|
-
- <!doctype html>
- <html lang="en" class="no-js">
- <head>
-
- <meta charset="utf-8">
- <meta name="viewport" content="width=device-width,initial-scale=1">
-
-
-
- <link rel="canonical" href="https://mindspore-lab.github.io/mindocr/cn/">
-
-
-
- <link rel="next" href="mkdocs/modelzoo_training/">
-
-
- <link rel="icon" href="../assets/images/favicon.png">
- <meta name="generator" content="mkdocs-1.5.3, mkdocs-material-9.5.18">
-
-
-
- <title>主页 - MindOCR Docs</title>
-
-
-
- <link rel="stylesheet" href="../assets/stylesheets/main.66ac8b77.min.css">
-
-
- <link rel="stylesheet" href="../assets/stylesheets/palette.06af60db.min.css">
-
-
-
-
-
-
-
-
-
-
-
-
- <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
- <link rel="stylesheet" href="https://fonts.googleapis.com/css?family=Roboto:300,300i,400,400i,700,700i%7CRoboto+Mono:400,400i,700,700i&display=fallback">
- <style>:root{--md-text-font:"Roboto";--md-code-font:"Roboto Mono"}</style>
-
-
-
- <link rel="stylesheet" href="../assets/_mkdocstrings.css">
-
- <script>__md_scope=new URL("..",location),__md_hash=e=>[...e].reduce((e,_)=>(e<<5)-e+_.charCodeAt(0),0),__md_get=(e,_=localStorage,t=__md_scope)=>JSON.parse(_.getItem(t.pathname+"."+e)),__md_set=(e,_,t=localStorage,a=__md_scope)=>{try{t.setItem(a.pathname+"."+e,JSON.stringify(_))}catch(e){}}</script>
-
-
-
-
-
-
- </head>
-
-
-
-
-
-
-
-
-
- <body dir="ltr" data-md-color-scheme="default" data-md-color-primary="black" data-md-color-accent="indigo">
-
-
- <input class="md-toggle" data-md-toggle="drawer" type="checkbox" id="__drawer" autocomplete="off">
- <input class="md-toggle" data-md-toggle="search" type="checkbox" id="__search" autocomplete="off">
- <label class="md-overlay" for="__drawer"></label>
- <div data-md-component="skip">
-
-
- <a href="#mindocr" class="md-skip">
- Skip to content
- </a>
-
- </div>
- <div data-md-component="announce">
-
- </div>
-
-
-
-
- <header class="md-header" data-md-component="header">
- <nav class="md-header__inner md-grid" aria-label="Header">
- <a href=".." title="MindOCR Docs" class="md-header__button md-logo" aria-label="MindOCR Docs" data-md-component="logo">
-
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M12 8a3 3 0 0 0 3-3 3 3 0 0 0-3-3 3 3 0 0 0-3 3 3 3 0 0 0 3 3m0 3.54C9.64 9.35 6.5 8 3 8v11c3.5 0 6.64 1.35 9 3.54 2.36-2.19 5.5-3.54 9-3.54V8c-3.5 0-6.64 1.35-9 3.54Z"/></svg>
-
- </a>
- <label class="md-header__button md-icon" for="__drawer">
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M3 6h18v2H3V6m0 5h18v2H3v-2m0 5h18v2H3v-2Z"/></svg>
- </label>
- <div class="md-header__title" data-md-component="header-title">
- <div class="md-header__ellipsis">
- <div class="md-header__topic">
- <span class="md-ellipsis">
- MindOCR Docs
- </span>
- </div>
- <div class="md-header__topic" data-md-component="header-topic">
- <span class="md-ellipsis">
-
- 主页
-
- </span>
- </div>
- </div>
- </div>
-
-
- <form class="md-header__option" data-md-component="palette">
-
-
-
-
- <input class="md-option" data-md-color-media="(prefers-color-scheme: light)" data-md-color-scheme="default" data-md-color-primary="black" data-md-color-accent="indigo" aria-label="Switch to dark mode" type="radio" name="__palette" id="__palette_0">
-
- <label class="md-header__button md-icon" title="Switch to dark mode" for="__palette_1" hidden>
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M12 7a5 5 0 0 1 5 5 5 5 0 0 1-5 5 5 5 0 0 1-5-5 5 5 0 0 1 5-5m0 2a3 3 0 0 0-3 3 3 3 0 0 0 3 3 3 3 0 0 0 3-3 3 3 0 0 0-3-3m0-7 2.39 3.42C13.65 5.15 12.84 5 12 5c-.84 0-1.65.15-2.39.42L12 2M3.34 7l4.16-.35A7.2 7.2 0 0 0 5.94 8.5c-.44.74-.69 1.5-.83 2.29L3.34 7m.02 10 1.76-3.77a7.131 7.131 0 0 0 2.38 4.14L3.36 17M20.65 7l-1.77 3.79a7.023 7.023 0 0 0-2.38-4.15l4.15.36m-.01 10-4.14.36c.59-.51 1.12-1.14 1.54-1.86.42-.73.69-1.5.83-2.29L20.64 17M12 22l-2.41-3.44c.74.27 1.55.44 2.41.44.82 0 1.63-.17 2.37-.44L12 22Z"/></svg>
- </label>
-
-
-
-
-
- <input class="md-option" data-md-color-media="(prefers-color-scheme: dark)" data-md-color-scheme="slate" data-md-color-primary="black" data-md-color-accent="indigo" aria-label="Switch to light mode" type="radio" name="__palette" id="__palette_1">
-
- <label class="md-header__button md-icon" title="Switch to light mode" for="__palette_0" hidden>
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="m17.75 4.09-2.53 1.94.91 3.06-2.63-1.81-2.63 1.81.91-3.06-2.53-1.94L12.44 4l1.06-3 1.06 3 3.19.09m3.5 6.91-1.64 1.25.59 1.98-1.7-1.17-1.7 1.17.59-1.98L15.75 11l2.06-.05L18.5 9l.69 1.95 2.06.05m-2.28 4.95c.83-.08 1.72 1.1 1.19 1.85-.32.45-.66.87-1.08 1.27C15.17 23 8.84 23 4.94 19.07c-3.91-3.9-3.91-10.24 0-14.14.4-.4.82-.76 1.27-1.08.75-.53 1.93.36 1.85 1.19-.27 2.86.69 5.83 2.89 8.02a9.96 9.96 0 0 0 8.02 2.89m-1.64 2.02a12.08 12.08 0 0 1-7.8-3.47c-2.17-2.19-3.33-5-3.49-7.82-2.81 3.14-2.7 7.96.31 10.98 3.02 3.01 7.84 3.12 10.98.31Z"/></svg>
- </label>
-
-
- </form>
-
-
-
- <script>var media,input,key,value,palette=__md_get("__palette");if(palette&&palette.color){"(prefers-color-scheme)"===palette.color.media&&(media=matchMedia("(prefers-color-scheme: light)"),input=document.querySelector(media.matches?"[data-md-color-media='(prefers-color-scheme: light)']":"[data-md-color-media='(prefers-color-scheme: dark)']"),palette.color.media=input.getAttribute("data-md-color-media"),palette.color.scheme=input.getAttribute("data-md-color-scheme"),palette.color.primary=input.getAttribute("data-md-color-primary"),palette.color.accent=input.getAttribute("data-md-color-accent"));for([key,value]of Object.entries(palette.color))document.body.setAttribute("data-md-color-"+key,value)}</script>
-
-
-
- <label class="md-header__button md-icon" for="__search">
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M9.5 3A6.5 6.5 0 0 1 16 9.5c0 1.61-.59 3.09-1.56 4.23l.27.27h.79l5 5-1.5 1.5-5-5v-.79l-.27-.27A6.516 6.516 0 0 1 9.5 16 6.5 6.5 0 0 1 3 9.5 6.5 6.5 0 0 1 9.5 3m0 2C7 5 5 7 5 9.5S7 14 9.5 14 14 12 14 9.5 12 5 9.5 5Z"/></svg>
- </label>
- <div class="md-search" data-md-component="search" role="dialog">
- <label class="md-search__overlay" for="__search"></label>
- <div class="md-search__inner" role="search">
- <form class="md-search__form" name="search">
