  <body data-bs-spy="scroll" data-bs-target=".bd-toc-nav" data-offset="180" data-bs-root-margin="0px 0px -60%" data-default-mode="">

  
  <div id="pst-skip-link" class="skip-link d-print-none"><a href="#main-content">Skip to main content</a></div>
  
  <div id="pst-scroll-pixel-helper"></div>
  
  <button type="button" class="btn rounded-pill" id="pst-back-to-top">
    <i class="fa-solid fa-arrow-up"></i>Back to top</button>

  
  <dialog id="pst-search-dialog">
    
<form class="bd-search d-flex align-items-center"
      action="../search.html"
      method="get">
  <i class="fa-solid fa-magnifying-glass"></i>
  <input type="search"
         class="form-control"
         name="q"
         placeholder="Search the docs ..."
         aria-label="Search the docs ..."
         autocomplete="off"
         autocorrect="off"
         autocapitalize="off"
         spellcheck="false"/>
  <span class="search-button__kbd-shortcut"><kbd class="kbd-shortcut__modifier">Ctrl</kbd>+<kbd>K</kbd></span>
</form>
  </dialog>

  <div class="pst-async-banner-revealer d-none">
  <aside id="bd-header-version-warning" class="d-none d-print-none" aria-label="Version warning"></aside>
</div>

  
    <header class="bd-header navbar navbar-expand-lg bd-navbar d-print-none">
<div class="bd-header__inner bd-page-width">
  <button class="pst-navbar-icon sidebar-toggle primary-toggle" aria-label="Site navigation">
    <span class="fa-solid fa-bars"></span>
  </button>
  
  
  <div class=" navbar-header-items__start">
    
      <div class="navbar-item">

  
<a class="navbar-brand logo" href="../index.html">
  
  
    <img src="../_static/scikit-learn-logo-small.png" class="logo__image only-light" alt="scikit-learn homepage"/>
    <img src="../_static/scikit-learn-logo-small.png" class="logo__image only-dark pst-js-only" alt="scikit-learn homepage"/>
  
  
</a></div>
    
  </div>
  
  <div class=" navbar-header-items">
    
    <div class="me-auto navbar-header-items__center">
      
        <div class="navbar-item">
<nav>
  <ul class="bd-navbar-elements navbar-nav">
    
<li class="nav-item ">
  <a class="nav-link nav-internal" href="../install.html">
    Install
  </a>
</li>


<li class="nav-item current active">
  <a class="nav-link nav-internal" href="../user_guide.html">
    User Guide
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../api/index.html">
    API
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../auto_examples/index.html">
    Examples
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-external" href="https://blog.scikit-learn.org/">
    Community
  </a>
</li>

            <li class="nav-item dropdown">
                <button class="btn dropdown-toggle nav-item" type="button"
                data-bs-toggle="dropdown" aria-expanded="false"
                aria-controls="pst-nav-more-links">
                    More
                </button>
                <ul id="pst-nav-more-links" class="dropdown-menu">
                    
<li class=" ">
  <a class="nav-link dropdown-item nav-internal" href="../getting_started.html">
    Getting Started
  </a>
</li>


<li class=" ">
  <a class="nav-link dropdown-item nav-internal" href="../whats_new.html">
    Release History
  </a>
</li>


<li class=" ">
  <a class="nav-link dropdown-item nav-internal" href="../glossary.html">
    Glossary
  </a>
</li>


<li class=" ">
  <a class="nav-link dropdown-item nav-internal" href="../developers/index.html">
    Development
  </a>
</li>


<li class=" ">
  <a class="nav-link dropdown-item nav-internal" href="../faq.html">
    FAQ
  </a>
</li>


<li class=" ">
  <a class="nav-link dropdown-item nav-internal" href="../support.html">
    Support
  </a>
</li>


<li class=" ">
  <a class="nav-link dropdown-item nav-internal" href="../related_projects.html">
    Related Projects
  </a>
</li>


<li class=" ">
  <a class="nav-link dropdown-item nav-internal" href="../roadmap.html">
    Roadmap
  </a>
</li>


<li class=" ">
  <a class="nav-link dropdown-item nav-internal" href="../governance.html">
    Governance
  </a>
</li>


<li class=" ">
  <a class="nav-link dropdown-item nav-internal" href="../about.html">
    About us
  </a>
</li>

                </ul>
            </li>
            
  </ul>
</nav></div>
      
    </div>
    
    
    <div class="navbar-header-items__end">
      
        <div class="navbar-item navbar-persistent--container">
          

<button class="btn btn-sm pst-navbar-icon search-button search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
    <i class="fa-solid fa-magnifying-glass fa-lg"></i>
</button>
        </div>
      
      
        <div class="navbar-item">

<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
</button></div>
      
        <div class="navbar-item"><ul class="navbar-icon-links"
    aria-label="Icon Links">
        <li class="nav-item">
          
          
          <a href="https://github.com/scikit-learn/scikit-learn" title="GitHub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-square-github fa-lg" aria-hidden="true"></i>
            <span class="sr-only">GitHub</span></a>
        </li>
</ul></div>
      
        <div class="navbar-item">
<div class="version-switcher__container dropdown pst-js-only">
  <button id="pst-version-switcher-button-2"
    type="button"
    class="version-switcher__button btn btn-sm dropdown-toggle"
    data-bs-toggle="dropdown"
    aria-haspopup="listbox"
    aria-controls="pst-version-switcher-list-2"
    aria-label="Version switcher list"
  >
    Choose version  <!-- this text may get changed later by javascript -->
    <span class="caret"></span>
  </button>
  <div id="pst-version-switcher-list-2"
    class="version-switcher__menu dropdown-menu list-group-flush py-0"
    role="listbox" aria-labelledby="pst-version-switcher-button-2">
    <!-- dropdown will be populated by javascript on page load -->
  </div>
</div></div>
      
