<link href="https://www.w3.org/StyleSheets/TR/2021/W3C-ED" rel="stylesheet">

<meta content="Bikeshed version 82ce88815, updated Thu Sep 7 16:33:55 2023 -0700" name="generator">

<link href="https://www.w3.org/TR/webnn/" rel="canonical">

<style>

/* Make <dl> blocks more distinct from their surroundings. */

main dl:not(.switch) {

<div class="head">

<p data-fill-with="logo"><a class="logo" href="https://www.w3.org/"> <img alt="W3C" height="48" src="https://www.w3.org/StyleSheets/TR/2021/logos/W3C" width="72"> </a> </p>

<h1 class="p-name no-ref" id="title">Web Neural Network API</h1>

<details open>

<summary>More details about this document</summary>

<div data-fill-with="spec-metadata">

<ol>

<li data-md>

<p>If <var>options</var>.<code class="idl"><a data-link-type="idl" href="#dom-mlclampoptions-minvalue" id="ref-for-dom-mlclampoptions-minvalue①">minValue</a></code> is greater than <var>options</var>.<code class="idl"><a data-link-type="idl" href="#dom-mlclampoptions-maxvalue" id="ref-for-dom-mlclampoptions-maxvalue①">maxValue</a></code>, then return false.</p>

<li data-md>

<p>Return true.</p>

</ol>

<li data-md>

<p>Let <var>output</var> be the result of <a data-link-type="dfn" href="#copy-an-mloperand" id="ref-for-copy-an-mloperand①">copying an MLOperand</a> given <var>input</var>.</p>

<li data-md>

<li data-md>

<p>Set <var>output</var>.<code class="idl"><a data-link-type="idl" href="#dom-mloperand-operator-slot" id="ref-for-dom-mloperand-operator-slot⑤">[[operator]]</a></code> to <var>operator</var>.</p>

<li data-md>

<div class="issue"> The maximum number of operand dimensions is not defined, but native ML APIs usually have a maximum supported size. <a href="https://github.com/webmachinelearning/webnn/issues/456">[Issue #456]</a> <a class="issue-return" href="#issue-8054fa8e" title="Jump to section">↵</a></div>

<div class="issue"> If <code class="idl"><a data-link-type="idl" href="#mlgraphbuilder">MLGraphBuilder</a></code> can’t be re-used, then this can be simplified: enforce uniqueness in <code class="idl"><a data-link-type="idl" href="#dom-mlgraphbuilder-input">input()</a></code> instead, and iteration can be done over all of the graph’s inputs instead of needing this traversal. <a href="https://github.com/webmachinelearning/webnn/issues/567">[Issue #567]</a> <a class="issue-return" href="#issue-91f4b8a3" title="Jump to section">↵</a></div>

<div class="issue"> If <code class="idl"><a data-link-type="idl" href="#dom-mlgraphbuilder-constant">constants'</a></code> <code class="idl"><a data-link-type="idl" href="https://webidl.spec.whatwg.org/#idl-ArrayBuffer">ArrayBuffer</a></code>s are not <a data-link-type="dfn" href="https://webidl.spec.whatwg.org/#arraybuffer-transfer">transferred</a>, make copies for <a data-link-type="dfn" href="#mlgraphbuilder-graph">graph</a>'s <a data-link-type="dfn" href="#computational-graph-constants">constants</a> here. <a href="https://github.com/webmachinelearning/webnn/issues/566">[Issue #566]</a> <a class="issue-return" href="#issue-6f962d4f" title="Jump to section">↵</a></div>

</div>

<script>/* Boilerplate: script-dfn-panel */

"use strict";

window.dfnpanelData['dom-mlgraphbuilder-batchnormalization'] = {"dfnID": "dom-mlgraphbuilder-batchnormalization", "url": "#dom-mlgraphbuilder-batchnormalization", "dfnText": "batchNormalization(input, mean, variance, options)", "refSections": [{"refs": [{"id": "ref-for-dom-mlgraphbuilder-batchnormalization"}], "title": "7.8.6. batchNormalization"}, {"refs": [{"id": "ref-for-dom-mlgraphbuilder-batchnormalization\u2460"}], "title": "7.8.24. instanceNormalization"}, {"refs": [{"id": "ref-for-dom-mlgraphbuilder-batchnormalization\u2461"}], "title": "7.8.25. layerNormalization"}], "external": false};

window.dfnpanelData['dom-mlgraphbuilder-cast'] = {"dfnID": "dom-mlgraphbuilder-cast", "url": "#dom-mlgraphbuilder-cast", "dfnText": "cast(input, type)", "refSections": [{"refs": [{"id": "ref-for-dom-mlgraphbuilder-cast"}], "title": "7.8.7. cast"}], "external": false};

window.dfnpanelData['dictdef-mlclampoptions'] = {"dfnID": "dictdef-mlclampoptions", "url": "#dictdef-mlclampoptions", "dfnText": "MLClampOptions", "refSections": [{"refs": [{"id": "ref-for-dictdef-mlclampoptions"}, {"id": "ref-for-dictdef-mlclampoptions\u2460"}, {"id": "ref-for-dictdef-mlclampoptions\u2461"}], "title": "7.8.8. clamp"}, {"refs": [{"id": "ref-for-dictdef-mlclampoptions\u2462"}], "title": "7.8.8.1. clamp(input, options)"}, {"refs": [{"id": "ref-for-dictdef-mlclampoptions\u2463"}], "title": "7.8.8.2. clamp(options)"}], "external": false};

window.dfnpanelData['check-clamp-options'] = {"dfnID": "check-clamp-options", "url": "#check-clamp-options", "dfnText": "check clamp options", "refSections": [{"refs": [{"id": "ref-for-check-clamp-options"}], "title": "7.8.8.1. clamp(input, options)"}, {"refs": [{"id": "ref-for-check-clamp-options\u2460"}], "title": "7.8.8.2. clamp(options)"}], "external": false};

window.dfnpanelData['dom-mlgraphbuilder-clamp'] = {"dfnID": "dom-mlgraphbuilder-clamp", "url": "#dom-mlgraphbuilder-clamp", "dfnText": "clamp(input, options)", "refSections": [{"refs": [{"id": "ref-for-dom-mlgraphbuilder-clamp"}], "title": "7.8.8. clamp"}, {"refs": [{"id": "ref-for-dom-mlgraphbuilder-clamp\u2460"}], "title": "7.8.8.1. clamp(input, options)"}], "external": false};

window.dfnpanelData['dom-mlgraphbuilder-clamp-options'] = {"dfnID": "dom-mlgraphbuilder-clamp-options", "url": "#dom-mlgraphbuilder-clamp-options", "dfnText": "clamp(options)", "refSections": [{"refs": [{"id": "ref-for-dom-mlgraphbuilder-clamp-options"}], "title": "7.8.8. clamp"}, {"refs": [{"id": "ref-for-dom-mlgraphbuilder-clamp-options\u2460"}], "title": "7.8.8.2. clamp(options)"}, {"refs": [{"id": "ref-for-dom-mlgraphbuilder-clamp-options\u2461"}], "title": "7.8.17. gather"}], "external": false};