- <input type="text" class="md-search__input" name="query" aria-label="Search" placeholder="Search" autocapitalize="off" autocorrect="off" autocomplete="off" spellcheck="false" data-md-component="search-query" required>
- <label class="md-search__icon md-icon" for="__search">
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M9.5 3A6.5 6.5 0 0 1 16 9.5c0 1.61-.59 3.09-1.56 4.23l.27.27h.79l5 5-1.5 1.5-5-5v-.79l-.27-.27A6.516 6.516 0 0 1 9.5 16 6.5 6.5 0 0 1 3 9.5 6.5 6.5 0 0 1 9.5 3m0 2C7 5 5 7 5 9.5S7 14 9.5 14 14 12 14 9.5 12 5 9.5 5Z"/></svg>
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M20 11v2H8l5.5 5.5-1.42 1.42L4.16 12l7.92-7.92L13.5 5.5 8 11h12Z"/></svg>
- </label>
- <nav class="md-search__options" aria-label="Search">
-
- <a href="javascript:void(0)" class="md-search__icon md-icon" title="Share" aria-label="Share" data-clipboard data-clipboard-text="" data-md-component="search-share" tabindex="-1">
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M18 16.08c-.76 0-1.44.3-1.96.77L8.91 12.7c.05-.23.09-.46.09-.7 0-.24-.04-.47-.09-.7l7.05-4.11c.54.5 1.25.81 2.04.81a3 3 0 0 0 3-3 3 3 0 0 0-3-3 3 3 0 0 0-3 3c0 .24.04.47.09.7L8.04 9.81C7.5 9.31 6.79 9 6 9a3 3 0 0 0-3 3 3 3 0 0 0 3 3c.79 0 1.5-.31 2.04-.81l7.12 4.15c-.05.21-.08.43-.08.66 0 1.61 1.31 2.91 2.92 2.91 1.61 0 2.92-1.3 2.92-2.91A2.92 2.92 0 0 0 18 16.08Z"/></svg>
- </a>
-
- <button type="reset" class="md-search__icon md-icon" title="Clear" aria-label="Clear" tabindex="-1">
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M19 6.41 17.59 5 12 10.59 6.41 5 5 6.41 10.59 12 5 17.59 6.41 19 12 13.41 17.59 19 19 17.59 13.41 12 19 6.41Z"/></svg>
- </button>
- </nav>
-
- <div class="md-search__suggest" data-md-component="search-suggest"></div>
-
- </form>
- <div class="md-search__output">
- <div class="md-search__scrollwrap" data-md-scrollfix>
- <div class="md-search-result" data-md-component="search-result">
- <div class="md-search-result__meta">
- Initializing search
- </div>
- <ol class="md-search-result__list" role="presentation"></ol>
- </div>
- </div>
- </div>
- </div>
- </div>
-
-
- <div class="md-header__source">
- <a href="https://github.com/mindspore-lab/mindocr" title="Go to repository" class="md-source" data-md-component="source">
- <div class="md-source__icon md-icon">
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 448 512"><!--! Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License) Copyright 2024 Fonticons, Inc.--><path d="M439.55 236.05 244 40.45a28.87 28.87 0 0 0-40.81 0l-40.66 40.63 51.52 51.52c27.06-9.14 52.68 16.77 43.39 43.68l49.66 49.66c34.23-11.8 61.18 31 35.47 56.69-26.49 26.49-70.21-2.87-56-37.34L240.22 199v121.85c25.3 12.54 22.26 41.85 9.08 55a34.34 34.34 0 0 1-48.55 0c-17.57-17.6-11.07-46.91 11.25-56v-123c-20.8-8.51-24.6-30.74-18.64-45L142.57 101 8.45 235.14a28.86 28.86 0 0 0 0 40.81l195.61 195.6a28.86 28.86 0 0 0 40.8 0l194.69-194.69a28.86 28.86 0 0 0 0-40.81z"/></svg>
- </div>
- <div class="md-source__repository">
- mindspore-lab/mindocr
- </div>
- </a>
- </div>
-
- </nav>
-
- </header>
-
- <div class="md-container" data-md-component="container">
-
-
-
-
-
- <nav class="md-tabs" aria-label="Tabs" data-md-component="tabs">
- <div class="md-grid">
- <ul class="md-tabs__list">
-
-
-
-
-
-
-
- <li class="md-tabs__item md-tabs__item--active">
- <a href="./" class="md-tabs__link">
-
-
-
-
- 主页
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
- <li class="md-tabs__item">
- <a href="mkdocs/modelzoo_training/" class="md-tabs__link">
-
-
-
-
- 模型仓库
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- <li class="md-tabs__item">
- <a href="datasets/converters/" class="md-tabs__link">
-
-
-
-
- 教程
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
-
-
-
- <li class="md-tabs__item">
- <a href="reference/api_doc/" class="md-tabs__link">
-
-
-
-
- Reference
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-tabs__item">
- <a href="mkdocs/contributing/" class="md-tabs__link">
-
-
-
-
- 说明
-
- </a>
- </li>
-
-
-
-
- </ul>
- </div>
- </nav>
-
-
-
- <main class="md-main" data-md-component="main">
- <div class="md-main__inner md-grid">
-
-
-
- <div class="md-sidebar md-sidebar--primary" data-md-component="sidebar" data-md-type="navigation" >
- <div class="md-sidebar__scrollwrap">
- <div class="md-sidebar__inner">
-
-
-
-
-
-
- <nav class="md-nav md-nav--primary md-nav--lifted" aria-label="Navigation" data-md-level="0">
- <label class="md-nav__title" for="__drawer">
- <a href=".." title="MindOCR Docs" class="md-nav__button md-logo" aria-label="MindOCR Docs" data-md-component="logo">
-
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M12 8a3 3 0 0 0 3-3 3 3 0 0 0-3-3 3 3 0 0 0-3 3 3 3 0 0 0 3 3m0 3.54C9.64 9.35 6.5 8 3 8v11c3.5 0 6.64 1.35 9 3.54 2.36-2.19 5.5-3.54 9-3.54V8c-3.5 0-6.64 1.35-9 3.54Z"/></svg>
-
- </a>
- MindOCR Docs
- </label>
-
- <div class="md-nav__source">
- <a href="https://github.com/mindspore-lab/mindocr" title="Go to repository" class="md-source" data-md-component="source">
- <div class="md-source__icon md-icon">
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 448 512"><!--! Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License) Copyright 2024 Fonticons, Inc.--><path d="M439.55 236.05 244 40.45a28.87 28.87 0 0 0-40.81 0l-40.66 40.63 51.52 51.52c27.06-9.14 52.68 16.77 43.39 43.68l49.66 49.66c34.23-11.8 61.18 31 35.47 56.69-26.49 26.49-70.21-2.87-56-37.34L240.22 199v121.85c25.3 12.54 22.26 41.85 9.08 55a34.34 34.34 0 0 1-48.55 0c-17.57-17.6-11.07-46.91 11.25-56v-123c-20.8-8.51-24.6-30.74-18.64-45L142.57 101 8.45 235.14a28.86 28.86 0 0 0 0 40.81l195.61 195.6a28.86 28.86 0 0 0 40.8 0l194.69-194.69a28.86 28.86 0 0 0 0-40.81z"/></svg>
- </div>
- <div class="md-source__repository">
- mindspore-lab/mindocr
- </div>
- </a>
- </div>
-
- <ul class="md-nav__list" data-md-scrollfix>
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item md-nav__item--active">
-
- <input class="md-nav__toggle md-toggle" type="checkbox" id="__toc">
-
-
-
-
-
- <label class="md-nav__link md-nav__link--active" for="__toc">
-
-
- <span class="md-ellipsis">
- 主页
- </span>
-
-
- <span class="md-nav__icon md-icon"></span>
- </label>
-
- <a href="./" class="md-nav__link md-nav__link--active">
-
-
- <span class="md-ellipsis">
- 主页
- </span>
-
-
- </a>
-
-
-
- <nav class="md-nav md-nav--secondary" aria-label="Table of contents">