    </div>
    
  </div>
  
  
    <div class="navbar-persistent--mobile">

<button class="btn btn-sm pst-navbar-icon search-button search-button__button pst-js-only" title="Search" aria-label="Search" data-bs-placement="bottom" data-bs-toggle="tooltip">
    <i class="fa-solid fa-magnifying-glass fa-lg"></i>
</button>
    </div>
  

    <button class="pst-navbar-icon sidebar-toggle secondary-toggle" aria-label="On this page">
      <span class="fa-solid fa-outdent"></span>
    </button>
  
</div>

    </header>
  

  <div class="bd-container">
    <div class="bd-container__inner bd-page-width">
      
      
      <dialog id="pst-primary-sidebar-modal"></dialog>
      <div id="pst-primary-sidebar" class="bd-sidebar-primary bd-sidebar">
        

  <div class="sidebar-header-items sidebar-primary__section">
    
    
      <div class="sidebar-header-items__center">
        
          
            <div class="navbar-item">
<nav>
  <ul class="bd-navbar-elements navbar-nav">
    
<li class="nav-item ">
  <a class="nav-link nav-internal" href="../install.html">
    Install
  </a>
</li>


<li class="nav-item current active">
  <a class="nav-link nav-internal" href="../user_guide.html">
    User Guide
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../api/index.html">
    API
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../auto_examples/index.html">
    Examples
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-external" href="https://blog.scikit-learn.org/">
    Community
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../getting_started.html">
    Getting Started
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../whats_new.html">
    Release History
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../glossary.html">
    Glossary
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../developers/index.html">
    Development
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../faq.html">
    FAQ
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../support.html">
    Support
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../related_projects.html">
    Related Projects
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../roadmap.html">
    Roadmap
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../governance.html">
    Governance
  </a>
</li>


<li class="nav-item ">
  <a class="nav-link nav-internal" href="../about.html">
    About us
  </a>
</li>

  </ul>
</nav></div>
          
        
      </div>
    
    
      <div class="sidebar-header-items__end">
        
          <div class="navbar-item">

<button class="btn btn-sm nav-link pst-navbar-icon theme-switch-button pst-js-only" aria-label="Color mode" data-bs-title="Color mode"  data-bs-placement="bottom" data-bs-toggle="tooltip">
  <i class="theme-switch fa-solid fa-sun                fa-lg" data-mode="light" title="Light"></i>
  <i class="theme-switch fa-solid fa-moon               fa-lg" data-mode="dark"  title="Dark"></i>
  <i class="theme-switch fa-solid fa-circle-half-stroke fa-lg" data-mode="auto"  title="System Settings"></i>
</button></div>
        
          <div class="navbar-item"><ul class="navbar-icon-links"
    aria-label="Icon Links">
        <li class="nav-item">
          
          
          <a href="https://github.com/scikit-learn/scikit-learn" title="GitHub" class="nav-link pst-navbar-icon" rel="noopener" target="_blank" data-bs-toggle="tooltip" data-bs-placement="bottom"><i class="fa-brands fa-square-github fa-lg" aria-hidden="true"></i>
            <span class="sr-only">GitHub</span></a>
        </li>
</ul></div>
        
          <div class="navbar-item">
<div class="version-switcher__container dropdown pst-js-only">
  <button id="pst-version-switcher-button-3"
    type="button"
    class="version-switcher__button btn btn-sm dropdown-toggle"
    data-bs-toggle="dropdown"
    aria-haspopup="listbox"
    aria-controls="pst-version-switcher-list-3"
    aria-label="Version switcher list"
  >
    Choose version  <!-- this text may get changed later by javascript -->
    <span class="caret"></span>
  </button>
  <div id="pst-version-switcher-list-3"
    class="version-switcher__menu dropdown-menu list-group-flush py-0"
    role="listbox" aria-labelledby="pst-version-switcher-button-3">
    <!-- dropdown will be populated by javascript on page load -->
  </div>
</div></div>
        
      </div>
    
  </div>
  
    <div class="sidebar-primary-items__start sidebar-primary__section">
        <div class="sidebar-primary-item">
<nav class="bd-docs-nav bd-links"
     aria-label="Section Navigation">
  <p class="bd-links__title" role="heading" aria-level="1">Section Navigation</p>
  <div class="bd-toc-item navbar-nav"><ul class="current nav bd-sidenav">
<li class="toctree-l1 has-children"><a class="reference internal" href="../supervised_learning.html">1. Supervised learning</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="linear_model.html">1.1. Linear Models</a></li>
<li class="toctree-l2"><a class="reference internal" href="lda_qda.html">1.2. Linear and Quadratic Discriminant Analysis</a></li>
<li class="toctree-l2"><a class="reference internal" href="kernel_ridge.html">1.3. Kernel ridge regression</a></li>
<li class="toctree-l2"><a class="reference internal" href="svm.html">1.4. Support Vector Machines</a></li>
<li class="toctree-l2"><a class="reference internal" href="sgd.html">1.5. Stochastic Gradient Descent</a></li>
<li class="toctree-l2"><a class="reference internal" href="neighbors.html">1.6. Nearest Neighbors</a></li>
<li class="toctree-l2"><a class="reference internal" href="gaussian_process.html">1.7. Gaussian Processes</a></li>
<li class="toctree-l2"><a class="reference internal" href="cross_decomposition.html">1.8. Cross decomposition</a></li>
<li class="toctree-l2"><a class="reference internal" href="naive_bayes.html">1.9. Naive Bayes</a></li>
<li class="toctree-l2"><a class="reference internal" href="tree.html">1.10. Decision Trees</a></li>
<li class="toctree-l2"><a class="reference internal" href="ensemble.html">1.11. Ensembles: Gradient boosting, random forests, bagging, voting, stacking</a></li>
<li class="toctree-l2"><a class="reference internal" href="multiclass.html">1.12. Multiclass and multioutput algorithms</a></li>
<li class="toctree-l2"><a class="reference internal" href="feature_selection.html">1.13. Feature selection</a></li>
<li class="toctree-l2"><a class="reference internal" href="semi_supervised.html">1.14. Semi-supervised learning</a></li>
<li class="toctree-l2"><a class="reference internal" href="isotonic.html">1.15. Isotonic regression</a></li>
<li class="toctree-l2"><a class="reference internal" href="calibration.html">1.16. Probability calibration</a></li>
<li class="toctree-l2"><a class="reference internal" href="neural_networks_supervised.html">1.17. Neural network models (supervised)</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../unsupervised_learning.html">2. Unsupervised learning</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="mixture.html">2.1. Gaussian mixture models</a></li>
<li class="toctree-l2"><a class="reference internal" href="manifold.html">2.2. Manifold learning</a></li>
<li class="toctree-l2"><a class="reference internal" href="clustering.html">2.3. Clustering</a></li>
<li class="toctree-l2"><a class="reference internal" href="biclustering.html">2.4. Biclustering</a></li>
<li class="toctree-l2"><a class="reference internal" href="decomposition.html">2.5. Decomposing signals in components (matrix factorization problems)</a></li>
<li class="toctree-l2"><a class="reference internal" href="covariance.html">2.6. Covariance estimation</a></li>
<li class="toctree-l2"><a class="reference internal" href="outlier_detection.html">2.7. Novelty and Outlier Detection</a></li>
<li class="toctree-l2"><a class="reference internal" href="density.html">2.8. Density Estimation</a></li>
<li class="toctree-l2"><a class="reference internal" href="neural_networks_unsupervised.html">2.9. Neural network models (unsupervised)</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../model_selection.html">3. Model selection and evaluation</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="cross_validation.html">3.1. Cross-validation: evaluating estimator performance</a></li>
<li class="toctree-l2"><a class="reference internal" href="grid_search.html">3.2. Tuning the hyper-parameters of an estimator</a></li>
<li class="toctree-l2"><a class="reference internal" href="classification_threshold.html">3.3. Tuning the decision threshold for class prediction</a></li>
<li class="toctree-l2"><a class="reference internal" href="model_evaluation.html">3.4. Metrics and scoring: quantifying the quality of predictions</a></li>
<li class="toctree-l2"><a class="reference internal" href="learning_curve.html">3.5. Validation curves: plotting scores to evaluate models</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../metadata_routing.html">4. Metadata Routing</a></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../inspection.html">5. Inspection</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="partial_dependence.html">5.1. Partial Dependence and Individual Conditional Expectation plots</a></li>
<li class="toctree-l2"><a class="reference internal" href="permutation_importance.html">5.2. Permutation feature importance</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../visualizations.html">6. Visualizations</a></li>
<li class="toctree-l1 current active has-children"><a class="reference internal" href="../data_transforms.html">7. Dataset transformations</a><details open="open"><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul class="current">
<li class="toctree-l2"><a class="reference internal" href="compose.html">7.1. Pipelines and composite estimators</a></li>
<li class="toctree-l2"><a class="reference internal" href="feature_extraction.html">7.2. Feature extraction</a></li>
<li class="toctree-l2"><a class="reference internal" href="preprocessing.html">7.3. Preprocessing data</a></li>
<li class="toctree-l2"><a class="reference internal" href="impute.html">7.4. Imputation of missing values</a></li>
<li class="toctree-l2"><a class="reference internal" href="unsupervised_reduction.html">7.5. Unsupervised dimensionality reduction</a></li>
<li class="toctree-l2"><a class="reference internal" href="random_projection.html">7.6. Random Projection</a></li>
<li class="toctree-l2"><a class="reference internal" href="kernel_approximation.html">7.7. Kernel Approximation</a></li>
<li class="toctree-l2 current active"><a class="current reference internal" href="#">7.8. Pairwise metrics, Affinities and Kernels</a></li>
<li class="toctree-l2"><a class="reference internal" href="preprocessing_targets.html">7.9. Transforming the prediction target (<code class="docutils literal notranslate"><span class="pre">y</span></code>)</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../datasets.html">8. Dataset loading utilities</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="../datasets/toy_dataset.html">8.1. Toy datasets</a></li>
<li class="toctree-l2"><a class="reference internal" href="../datasets/real_world.html">8.2. Real world datasets</a></li>
<li class="toctree-l2"><a class="reference internal" href="../datasets/sample_generators.html">8.3. Generated datasets</a></li>
<li class="toctree-l2"><a class="reference internal" href="../datasets/loading_other_datasets.html">8.4. Loading other datasets</a></li>
</ul>
</details></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../computing.html">9. Computing with scikit-learn</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="../computing/scaling_strategies.html">9.1. Strategies to scale computationally: bigger data</a></li>
<li class="toctree-l2"><a class="reference internal" href="../computing/computational_performance.html">9.2. Computational Performance</a></li>
<li class="toctree-l2"><a class="reference internal" href="../computing/parallelism.html">9.3. Parallelism, resource management, and configuration</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../model_persistence.html">10. Model persistence</a></li>
<li class="toctree-l1"><a class="reference internal" href="../common_pitfalls.html">11. Common pitfalls and recommended practices</a></li>
<li class="toctree-l1 has-children"><a class="reference internal" href="../dispatching.html">12. Dispatching</a><details><summary><span class="toctree-toggle" role="presentation"><i class="fa-solid fa-chevron-down"></i></span></summary><ul>
<li class="toctree-l2"><a class="reference internal" href="array_api.html">12.1. Array API support (experimental)</a></li>
</ul>
</details></li>
<li class="toctree-l1"><a class="reference internal" href="../machine_learning_map.html">13. Choosing the right estimator</a></li>
<li class="toctree-l1"><a class="reference internal" href="../presentations.html">14. External Resources, Videos and Talks</a></li>
</ul>
</div>
</nav></div>
    </div>
  