-
-
-
-
-
-
- <label class="md-nav__title" for="__toc">
- <span class="md-nav__icon md-icon"></span>
- Table of contents
- </label>
- <ul class="md-nav__list" data-md-component="toc" data-md-scrollfix>
-
- <li class="md-nav__item">
- <a href="#_1" class="md-nav__link">
- <span class="md-ellipsis">
- 简介
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_2" class="md-nav__link">
- <span class="md-ellipsis">
- 安装教程
- </span>
- </a>
-
- <nav class="md-nav" aria-label="安装教程">
- <ul class="md-nav__list">
-
- <li class="md-nav__item">
- <a href="#mindspore" class="md-nav__link">
- <span class="md-ellipsis">
- MindSpore相关环境准备
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_3" class="md-nav__link">
- <span class="md-ellipsis">
- 包依赖
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_4" class="md-nav__link">
- <span class="md-ellipsis">
- 通过源文件安装(推荐)
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#docker" class="md-nav__link">
- <span class="md-ellipsis">
- 通过docker安装
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#pypi" class="md-nav__link">
- <span class="md-ellipsis">
- 通过PyPI安装
- </span>
- </a>
-
- </li>
-
- </ul>
- </nav>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_5" class="md-nav__link">
- <span class="md-ellipsis">
- 快速开始
- </span>
- </a>
-
- <nav class="md-nav" aria-label="快速开始">
- <ul class="md-nav__list">
-
- <li class="md-nav__item">
- <a href="#1" class="md-nav__link">
- <span class="md-ellipsis">
- 1. 文字检测和识别示例
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#2-" class="md-nav__link">
- <span class="md-ellipsis">
- 2. 模型训练、评估与推理-快速指南
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#3-" class="md-nav__link">
- <span class="md-ellipsis">
- 3. 模型离线推理-快速指南
- </span>
- </a>
-
- </li>
-
- </ul>
- </nav>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_6" class="md-nav__link">
- <span class="md-ellipsis">
- 使用教程
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_7" class="md-nav__link">
- <span class="md-ellipsis">
- 模型列表
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_8" class="md-nav__link">
- <span class="md-ellipsis">
- 数据集列表
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_9" class="md-nav__link">
- <span class="md-ellipsis">
- 常见问题
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_10" class="md-nav__link">
- <span class="md-ellipsis">
- 重要信息
- </span>
- </a>
-
- <nav class="md-nav" aria-label="重要信息">
- <ul class="md-nav__list">
-
- <li class="md-nav__item">
- <a href="#_11" class="md-nav__link">
- <span class="md-ellipsis">
- 更新日志
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_12" class="md-nav__link">
- <span class="md-ellipsis">
- 如何贡献
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_13" class="md-nav__link">
- <span class="md-ellipsis">
- 许可
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_14" class="md-nav__link">
- <span class="md-ellipsis">
- 引用
- </span>
- </a>
-
- </li>
-
- </ul>
- </nav>
-
- </li>
-
- </ul>
-
- </nav>
-
- </li>
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item md-nav__item--nested">
-
-
-
- <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_2" >
-
-
- <label class="md-nav__link" for="__nav_2" id="__nav_2_label" tabindex="0">
-
-
- <span class="md-ellipsis">
- 模型仓库
- </span>
-
-
- <span class="md-nav__icon md-icon"></span>
- </label>
-
- <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_2_label" aria-expanded="false">
- <label class="md-nav__title" for="__nav_2">
- <span class="md-nav__icon md-icon"></span>
- 模型仓库
- </label>
- <ul class="md-nav__list" data-md-scrollfix>
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="mkdocs/modelzoo_training/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- 1. Training
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="inference/inference_quickstart/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- 2. Inference - MindOCR Models
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="inference/inference_thirdparty_quickstart/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- 3. Inference - Third-party Models
- </span>
-
-
- </a>
- </li>
-
-
-
-
- </ul>
- </nav>
-
- </li>
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item md-nav__item--nested">
-
-
-
- <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_3" >
-
-
- <label class="md-nav__link" for="__nav_3" id="__nav_3_label" tabindex="0">
-
-
- <span class="md-ellipsis">
- 教程
- </span>
-
-
- <span class="md-nav__icon md-icon"></span>
- </label>
-
- <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_3_label" aria-expanded="false">
- <label class="md-nav__title" for="__nav_3">
- <span class="md-nav__icon md-icon"></span>
- 教程
- </label>
- <ul class="md-nav__list" data-md-scrollfix>
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item md-nav__item--nested">
-
-
-
- <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_3_1" >
-
-
- <label class="md-nav__link" for="__nav_3_1" id="__nav_3_1_label" tabindex="0">
-
-
- <span class="md-ellipsis">
- 1. Datasets
- </span>
-
-
- <span class="md-nav__icon md-icon"></span>
- </label>
-
- <nav class="md-nav" data-md-level="2" aria-labelledby="__nav_3_1_label" aria-expanded="false">
- <label class="md-nav__title" for="__nav_3_1">
- <span class="md-nav__icon md-icon"></span>
- 1. Datasets
- </label>
- <ul class="md-nav__list" data-md-scrollfix>
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="datasets/converters/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Dataset Preparation
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="tutorials/transform_tutorial/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Data Transformation Mechanism
- </span>
-
-
- </a>
- </li>
-
-
-
-
- </ul>
- </nav>
-
- </li>
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item md-nav__item--nested">
-
-
-
- <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_3_2" >
-
-
- <label class="md-nav__link" for="__nav_3_2" id="__nav_3_2_label" tabindex="0">
-
-
- <span class="md-ellipsis">
- 2. Model Training
- </span>
-
-
- <span class="md-nav__icon md-icon"></span>
- </label>
-
- <nav class="md-nav" data-md-level="2" aria-labelledby="__nav_3_2_label" aria-expanded="false">
- <label class="md-nav__title" for="__nav_3_2">