  
  <div class="sidebar-primary-items__end sidebar-primary__section">
  </div>


      </div>
      
      <main id="main-content" class="bd-main" role="main">
        
        
          <div class="bd-content">
            <div class="bd-article-container">
              
              <div class="bd-header-article d-print-none">
<div class="header-article-items header-article__inner">
  
    <div class="header-article-items__start">
      
        <div class="header-article-item">

<nav aria-label="Breadcrumb" class="d-print-none">
  <ul class="bd-breadcrumbs">
    
    <li class="breadcrumb-item breadcrumb-home">
      <a href="../index.html" class="nav-link" aria-label="Home">
        <i class="fa-solid fa-home"></i>
      </a>
    </li>
    
    <li class="breadcrumb-item"><a href="../user_guide.html" class="nav-link">User Guide</a></li>
    
    
    <li class="breadcrumb-item"><a href="../data_transforms.html" class="nav-link"><span class="section-number">7. </span>Dataset transformations</a></li>
    
    <li class="breadcrumb-item active" aria-current="page"><span class="ellipsis"><span class="section-number">7.8. </span>Pairwise metrics, Affinities and Kernels</span></li>
  </ul>
</nav>
</div>
      
    </div>
  
  
</div>
</div>
              
              
<div id="searchbox"></div>
                <article class="bd-article">
                  
  <section id="pairwise-metrics-affinities-and-kernels">
<span id="metrics"></span><h1><span class="section-number">7.8. </span>Pairwise metrics, Affinities and Kernels<a class="headerlink" href="#pairwise-metrics-affinities-and-kernels" title="Link to this heading">#</a></h1>
<p>The <a class="reference internal" href="../api/sklearn.metrics.html#module-sklearn.metrics.pairwise" title="sklearn.metrics.pairwise"><code class="xref py py-mod docutils literal notranslate"><span class="pre">sklearn.metrics.pairwise</span></code></a> submodule implements utilities to evaluate
pairwise distances or affinity of sets of samples.</p>
<p>This module contains both distance metrics and kernels. A brief summary is
given on the two here.</p>
<p>Distance metrics are functions <code class="docutils literal notranslate"><span class="pre">d(a,</span> <span class="pre">b)</span></code> such that <code class="docutils literal notranslate"><span class="pre">d(a,</span> <span class="pre">b)</span> <span class="pre">&lt;</span> <span class="pre">d(a,</span> <span class="pre">c)</span></code>
if objects <code class="docutils literal notranslate"><span class="pre">a</span></code> and <code class="docutils literal notranslate"><span class="pre">b</span></code> are considered “more similar” than objects <code class="docutils literal notranslate"><span class="pre">a</span></code>
and <code class="docutils literal notranslate"><span class="pre">c</span></code>. Two objects exactly alike would have a distance of zero.
One of the most popular examples is Euclidean distance.
To be a ‘true’ metric, it must obey the following four conditions:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="mf">1.</span> <span class="n">d</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span> <span class="o">&gt;=</span> <span class="mi">0</span><span class="p">,</span> <span class="k">for</span> <span class="nb">all</span> <span class="n">a</span> <span class="ow">and</span> <span class="n">b</span>
<span class="mf">2.</span> <span class="n">d</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span> <span class="o">==</span> <span class="mi">0</span><span class="p">,</span> <span class="k">if</span> <span class="ow">and</span> <span class="n">only</span> <span class="k">if</span> <span class="n">a</span> <span class="o">=</span> <span class="n">b</span><span class="p">,</span> <span class="n">positive</span> <span class="n">definiteness</span>
<span class="mf">3.</span> <span class="n">d</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span> <span class="o">==</span> <span class="n">d</span><span class="p">(</span><span class="n">b</span><span class="p">,</span> <span class="n">a</span><span class="p">),</span> <span class="n">symmetry</span>
<span class="mf">4.</span> <span class="n">d</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span> <span class="o">&lt;=</span> <span class="n">d</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span> <span class="o">+</span> <span class="n">d</span><span class="p">(</span><span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">),</span> <span class="n">the</span> <span class="n">triangle</span> <span class="n">inequality</span>
</pre></div>
</div>
<p>Kernels are measures of similarity, i.e. <code class="docutils literal notranslate"><span class="pre">s(a,</span> <span class="pre">b)</span> <span class="pre">&gt;</span> <span class="pre">s(a,</span> <span class="pre">c)</span></code>
if objects <code class="docutils literal notranslate"><span class="pre">a</span></code> and <code class="docutils literal notranslate"><span class="pre">b</span></code> are considered “more similar” than objects
<code class="docutils literal notranslate"><span class="pre">a</span></code> and <code class="docutils literal notranslate"><span class="pre">c</span></code>. A kernel must also be positive semi-definite.</p>
<p>There are a number of ways to convert between a distance metric and a
similarity measure, such as a kernel. Let <code class="docutils literal notranslate"><span class="pre">D</span></code> be the distance, and <code class="docutils literal notranslate"><span class="pre">S</span></code> be
the kernel:</p>
<ol class="arabic simple">
<li><dl class="simple">
<dt><code class="docutils literal notranslate"><span class="pre">S</span> <span class="pre">=</span> <span class="pre">np.exp(-D</span> <span class="pre">*</span> <span class="pre">gamma)</span></code>, where one heuristic for choosing</dt><dd><p><code class="docutils literal notranslate"><span class="pre">gamma</span></code> is <code class="docutils literal notranslate"><span class="pre">1</span> <span class="pre">/</span> <span class="pre">num_features</span></code></p>
</dd>
</dl>
</li>
<li><p><code class="docutils literal notranslate"><span class="pre">S</span> <span class="pre">=</span> <span class="pre">1.</span> <span class="pre">/</span> <span class="pre">(D</span> <span class="pre">/</span> <span class="pre">np.max(D))</span></code></p></li>
</ol>
<p>The distances between the row vectors of <code class="docutils literal notranslate"><span class="pre">X</span></code> and the row vectors of <code class="docutils literal notranslate"><span class="pre">Y</span></code>
can be evaluated using <a class="reference internal" href="generated/sklearn.metrics.pairwise_distances.html#sklearn.metrics.pairwise_distances" title="sklearn.metrics.pairwise_distances"><code class="xref py py-func docutils literal notranslate"><span class="pre">pairwise_distances</span></code></a>. If <code class="docutils literal notranslate"><span class="pre">Y</span></code> is omitted the
pairwise distances of the row vectors of <code class="docutils literal notranslate"><span class="pre">X</span></code> are calculated. Similarly,
<a class="reference internal" href="generated/sklearn.metrics.pairwise.pairwise_kernels.html#sklearn.metrics.pairwise.pairwise_kernels" title="sklearn.metrics.pairwise.pairwise_kernels"><code class="xref py py-func docutils literal notranslate"><span class="pre">pairwise.pairwise_kernels</span></code></a> can be used to calculate the kernel between <code class="docutils literal notranslate"><span class="pre">X</span></code>
and <code class="docutils literal notranslate"><span class="pre">Y</span></code> using different kernel functions. See the API reference for more
details.</p>
<div class="doctest highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="kn">import</span><span class="w"> </span><span class="nn">numpy</span><span class="w"> </span><span class="k">as</span><span class="w"> </span><span class="nn">np</span>
<span class="gp">&gt;&gt;&gt; </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn.metrics</span><span class="w"> </span><span class="kn">import</span> <span class="n">pairwise_distances</span>
<span class="gp">&gt;&gt;&gt; </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn.metrics.pairwise</span><span class="w"> </span><span class="kn">import</span> <span class="n">pairwise_kernels</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">X</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">3</span><span class="p">],</span> <span class="p">[</span><span class="mi">3</span><span class="p">,</span> <span class="mi">5</span><span class="p">],</span> <span class="p">[</span><span class="mi">5</span><span class="p">,</span> <span class="mi">8</span><span class="p">]])</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">Y</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">([[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="mi">2</span><span class="p">,</span> <span class="mi">1</span><span class="p">]])</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">pairwise_distances</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">Y</span><span class="p">,</span> <span class="n">metric</span><span class="o">=</span><span class="s1">&#39;manhattan&#39;</span><span class="p">)</span>
<span class="go">array([[ 4.,  2.],</span>
<span class="go">       [ 7.,  5.],</span>
<span class="go">       [12., 10.]])</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">pairwise_distances</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">metric</span><span class="o">=</span><span class="s1">&#39;manhattan&#39;</span><span class="p">)</span>
<span class="go">array([[0., 3., 8.],</span>
<span class="go">       [3., 0., 5.],</span>
<span class="go">       [8., 5., 0.]])</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">pairwise_kernels</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">Y</span><span class="p">,</span> <span class="n">metric</span><span class="o">=</span><span class="s1">&#39;linear&#39;</span><span class="p">)</span>
<span class="go">array([[ 2.,  7.],</span>
<span class="go">       [ 3., 11.],</span>
<span class="go">       [ 5., 18.]])</span>
</pre></div>
</div>
<section id="cosine-similarity">
<span id="id1"></span><h2><span class="section-number">7.8.1. </span>Cosine similarity<a class="headerlink" href="#cosine-similarity" title="Link to this heading">#</a></h2>
<p><a class="reference internal" href="generated/sklearn.metrics.pairwise.cosine_similarity.html#sklearn.metrics.pairwise.cosine_similarity" title="sklearn.metrics.pairwise.cosine_similarity"><code class="xref py py-func docutils literal notranslate"><span class="pre">cosine_similarity</span></code></a> computes the L2-normalized dot product of vectors.
That is, if <span class="math notranslate nohighlight">\(x\)</span> and <span class="math notranslate nohighlight">\(y\)</span> are row vectors,
their cosine similarity <span class="math notranslate nohighlight">\(k\)</span> is defined as:</p>
<div class="math notranslate nohighlight">
\[k(x, y) = \frac{x y^\top}{\|x\| \|y\|}\]</div>
<p>This is called cosine similarity, because Euclidean (L2) normalization
projects the vectors onto the unit sphere,
and their dot product is then the cosine of the angle between the points
denoted by the vectors.</p>
<p>This kernel is a popular choice for computing the similarity of documents
represented as tf-idf vectors.
<a class="reference internal" href="generated/sklearn.metrics.pairwise.cosine_similarity.html#sklearn.metrics.pairwise.cosine_similarity" title="sklearn.metrics.pairwise.cosine_similarity"><code class="xref py py-func docutils literal notranslate"><span class="pre">cosine_similarity</span></code></a> accepts <code class="docutils literal notranslate"><span class="pre">scipy.sparse</span></code> matrices.
(Note that the tf-idf functionality in <code class="docutils literal notranslate"><span class="pre">sklearn.feature_extraction.text</span></code>
can produce normalized vectors, in which case <a class="reference internal" href="generated/sklearn.metrics.pairwise.cosine_similarity.html#sklearn.metrics.pairwise.cosine_similarity" title="sklearn.metrics.pairwise.cosine_similarity"><code class="xref py py-func docutils literal notranslate"><span class="pre">cosine_similarity</span></code></a>
is equivalent to <a class="reference internal" href="generated/sklearn.metrics.pairwise.linear_kernel.html#sklearn.metrics.pairwise.linear_kernel" title="sklearn.metrics.pairwise.linear_kernel"><code class="xref py py-func docutils literal notranslate"><span class="pre">linear_kernel</span></code></a>, only slower.)</p>
<p class="rubric">References</p>
<ul class="simple">
<li><p>C.D. Manning, P. Raghavan and H. Schütze (2008). Introduction to
Information Retrieval. Cambridge University Press.
<a class="reference external" href="https://nlp.stanford.edu/IR-book/html/htmledition/the-vector-space-model-for-scoring-1.html">https://nlp.stanford.edu/IR-book/html/htmledition/the-vector-space-model-for-scoring-1.html</a></p></li>
</ul>
</section>
<section id="linear-kernel">
<span id="id2"></span><h2><span class="section-number">7.8.2. </span>Linear kernel<a class="headerlink" href="#linear-kernel" title="Link to this heading">#</a></h2>
<p>The function <a class="reference internal" href="generated/sklearn.metrics.pairwise.linear_kernel.html#sklearn.metrics.pairwise.linear_kernel" title="sklearn.metrics.pairwise.linear_kernel"><code class="xref py py-func docutils literal notranslate"><span class="pre">linear_kernel</span></code></a> computes the linear kernel, that is, a
special case of <a class="reference internal" href="generated/sklearn.metrics.pairwise.polynomial_kernel.html#sklearn.metrics.pairwise.polynomial_kernel" title="sklearn.metrics.pairwise.polynomial_kernel"><code class="xref py py-func docutils literal notranslate"><span class="pre">polynomial_kernel</span></code></a> with <code class="docutils literal notranslate"><span class="pre">degree=1</span></code> and <code class="docutils literal notranslate"><span class="pre">coef0=0</span></code> (homogeneous).
If <code class="docutils literal notranslate"><span class="pre">x</span></code> and <code class="docutils literal notranslate"><span class="pre">y</span></code> are column vectors, their linear kernel is:</p>
<div class="math notranslate nohighlight">
\[k(x, y) = x^\top y\]</div>
</section>
<section id="polynomial-kernel">
<span id="id3"></span><h2><span class="section-number">7.8.3. </span>Polynomial kernel<a class="headerlink" href="#polynomial-kernel" title="Link to this heading">#</a></h2>