- <span class="md-nav__icon md-icon"></span>
- 2. Model Training
- </label>
- <ul class="md-nav__list" data-md-scrollfix>
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="tutorials/yaml_configuration/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Yaml Configuration
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="tutorials/training_detection_custom_dataset/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Text Detection
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="tutorials/training_recognition_custom_dataset/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Text Recognition
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="tutorials/distribute_train/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Distributed Training
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="tutorials/advanced_train/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Advance Training
- </span>
-
-
- </a>
- </li>
-
-
-
-
- </ul>
- </nav>
-
- </li>
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item md-nav__item--nested">
-
-
-
- <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_3_3" >
-
-
- <label class="md-nav__link" for="__nav_3_3" id="__nav_3_3_label" tabindex="0">
-
-
- <span class="md-ellipsis">
- 3. Inference and Deployment
- </span>
-
-
- <span class="md-nav__icon md-icon"></span>
- </label>
-
- <nav class="md-nav" data-md-level="2" aria-labelledby="__nav_3_3_label" aria-expanded="false">
- <label class="md-nav__title" for="__nav_3_3">
- <span class="md-nav__icon md-icon"></span>
- 3. Inference and Deployment
- </label>
- <ul class="md-nav__list" data-md-scrollfix>
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="mkdocs/online_inference/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Python Online Inference
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="inference/inference_tutorial/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Python/C++ Inference on Ascend 310
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="inference/inference_quickstart/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- MindOCR Models Offline Inference - Quick Start
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="inference/inference_thirdparty_quickstart/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Third-party Models Offline Inference - Quick Start
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="inference/convert_tutorial/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Model Conversion
- </span>
-
-
- </a>
- </li>
-
-
-
-
- </ul>
- </nav>
-
- </li>
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item md-nav__item--nested">
-
-
-
- <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_3_4" >
-
-
- <label class="md-nav__link" for="__nav_3_4" id="__nav_3_4_label" tabindex="0">
-
-
- <span class="md-ellipsis">
- 4. Developer Guides
- </span>
-
-
- <span class="md-nav__icon md-icon"></span>
- </label>
-
- <nav class="md-nav" data-md-level="2" aria-labelledby="__nav_3_4_label" aria-expanded="false">
- <label class="md-nav__title" for="__nav_3_4">
- <span class="md-nav__icon md-icon"></span>
- 4. Developer Guides
- </label>
- <ul class="md-nav__list" data-md-scrollfix>
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="mkdocs/customize_dataset/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Customize Dataset
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="mkdocs/customize_data_transform/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Customize Data Transformation
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="mkdocs/customize_model/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Customize a New Model
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="mkdocs/customize_postprocess/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Customize Postprocessing Method
- </span>
-
-
- </a>
- </li>
-
-
-
-
- </ul>
- </nav>
-
- </li>
-
-
-
-
- </ul>
- </nav>
-
- </li>
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item md-nav__item--nested">
-
-
-
- <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_4" >
-
-
- <label class="md-nav__link" for="__nav_4" id="__nav_4_label" tabindex="0">
-
-
- <span class="md-ellipsis">
- Reference
- </span>
-
-
- <span class="md-nav__icon md-icon"></span>
- </label>
-
- <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_4_label" aria-expanded="false">
- <label class="md-nav__title" for="__nav_4">
- <span class="md-nav__icon md-icon"></span>
- Reference
- </label>
- <ul class="md-nav__list" data-md-scrollfix>
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="reference/api_doc/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- API doc
- </span>
-
-
- </a>
- </li>
-
-
-
-
- </ul>
- </nav>
-
- </li>
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item md-nav__item--nested">
-
-
-
- <input class="md-nav__toggle md-toggle " type="checkbox" id="__nav_5" >
-
-
- <label class="md-nav__link" for="__nav_5" id="__nav_5_label" tabindex="0">
-
-
- <span class="md-ellipsis">
- 说明
- </span>
-
-
- <span class="md-nav__icon md-icon"></span>
- </label>
-
- <nav class="md-nav" data-md-level="1" aria-labelledby="__nav_5_label" aria-expanded="false">
- <label class="md-nav__title" for="__nav_5">
- <span class="md-nav__icon md-icon"></span>
- 说明
- </label>
- <ul class="md-nav__list" data-md-scrollfix>
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="mkdocs/contributing/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- Contributing
- </span>
-
-
- </a>
- </li>
-
-
-
-
-
-
-
-
-
-
- <li class="md-nav__item">
- <a href="tutorials/frequently_asked_questions/" class="md-nav__link">
-
-
- <span class="md-ellipsis">
- 常见问题
- </span>
-
-
- </a>
- </li>
-
-
-
-
- </ul>
- </nav>
-
- </li>
-
-
-
- </ul>
- </nav>
- </div>
- </div>
- </div>
-
-
-
- <div class="md-sidebar md-sidebar--secondary" data-md-component="sidebar" data-md-type="toc" >
- <div class="md-sidebar__scrollwrap">
- <div class="md-sidebar__inner">
-
-
- <nav class="md-nav md-nav--secondary" aria-label="Table of contents">
-
-
-
-
-
-
- <label class="md-nav__title" for="__toc">
- <span class="md-nav__icon md-icon"></span>
- Table of contents
- </label>
- <ul class="md-nav__list" data-md-component="toc" data-md-scrollfix>
-
- <li class="md-nav__item">
- <a href="#_1" class="md-nav__link">
- <span class="md-ellipsis">
- 简介
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_2" class="md-nav__link">
- <span class="md-ellipsis">
- 安装教程
- </span>
- </a>
-
- <nav class="md-nav" aria-label="安装教程">
- <ul class="md-nav__list">
-
- <li class="md-nav__item">
- <a href="#mindspore" class="md-nav__link">
- <span class="md-ellipsis">
- MindSpore相关环境准备
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_3" class="md-nav__link">
- <span class="md-ellipsis">
- 包依赖
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_4" class="md-nav__link">
- <span class="md-ellipsis">
- 通过源文件安装(推荐)
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#docker" class="md-nav__link">
- <span class="md-ellipsis">
- 通过docker安装
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#pypi" class="md-nav__link">
- <span class="md-ellipsis">
- 通过PyPI安装
- </span>
- </a>
-
- </li>
-
- </ul>
- </nav>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_5" class="md-nav__link">
- <span class="md-ellipsis">
- 快速开始
- </span>
- </a>
-
- <nav class="md-nav" aria-label="快速开始">
- <ul class="md-nav__list">
-
- <li class="md-nav__item">
- <a href="#1" class="md-nav__link">
- <span class="md-ellipsis">
- 1. 文字检测和识别示例
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#2-" class="md-nav__link">
- <span class="md-ellipsis">
- 2. 模型训练、评估与推理-快速指南
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#3-" class="md-nav__link">
- <span class="md-ellipsis">
- 3. 模型离线推理-快速指南
- </span>
- </a>
-
- </li>
-
- </ul>
- </nav>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_6" class="md-nav__link">
- <span class="md-ellipsis">
- 使用教程
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_7" class="md-nav__link">
- <span class="md-ellipsis">
- 模型列表
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_8" class="md-nav__link">
- <span class="md-ellipsis">
- 数据集列表
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_9" class="md-nav__link">
- <span class="md-ellipsis">
- 常见问题
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_10" class="md-nav__link">
- <span class="md-ellipsis">
- 重要信息
- </span>
- </a>
-
- <nav class="md-nav" aria-label="重要信息">
- <ul class="md-nav__list">
-
- <li class="md-nav__item">
- <a href="#_11" class="md-nav__link">
- <span class="md-ellipsis">
- 更新日志
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_12" class="md-nav__link">
- <span class="md-ellipsis">
- 如何贡献
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_13" class="md-nav__link">
- <span class="md-ellipsis">
- 许可
- </span>
- </a>
-
- </li>
-
- <li class="md-nav__item">
- <a href="#_14" class="md-nav__link">
- <span class="md-ellipsis">
- 引用
- </span>
- </a>
-
- </li>
-
- </ul>
- </nav>
-
- </li>
-
- </ul>
-
- </nav>
- </div>
- </div>
- </div>
-
-
-
- <div class="md-content" data-md-component="content">
- <article class="md-content__inner md-typeset">
-
-
-
-
- <a href="https://github.com/mindspore-lab/mindocr/edit/master/docs/cn/index.md" title="Edit this page" class="md-content__button md-icon">
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M10 20H6V4h7v5h5v3.1l2-2V8l-6-6H6c-1.1 0-2 .9-2 2v16c0 1.1.9 2 2 2h4v-2m10.2-7c.1 0 .3.1.4.2l1.3 1.3c.2.2.2.6 0 .8l-1 1-2.1-2.1 1-1c.1-.1.2-.2.4-.2m0 3.9L14.1 23H12v-2.1l6.1-6.1 2.1 2.1Z"/></svg>
- </a>
-
-
-
-
-
- <a href="https://github.com/mindspore-lab/mindocr/raw/master/docs/cn/index.md" title="View source of this page" class="md-content__button md-icon">
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M17 18c.56 0 1 .44 1 1s-.44 1-1 1-1-.44-1-1 .44-1 1-1m0-3c-2.73 0-5.06 1.66-6 4 .94 2.34 3.27 4 6 4s5.06-1.66 6-4c-.94-2.34-3.27-4-6-4m0 6.5a2.5 2.5 0 0 1-2.5-2.5 2.5 2.5 0 0 1 2.5-2.5 2.5 2.5 0 0 1 2.5 2.5 2.5 2.5 0 0 1-2.5 2.5M9.27 20H6V4h7v5h5v4.07c.7.08 1.36.25 2 .49V8l-6-6H6a2 2 0 0 0-2 2v16a2 2 0 0 0 2 2h4.5a8.15 8.15 0 0 1-1.23-2Z"/></svg>
- </a>
-
-
-
- <!-- BEGIN INCLUDE ../../README_CN.md '<!--start-->' '<!--end-->' -->
-
- <div align="center">
- <h1 id="mindocr">MindOCR<a class="headerlink" href="#mindocr" title="Permanent link">¶</a></h1>
- <p><a href="../../README.md">English</a> | 中文</p>
- <p><a href="../..#简介">📝简介</a> |
- <a href="../..#安装教程">🔨安装教程</a> |
- <a href="../..#快速开始">🚀快速开始</a> |
- <a href="../..#使用教程">📚使用教程</a> |
- <a href="../..#模型列表">🎁模型列表</a> |
- <a href="../..#数据集列表">📰数据集列表</a> |
- <a href="../..#常见问题">📖常见问题</a> |
- <a href="../..#更新日志">🎉更新日志</a></p>
- </div>
- <h2 id="_1">简介<a class="headerlink" href="#_1" title="Permanent link">¶</a></h2>
- <p>MindOCR是一个基于<a href="https://www.mindspore.cn/en">MindSpore</a> 框架开发的OCR开源工具箱,集成系列主流文字检测识别的算法、模型,并提供易用的训练和推理工具,可以帮助用户快速开发和应用业界SoTA文本检测、文本识别模型,如DBNet/DBNet++和CRNN/SVTR,满足图像文档理解的需求。</p>
- <details open="open">
- <summary> 主要特性 </summary>
- <ul>
- <li><strong>模块化设计</strong>: MindOCR将OCR任务解耦成多个可配置模块,用户只需修改几行代码,就可以轻松地在定制化的数据和模型上配置训练、评估的全流程;</li>
- <li><strong>高性能</strong>: MindOCR提供的预训练权重和训练方法可以使其达到OCR任务上具有竞争力的表现;</li>
- <li><strong>易用性</strong>: MindOCR提供易用工具帮助在真实世界数据中进行文本的检测和识别。</li>
- </ul>
- </details>
- <h2 id="_2">安装教程<a class="headerlink" href="#_2" title="Permanent link">¶</a></h2>
- <details open="open">
- <h4 id="mindspore">MindSpore相关环境准备<a class="headerlink" href="#mindspore" title="Permanent link">¶</a></h4>
- <p>MindOCR基于MindSpore AI框架(支持CPU/GPU/NPU)开发,并适配以下框架版本。安装方式请参见下方的安装链接。</p>
- <ul>
- <li>mindspore >= 2.2.0 [<a href="https://www.mindspore.cn/install">安装</a>]</li>
- <li>python >= 3.7</li>
- <li>openmpi 4.0.3 (用于分布式训练与验证) [<a href="https://www.open-mpi.org/software/ompi/v4.0/">安装</a>]</li>
- <li>mindspore lite (用于离线推理) >= 2.2.0 [<a href="inference/environment/">安装</a>]</li>
- </ul>
- <h4 id="_3">包依赖<a class="headerlink" href="#_3" title="Permanent link">¶</a></h4>
- <div class="highlight"><pre><span></span><code>pip<span class="w"> </span>install<span class="w"> </span>-r<span class="w"> </span>requirements.txt
- </code></pre></div>
- <h4 id="_4">通过源文件安装(推荐)<a class="headerlink" href="#_4" title="Permanent link">¶</a></h4>
- <div class="highlight"><pre><span></span><code>git<span class="w"> </span>clone<span class="w"> </span>https://github.com/mindspore-lab/mindocr.git
- <span class="nb">cd</span><span class="w"> </span>mindocr
- pip<span class="w"> </span>install<span class="w"> </span>-e<span class="w"> </span>.