<p>The function <a class="reference internal" href="generated/sklearn.metrics.pairwise.polynomial_kernel.html#sklearn.metrics.pairwise.polynomial_kernel" title="sklearn.metrics.pairwise.polynomial_kernel"><code class="xref py py-func docutils literal notranslate"><span class="pre">polynomial_kernel</span></code></a> computes the degree-d polynomial kernel
between two vectors. The polynomial kernel represents the similarity between two
vectors. Conceptually, the polynomial kernel considers not only the similarity
between vectors under the same dimension, but also across dimensions. When used
in machine learning algorithms, this allows to account for feature interaction.</p>
<p>The polynomial kernel is defined as:</p>
<div class="math notranslate nohighlight">
\[k(x, y) = (\gamma x^\top y +c_0)^d\]</div>
<p>where:</p>
<ul class="simple">
<li><p><code class="docutils literal notranslate"><span class="pre">x</span></code>, <code class="docutils literal notranslate"><span class="pre">y</span></code> are the input vectors</p></li>
<li><p><code class="docutils literal notranslate"><span class="pre">d</span></code> is the kernel degree</p></li>
</ul>
<p>If <span class="math notranslate nohighlight">\(c_0 = 0\)</span> the kernel is said to be homogeneous.</p>
</section>
<section id="sigmoid-kernel">
<span id="id4"></span><h2><span class="section-number">7.8.4. </span>Sigmoid kernel<a class="headerlink" href="#sigmoid-kernel" title="Link to this heading">#</a></h2>
<p>The function <a class="reference internal" href="generated/sklearn.metrics.pairwise.sigmoid_kernel.html#sklearn.metrics.pairwise.sigmoid_kernel" title="sklearn.metrics.pairwise.sigmoid_kernel"><code class="xref py py-func docutils literal notranslate"><span class="pre">sigmoid_kernel</span></code></a> computes the sigmoid kernel between two
vectors. The sigmoid kernel is also known as hyperbolic tangent, or Multilayer
Perceptron (because, in the neural network field, it is often used as neuron
activation function). It is defined as:</p>
<div class="math notranslate nohighlight">
\[k(x, y) = \tanh( \gamma x^\top y + c_0)\]</div>
<p>where:</p>
<ul class="simple">
<li><p><code class="docutils literal notranslate"><span class="pre">x</span></code>, <code class="docutils literal notranslate"><span class="pre">y</span></code> are the input vectors</p></li>
<li><p><span class="math notranslate nohighlight">\(\gamma\)</span> is known as slope</p></li>
<li><p><span class="math notranslate nohighlight">\(c_0\)</span> is known as intercept</p></li>
</ul>
</section>
<section id="rbf-kernel">
<span id="id5"></span><h2><span class="section-number">7.8.5. </span>RBF kernel<a class="headerlink" href="#rbf-kernel" title="Link to this heading">#</a></h2>
<p>The function <a class="reference internal" href="generated/sklearn.metrics.pairwise.rbf_kernel.html#sklearn.metrics.pairwise.rbf_kernel" title="sklearn.metrics.pairwise.rbf_kernel"><code class="xref py py-func docutils literal notranslate"><span class="pre">rbf_kernel</span></code></a> computes the radial basis function (RBF) kernel
between two vectors. This kernel is defined as:</p>
<div class="math notranslate nohighlight">
\[k(x, y) = \exp( -\gamma \| x-y \|^2)\]</div>
<p>where <code class="docutils literal notranslate"><span class="pre">x</span></code> and <code class="docutils literal notranslate"><span class="pre">y</span></code> are the input vectors. If <span class="math notranslate nohighlight">\(\gamma = \sigma^{-2}\)</span>
the kernel is known as the Gaussian kernel of variance <span class="math notranslate nohighlight">\(\sigma^2\)</span>.</p>
</section>
<section id="laplacian-kernel">
<span id="id6"></span><h2><span class="section-number">7.8.6. </span>Laplacian kernel<a class="headerlink" href="#laplacian-kernel" title="Link to this heading">#</a></h2>
<p>The function <a class="reference internal" href="generated/sklearn.metrics.pairwise.laplacian_kernel.html#sklearn.metrics.pairwise.laplacian_kernel" title="sklearn.metrics.pairwise.laplacian_kernel"><code class="xref py py-func docutils literal notranslate"><span class="pre">laplacian_kernel</span></code></a> is a variant on the radial basis
function kernel defined as:</p>
<div class="math notranslate nohighlight">
\[k(x, y) = \exp( -\gamma \| x-y \|_1)\]</div>
<p>where <code class="docutils literal notranslate"><span class="pre">x</span></code> and <code class="docutils literal notranslate"><span class="pre">y</span></code> are the input vectors and <span class="math notranslate nohighlight">\(\|x-y\|_1\)</span> is the
Manhattan distance between the input vectors.</p>
<p>It has proven useful in ML applied to noiseless data.
See e.g. <a class="reference external" href="https://onlinelibrary.wiley.com/doi/10.1002/qua.24954/abstract/">Machine learning for quantum mechanics in a nutshell</a>.</p>
</section>
<section id="chi-squared-kernel">
<span id="chi2-kernel"></span><h2><span class="section-number">7.8.7. </span>Chi-squared kernel<a class="headerlink" href="#chi-squared-kernel" title="Link to this heading">#</a></h2>
<p>The chi-squared kernel is a very popular choice for training non-linear SVMs in
computer vision applications.
It can be computed using <a class="reference internal" href="generated/sklearn.metrics.pairwise.chi2_kernel.html#sklearn.metrics.pairwise.chi2_kernel" title="sklearn.metrics.pairwise.chi2_kernel"><code class="xref py py-func docutils literal notranslate"><span class="pre">chi2_kernel</span></code></a> and then passed to an
<a class="reference internal" href="generated/sklearn.svm.SVC.html#sklearn.svm.SVC" title="sklearn.svm.SVC"><code class="xref py py-class docutils literal notranslate"><span class="pre">SVC</span></code></a> with <code class="docutils literal notranslate"><span class="pre">kernel=&quot;precomputed&quot;</span></code>:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn.svm</span><span class="w"> </span><span class="kn">import</span> <span class="n">SVC</span>
<span class="gp">&gt;&gt;&gt; </span><span class="kn">from</span><span class="w"> </span><span class="nn">sklearn.metrics.pairwise</span><span class="w"> </span><span class="kn">import</span> <span class="n">chi2_kernel</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">X</span> <span class="o">=</span> <span class="p">[[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">],</span> <span class="p">[</span><span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">],</span> <span class="p">[</span><span class="mf">.2</span><span class="p">,</span> <span class="mf">.8</span><span class="p">],</span> <span class="p">[</span><span class="mf">.7</span><span class="p">,</span> <span class="mf">.3</span><span class="p">]]</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">y</span> <span class="o">=</span> <span class="p">[</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">]</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">K</span> <span class="o">=</span> <span class="n">chi2_kernel</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">gamma</span><span class="o">=</span><span class="mf">.5</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">K</span>
<span class="go">array([[1.        , 0.36787944, 0.89483932, 0.58364548],</span>
<span class="go">       [0.36787944, 1.        , 0.51341712, 0.83822343],</span>
<span class="go">       [0.89483932, 0.51341712, 1.        , 0.7768366 ],</span>
<span class="go">       [0.58364548, 0.83822343, 0.7768366 , 1.        ]])</span>