- </code></pre></div>
- <blockquote>
- <p>使用 <code>-e</code> 代表可编辑模式,可以帮助解决潜在的模块导入问题。</p>
- </blockquote>
- <h4 id="docker">通过docker安装<a class="headerlink" href="#docker" title="Permanent link">¶</a></h4>
- <p>目前提供的docker,环境信息如下
- - 操作系统版本:Euler2.8
- - CANN版本:7.0
- - Python版本:3.9
- - MindSpore 版本:2.2.10
- - MindSpore Lite 版本:2.2.10</p>
- <p>使用docker安装,根据以下步骤:</p>
- <ol>
- <li>下载docker<ul>
- <li>910:
- <div class="highlight"><pre><span></span><code>docker<span class="w"> </span>pull<span class="w"> </span>swr.cn-central-221.ovaijisuan.com/mindocr/mindocr_dev_910_ms_2_2_10_cann7_0_py39:v1
- </code></pre></div></li>
- <li>910*:
- <div class="highlight"><pre><span></span><code>docker<span class="w"> </span>pull<span class="w"> </span>swr.cn-central-221.ovaijisuan.com/mindocr/mindocr_dev_ms_2_2_10_cann7_0_py39:v1
- </code></pre></div></li>
- </ul>
- </li>
- <li>
- <p>新建容器
- <div class="highlight"><pre><span></span><code><span class="nv">docker_name</span><span class="o">=</span><span class="s2">"temp_mindocr"</span>
- <span class="c1"># 910</span>
- <span class="nv">image_name</span><span class="o">=</span><span class="s2">"swr.cn-central-221.ovaijisuan.com/mindocr/mindocr_dev_910_ms_2_2_10_cann7_0_py39:v1"</span>
- <span class="c1"># 910*</span>
- <span class="nv">image_name</span><span class="o">=</span><span class="s2">"swr.cn-central-221.ovaijisuan.com/mindocr/mindocr_dev_ms_2_2_10_cann7_0_py39:v1"</span>
-
- docker<span class="w"> </span>run<span class="w"> </span>--privileged<span class="w"> </span>--name<span class="w"> </span><span class="si">${</span><span class="nv">docker_name</span><span class="si">}</span><span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--tmpfs<span class="w"> </span>/tmp<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--tmpfs<span class="w"> </span>/run<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>-v<span class="w"> </span>/sys/fs/cgroup:/sys/fs/cgroup:ro<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--device<span class="o">=</span>/dev/davinci1<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--device<span class="o">=</span>/dev/davinci2<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--device<span class="o">=</span>/dev/davinci3<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--device<span class="o">=</span>/dev/davinci4<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--device<span class="o">=</span>/dev/davinci5<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--device<span class="o">=</span>/dev/davinci6<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--device<span class="o">=</span>/dev/davinci7<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--device<span class="o">=</span>/dev/davinci_manager<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--device<span class="o">=</span>/dev/hisi_hdc<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--device<span class="o">=</span>/dev/devmm_svm<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>-v<span class="w"> </span>/etc/localtime:/etc/localtime<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>-v<span class="w"> </span>/usr/local/Ascend/driver:/usr/local/Ascend/driver<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>-v<span class="w"> </span>/usr/local/bin/npu-smi:/usr/local/bin/npu-smi<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--shm-size<span class="w"> </span>800g<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--cpus<span class="w"> </span><span class="m">96</span><span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--security-opt<span class="w"> </span><span class="nv">seccomp</span><span class="o">=</span>unconfined<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--network<span class="o">=</span>bridge<span class="w"> </span>-itd<span class="w"> </span><span class="si">${</span><span class="nv">image_name</span><span class="si">}</span><span class="w"> </span>bash
- </code></pre></div></p>
- </li>
- <li>
- <p>进入容器
- <div class="highlight"><pre><span></span><code><span class="c1"># 设置docker id</span>
- <span class="nv">container_id</span><span class="o">=</span><span class="s2">"your docker id"</span>
- docker<span class="w"> </span><span class="nb">exec</span><span class="w"> </span>-it<span class="w"> </span>--user<span class="w"> </span>root<span class="w"> </span><span class="nv">$container_id</span><span class="w"> </span>bash
- </code></pre></div></p>
- </li>
- <li>
- <p>设置环境变量
- 进入容器后,设置环境变量:
- <div class="highlight"><pre><span></span><code><span class="nb">source</span><span class="w"> </span>env_setup.sh
- </code></pre></div></p>
- </li>
- </ol>
- <h4 id="pypi">通过PyPI安装<a class="headerlink" href="#pypi" title="Permanent link">¶</a></h4>
- <div class="highlight"><pre><span></span><code>pip<span class="w"> </span>install<span class="w"> </span>mindocr
- </code></pre></div>
- <blockquote>
- <p>由于此项目正在积极开发中,从PyPI安装的版本目前已过期,我们将很快更新,敬请期待。</p>
- </blockquote>
- </details>
- <h2 id="_5">快速开始<a class="headerlink" href="#_5" title="Permanent link">¶</a></h2>
- <h3 id="1">1. 文字检测和识别示例<a class="headerlink" href="#1" title="Permanent link">¶</a></h3>
- <p>安装完MindOCR后,我们就很方便地进行任意图像的文本检测和识别,如下。</p>
- <div class="highlight"><pre><span></span><code>python<span class="w"> </span>tools/infer/text/predict_system.py<span class="w"> </span>--image_dir<span class="w"> </span><span class="o">{</span>path_to_img<span class="w"> </span>or<span class="w"> </span>dir_to_imgs<span class="o">}</span><span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--det_algorithm<span class="w"> </span>DB++<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--rec_algorithm<span class="w"> </span>CRNN
- </code></pre></div>
- <p>运行结束后,结果将被默认保存在<code>./inference_results</code>路径,可视化结果如下:</p>
- <p align="center">
- <img src="https://github.com/SamitHuang/mindocr-1/assets/8156835/c1f53970-8618-4039-994f-9f6dc1eee1dd" width=600 />
- </p>
- <p align="center">
- <em> 文本检测、识别结果可视化 </em>
- </p>
-
- <p>可以看到图像中的文字块均被检测出来并正确识别。更详细的用法介绍,请参考推理<a href="../..#使用教程">教程</a>。</p>
- <h3 id="2-">2. 模型训练、评估与推理-快速指南<a class="headerlink" href="#2-" title="Permanent link">¶</a></h3>
- <p>使用<code>tools/train.py</code>脚本可以进行OCR模型训练,该脚本可支持文本检测和识别模型训练。
- <div class="highlight"><pre><span></span><code>python<span class="w"> </span>tools/train.py<span class="w"> </span>--config<span class="w"> </span><span class="o">{</span>path/to/model_config.yaml<span class="o">}</span>
- </code></pre></div>
- <code>--config</code> 参数用于指定yaml文件的路径,该文件定义要训练的模型和训练策略,包括数据处理流程、优化器、学习率调度器等。</p>
- <p>MindOCR在<code>configs</code>文件夹中提供系列SoTA的OCR模型及其训练策略,用户可以快速将其适配到自己的任务或数据集上,参考例子如下</p>