<span class="gp">&gt;&gt;&gt; </span><span class="n">svm</span> <span class="o">=</span> <span class="n">SVC</span><span class="p">(</span><span class="n">kernel</span><span class="o">=</span><span class="s1">&#39;precomputed&#39;</span><span class="p">)</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">K</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">svm</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">K</span><span class="p">)</span>
<span class="go">array([0, 1, 0, 1])</span>
</pre></div>
</div>
<p>It can also be directly used as the <code class="docutils literal notranslate"><span class="pre">kernel</span></code> argument:</p>
<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="gp">&gt;&gt;&gt; </span><span class="n">svm</span> <span class="o">=</span> <span class="n">SVC</span><span class="p">(</span><span class="n">kernel</span><span class="o">=</span><span class="n">chi2_kernel</span><span class="p">)</span><span class="o">.</span><span class="n">fit</span><span class="p">(</span><span class="n">X</span><span class="p">,</span> <span class="n">y</span><span class="p">)</span>
<span class="gp">&gt;&gt;&gt; </span><span class="n">svm</span><span class="o">.</span><span class="n">predict</span><span class="p">(</span><span class="n">X</span><span class="p">)</span>
<span class="go">array([0, 1, 0, 1])</span>
</pre></div>
</div>
<p>The chi squared kernel is given by</p>
<div class="math notranslate nohighlight">
\[k(x, y) = \exp \left (-\gamma \sum_i \frac{(x[i] - y[i]) ^ 2}{x[i] + y[i]} \right )\]</div>
<p>The data is assumed to be non-negative, and is often normalized to have an L1-norm of one.
The normalization is rationalized with the connection to the chi squared distance,
which is a distance between discrete probability distributions.</p>
<p>The chi squared kernel is most commonly used on histograms (bags) of visual words.</p>
<p class="rubric">References</p>
<ul class="simple">
<li><p>Zhang, J. and Marszalek, M. and Lazebnik, S. and Schmid, C.
Local features and kernels for classification of texture and object
categories: A comprehensive study
International Journal of Computer Vision 2007
<a class="reference external" href="https://hal.archives-ouvertes.fr/hal-00171412/document">https://hal.archives-ouvertes.fr/hal-00171412/document</a></p></li>
</ul>
</section>
</section>


                </article>
              
              
                <footer class="bd-footer-article">
                  <div class="footer-article-items footer-article__inner">
  
    <div class="footer-article-item">
<div class="prev-next-area">
    <a class="left-prev"
       href="kernel_approximation.html"
       title="previous page">
      <i class="fa-solid fa-angle-left"></i>
      <div class="prev-next-info">
        <p class="prev-next-subtitle">previous</p>
        <p class="prev-next-title"><span class="section-number">7.7. </span>Kernel Approximation</p>
      </div>
    </a>
    <a class="right-next"
       href="preprocessing_targets.html"
       title="next page">
      <div class="prev-next-info">
        <p class="prev-next-subtitle">next</p>
        <p class="prev-next-title"><span class="section-number">7.9. </span>Transforming the prediction target (<code class="docutils literal notranslate"><span class="pre">y</span></code>)</p>
      </div>
      <i class="fa-solid fa-angle-right"></i>
    </a>
</div></div>
  
</div>
                </footer>
              
              
            </div>
            
            
                <dialog id="pst-secondary-sidebar-modal"></dialog>
                <div id="pst-secondary-sidebar" class="bd-sidebar-secondary bd-toc"><div class="sidebar-secondary-items sidebar-secondary__inner">


  <div class="sidebar-secondary-item">
<div
    id="pst-page-navigation-heading-2"
    class="page-toc tocsection onthispage">
    <i class="fa-solid fa-list"></i> On this page
  </div>
  <nav class="bd-toc-nav page-toc" aria-labelledby="pst-page-navigation-heading-2">
    <ul class="visible nav section-nav flex-column">
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#cosine-similarity">7.8.1. Cosine similarity</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#linear-kernel">7.8.2. Linear kernel</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#polynomial-kernel">7.8.3. Polynomial kernel</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#sigmoid-kernel">7.8.4. Sigmoid kernel</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#rbf-kernel">7.8.5. RBF kernel</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#laplacian-kernel">7.8.6. Laplacian kernel</a></li>
<li class="toc-h2 nav-item toc-entry"><a class="reference internal nav-link" href="#chi-squared-kernel">7.8.7. Chi-squared kernel</a></li>
</ul>
  </nav></div>

  <div class="sidebar-secondary-item">
  <div role="note" aria-label="source link">
    <h3>This Page</h3>
    <ul class="this-page-menu">
      <li><a href="../_sources/modules/metrics.rst.txt"
            rel="nofollow">Show Source</a></li>
    </ul>
   </div></div>

</div></div>
              
            
          </div>
          <footer class="bd-footer-content">
            
          </footer>
        
      </main>
    </div>
  </div>
  
  <!-- Scripts loaded after <body> so the DOM is not blocked -->
  <script defer src="../_static/scripts/bootstrap.js?digest=8878045cc6db502f8baf"></script>
<script defer src="../_static/scripts/pydata-sphinx-theme.js?digest=8878045cc6db502f8baf"></script>

  <footer class="bd-footer">
<div class="bd-footer__inner bd-page-width">
  
    <div class="footer-items__start">
      
        <div class="footer-item">

  <p class="copyright">
    
      © Copyright 2007 - 2025, scikit-learn developers (BSD License).
      <br/>
    
  </p>
</div>
      
    </div>
  
  
</div>

  </footer>
  </body>
</html>