- <p><div class="highlight"><pre><span></span><code><span class="c1"># train text detection model DBNet++ on icdar15 dataset</span>
- python<span class="w"> </span>tools/train.py<span class="w"> </span>--config<span class="w"> </span>configs/det/dbnet/dbpp_r50_icdar15.yaml
- </code></pre></div>
- <div class="highlight"><pre><span></span><code><span class="c1"># train text recognition model CRNN on icdar15 dataset</span>
- python<span class="w"> </span>tools/train.py<span class="w"> </span>--config<span class="w"> </span>configs/rec/crnn/crnn_icdar15.yaml
- </code></pre></div></p>
- <p>使用<code>tools/eval.py</code> 脚本可以评估已训练好的模型,如下所示:
- <div class="highlight"><pre><span></span><code>python<span class="w"> </span>tools/eval.py<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--config<span class="w"> </span><span class="o">{</span>path/to/model_config.yaml<span class="o">}</span><span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--opt<span class="w"> </span>eval.dataset_root<span class="o">={</span>path/to/your_dataset<span class="o">}</span><span class="w"> </span>eval.ckpt_load_path<span class="o">={</span>path/to/ckpt_file<span class="o">}</span>
- </code></pre></div></p>
- <p>使用<code>tools/infer/text/predict_system.py</code> 脚本可进行模型的在线推理,如下所示:
- <div class="highlight"><pre><span></span><code>python<span class="w"> </span>tools/infer/text/predict_system.py<span class="w"> </span>--image_dir<span class="w"> </span><span class="o">{</span>path_to_img<span class="w"> </span>or<span class="w"> </span>dir_to_imgs<span class="o">}</span><span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--det_algorithm<span class="w"> </span>DB++<span class="w"> </span><span class="se">\</span>
- <span class="w"> </span>--rec_algorithm<span class="w"> </span>CRNN
- </code></pre></div></p>
- <p>更多使用方法,请参考<a href="../..#使用教程">使用教程</a>中的模型训练、推理章节。</p>
- <h3 id="3-">3. 模型离线推理-快速指南<a class="headerlink" href="#3-" title="Permanent link">¶</a></h3>
- <p>你可以在MindOCR中对**MindOCR原生模型**或**第三方模型**(如PaddleOCR、MMOCR等)进行MindSpore Lite推理。请参考以下文档
- - <a href="inference/inference_tutorial/">基于Python/C++和昇腾310的OCR推理</a>
- - <a href="inference/inference_quickstart/">MindOCR原生模型离线推理 - 快速开始</a>
- - <a href="inference/inference_thirdparty_quickstart/">第三方模型离线推理 - 快速开始</a></p>
- <h2 id="_6">使用教程<a class="headerlink" href="#_6" title="Permanent link">¶</a></h2>
- <ul>
- <li>数据集<ul>
- <li><a href="datasets/converters/">数据集准备</a></li>
- <li><a href="tutorials/transform_tutorial/">数据增强策略</a></li>
- </ul>
- </li>
- <li>模型训练<ul>
- <li><a href="tutorials/yaml_configuration/">Yaml配置文件</a></li>
- <li><a href="tutorials/training_detection_custom_dataset/">文本检测</a></li>
- <li><a href="tutorials/training_recognition_custom_dataset/">文本识别</a></li>
- <li><a href="tutorials/distribute_train/">分布式训练</a></li>
- <li><a href="tutorials/advanced_train/">进阶技巧:梯度累积,EMA,断点续训等</a></li>
- </ul>
- </li>
- <li>使用MindSpore进行在线推理<ul>
- <li><a href="../../tools/infer/text/README.md">基于Python的OCR在线推理</a></li>
- </ul>
- </li>
- <li>使用MindSpore Lite进行离线推理<ul>
- <li><a href="inference/inference_tutorial/">基于Python/C++和昇腾310的OCR推理</a></li>
- <li><a href="inference/inference_quickstart/">MindOCR原生模型离线推理 - 快速开始</a></li>
- <li><a href="inference/inference_thirdparty_quickstart/">第三方模型离线推理 - 快速开始</a></li>
- </ul>
- </li>
- <li>开发者指南<ul>
- <li><a href="../../mindocr/data/README.md">如何自定义数据集</a></li>
- <li><a href="../../mindocr/data/transforms/README.md">如何自定义数据增强方法</a></li>
- <li><a href="../../mindocr/models/README.md">如何创建新的OCR模型</a></li>
- <li><a href="../../mindocr/postprocess/README.md">如何自定义后处理方法</a></li>
- </ul>
- </li>
- </ul>
- <h2 id="_7">模型列表<a class="headerlink" href="#_7" title="Permanent link">¶</a></h2>
- <details open="open">
- <summary>文本检测</summary>
- <ul class="task-list">
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/det/dbnet/README_CN.md">DBNet</a> (AAAI'2020)</li>
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/det/dbnet/README_CN.md">DBNet++</a> (TPAMI'2022)</li>
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/det/psenet/README_CN.md">PSENet</a> (CVPR'2019)</li>
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/det/east/README_CN.md">EAST</a>(CVPR'2017)</li>
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/det/fcenet/README_CN.md">FCENet</a> (CVPR'2021)</li>
- </ul>
- </details>
- <details open="open">
- <summary>文本识别</summary>
- <ul class="task-list">
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/rec/crnn/README_CN.md">CRNN</a> (TPAMI'2016)</li>
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/rec/rare/README_CN.md">CRNN-Seq2Seq/RARE</a> (CVPR'2016)</li>
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/rec/svtr/README_CN.md">SVTR</a> (IJCAI'2022)</li>
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/rec/master/README_CN.md">MASTER</a> (PR'2019)</li>
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/rec/visionlan/README_CN.md">VISIONLAN</a> (ICCV'2021)</li>
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/rec/robustscanner/README_CN.md">RobustScanner</a> (ECCV'2020)</li>
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/rec/abinet/README_CN.md">ABINet</a> (CVPR'2021)</li>
- </ul>
- </details>
- <details open="open">
- <summary>版面分析</summary>
- <ul class="task-list">
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/layout/yolov8/README_CN.md">YOLOv8</a> (<a href="https://github.com/ultralytics/ultralytics">Ultralytics Inc.</a>)</li>
- </ul>
- </details>
- <details open="open">
- <summary>关键信息抽取</summary>
- <ul class="task-list">
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/kie/vi_layoutxlm/README_CN.md">LayoutXLM</a> (arXiv'2021)</li>
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/kie/layoutlmv3/README_CN.md">LayoutLMv3</a> (arXiv'2022)</li>
- </ul>
- </details>
- <details open="open">
- <summary>表格识别</summary>
- <ul class="task-list">
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/table/README_CN.md">TableMaster</a> (arXiv'2021)</li>
- </ul>
- </details>
- <details open="open">
- <summary>OCR大模型</summary>
- <ul class="task-list">
- <li class="task-list-item"><label class="task-list-control"><input type="checkbox" disabled checked/><span class="task-list-indicator"></span></label> <a href="../../configs/llm/vary/README_CN.md">Vary</a> (arXiv'2023)</li>
- </ul>
- </details>
- <p>关于以上模型的具体训练方法和结果,请参见<a href="../../configs">configs</a>下各模型子目录的readme文档。</p>
- <p>关于<a href="https://www.mindspore.cn/lite">MindSpore Lite</a>和<a href="https://www.hiascend.com/document/detail/zh/canncommercial/63RC1/inferapplicationdev/aclcppdevg/aclcppdevg_000004.html">ACL</a>模型推理的支持列表,
- 请参见<a href="inference/inference_quickstart/">MindOCR原生模型推理支持列表</a> 和 <a href="inference/inference_thirdparty_quickstart/">第三方模型推理支持列表</a>(如PaddleOCR、MMOCR等)。</p>
- <h2 id="_8">数据集列表<a class="headerlink" href="#_8" title="Permanent link">¶</a></h2>
- <p>MindOCR提供了<a href="../../tools/dataset_converters">数据格式转换工具</a> ,以支持不同格式的OCR数据集,支持用户自定义的数据集。
- 当前已在模型训练评估中验证过的公开OCR数据集如下。</p>
- <details close="close">
- <summary>通用OCR数据集</summary>
- <ul>
- <li><a href="https://rrc.cvc.uab.es/?ch=1">Born-Digital Images</a> [<a href="datasets/borndigital/">download</a>]</li>
- <li><a href="http://www.nlpr.ia.ac.cn/pal/CASIA10K.html">CASIA-10K</a> [<a href="datasets/casia10k/">download</a>]</li>
- <li><a href="https://github.com/detectRecog/CCPD">CCPD</a> [<a href="datasets/ccpd/">download</a>]</li>
- <li><a href="https://github.com/FudanVI/benchmarking-chinese-text-recognition">Chinese Text Recognition Benchmark</a> [<a href="https://arxiv.org/abs/2112.15093">paper</a>] [<a href="datasets/chinese_text_recognition/">download</a>]</li>
- <li><a href="https://rrc.cvc.uab.es/?ch=5">COCO-Text</a> [<a href="datasets/cocotext/">download</a>]</li>
- <li><a href="https://ctwdataset.github.io/">CTW</a> [<a href="datasets/ctw/">download</a>]</li>
- <li><a href="https://rrc.cvc.uab.es/?ch=4">ICDAR2015</a> [<a href="https://rrc.cvc.uab.es/files/short_rrc_2015.pdf">paper</a>] [<a href="datasets/icdar2015/">download</a>]</li>
- <li><a href="https://rrc.cvc.uab.es/?ch=14">ICDAR2019 ArT</a> [<a href="datasets/ic19_art/">download</a>]</li>
- <li><a href="https://rrc.cvc.uab.es/?ch=16">LSVT</a> [<a href="datasets/lsvt/">download</a>]</li>
- <li><a href="https://rrc.cvc.uab.es/?ch=8">MLT2017</a> [<a href="https://ieeexplore.ieee.org/abstract/document/8270168">paper</a>] [<a href="datasets/mlt2017/">download</a>]</li>
- <li><a href="http://www.iapr-tc11.org/mediawiki/index.php/MSRA_Text_Detection_500_Database_(MSRA-TD500)">MSRA-TD500</a> [<a href="https://ieeexplore.ieee.org/abstract/document/6247787">paper</a>] [<a href="datasets/td500/">download</a>]</li>
- <li><a href="https://tianchi.aliyun.com/competition/entrance/231651/introduction">MTWI-2018</a> [<a href="datasets/mtwi2018/">download</a>]</li>
- <li><a href="https://rctw.vlrlab.net/">RCTW-17</a> [<a href="datasets/rctw17/">download</a>]</li>
- <li><a href="https://rrc.cvc.uab.es/?ch=12">ReCTS</a> [<a href="datasets/rects/">download</a>]</li>
- <li><a href="https://github.com/Yuliang-Liu/Curve-Text-Detector">SCUT-CTW1500</a> [<a href="https://www.sciencedirect.com/science/article/pii/S0031320319300664">paper</a>] [<a href="datasets/ctw1500/">download</a>]</li>
- <li><a href="https://rrc.cvc.uab.es/?ch=13">SROIE</a> [<a href="datasets/sroie/">download</a>]</li>
- <li><a href="http://www.iapr-tc11.org/mediawiki/index.php/The_Street_View_Text_Dataset">SVT</a> [<a href="datasets/svt/">download</a>]</li>
- <li><a href="https://github.com/aim-uofa/AdelaiDet">SynText150k</a> [<a href="https://arxiv.org/abs/2002.10200">paper</a>] [<a href="datasets/syntext150k/">download</a>]</li>
- <li><a href="https://www.robots.ox.ac.uk/~vgg/data/scenetext/">SynthText</a> [<a href="https://www.robots.ox.ac.uk/~vgg/publications/2016/Gupta16/">paper</a>] [<a href="datasets/synthtext/">download</a>]</li>
- <li><a href="https://textvqa.org/textocr/">TextOCR</a> [<a href="datasets/textocr/">download</a>]</li>
- <li><a href="https://github.com/cs-chan/Total-Text-Dataset/tree/master/Dataset">Total-Text</a> [<a href="https://arxiv.org/abs/1710.10400">paper</a>] [<a href="datasets/totaltext/">download</a>]</li>
- </ul>
- </details>
- <details close="close">
- <summary>版面分析数据集</summary>
- <ul>
- <li><a href="https://github.com/ibm-aur-nlp/PubLayNet">PublayNet</a> [<a href="https://arxiv.org/abs/1908.07836">paper</a>] [<a href="https://dax-cdn.cdn.appdomain.cloud/dax-publaynet/1.0.0/publaynet.tar.gz">download</a>]</li>
- </ul>
- </details>
- <details close="close">
- <summary>关键信息抽取数据集</summary>
- <ul>
- <li><a href="https://github.com/doc-analysis/XFUND">XFUND</a> [<a href="https://aclanthology.org/2022.findings-acl.253/">paper</a>] [<a href="https://github.com/doc-analysis/XFUND/releases/tag/v1.0">download</a>]</li>
- </ul>
- </details>
- <details close="close">
- <summary>表格识别数据集</summary>
- <ul>
- <li><a href="https://github.com/ibm-aur-nlp/PubTabNet">PubTabNet</a> [<a href="https://arxiv.org/pdf/1911.10683.pdf">paper</a>] [<a href="https://dax-cdn.cdn.appdomain.cloud/dax-pubtabnet/2.0.0/pubtabnet.tar.gz">download</a>]</li>
- </ul>
- </details>
- <p>我们会在更多的数据集上进行模型训练和验证。该列表将持续更新。</p>
- <h2 id="_9">常见问题<a class="headerlink" href="#_9" title="Permanent link">¶</a></h2>
- <p>关于配置环境、使用mindocr遇到的高频问题,可以参考<a href="tutorials/frequently_asked_questions/">常见问题</a>。</p>
- <h2 id="_10">重要信息<a class="headerlink" href="#_10" title="Permanent link">¶</a></h2>
- <h3 id="_11">更新日志<a class="headerlink" href="#_11" title="Permanent link">¶</a></h3>
- <details close="close">
- <summary>详细</summary>
- <ul>
- <li>
- <p>2023/04/01
- 1. 增加新模型</p>
- <ul>
- <li>关键信息抽取<a href="../../configs/kie/layoutlmv3/">LayoutLMv3</a></li>
- </ul>
- </li>
- <li>
- <p>2024/03/20
- 1. 增加新模型</p>
- <ul>
- <li>OCR大模型<a href="../../configs/llm/vary/vary_toy.yaml">Vary-toy</a>,支持基于通义千问1.8B LLM的检测和OCR功能</li>
- </ul>
- </li>
- <li>
- <p>2023/12/25
- 1. 增加新模型</p>
- <ul>
- <li>表格识别<a href="../../configs/table/table_master.yaml">TableMaster</a>
- 2. 添加更多基准数据集及其结果</li>
- <li><a href="../../configs/table/README_CN.md">PubTabNet</a></li>
- </ul>
- </li>
- <li>
- <p>2023/12/14
- 1. 增加新模型</p>
- <ul>
- <li>关键信息抽取<a href="../../configs/kie/layoutxlm">LayoutXLM</a></li>
- <li>关键信息抽取<a href="../../configs/kie/vi_layoutxlm">VI-LayoutXLM</a></li>
- <li>文本检测<a href="../../configs/det/dbnet/db_mobilenetv3_ppocrv3.yaml">PP-OCRv3 DBNet</a>和文本识别<a href="../../configs/rec/svtr/svtr_ppocrv3_ch.yaml">PP-OCRv3 SVTR</a>,支持在线推理和微调训练
- 2. 添加更多基准数据集及其结果</li>
- <li><a href="../../configs/kie/vi_layoutxlm/README_CN.md">XFUND</a>
- 3. 昇腾910硬件多规格支持:DBNet ResNet-50、DBNet++ ResNet-50、CRNN VGG7、SVTR-Tiny、FCENet、ABINet</li>
- </ul>
- </li>
- <li>2023/11/28
- 1. 增加支持PP-OCRv4模型离线推理<ul>
- <li>文本检测 <a href="../../deploy/py_infer/src/configs/det/ppocr/ch_PP-OCRv4_det_cml.yaml">PP-OCRv4 DBNet</a>和文本识别 <a href="../../deploy/py_infer/src/configs/rec/ppocr/ch_PP-OCRv4_rec_distillation.yaml">PP-OCRv4 CRNN</a>,支持离线推理
- 2. 修复第三方模型离线推理bug</li>
- </ul>
- </li>
- <li>2023/11/17
- 1. 增加新模型<ul>
- <li>版面分析<a href="../../configs/layout/yolov8">YOLOv8</a>
- 2. 添加更多基准数据集及其结果</li>
- <li><a href="../../configs/layout/yolov8/README_CN.md">PublayNet</a></li>
- </ul>
- </li>
- <li>2023/07/06
- 1. 增加新模型<ul>
- <li>文本识别 <a href="../../configs/rec/robustscanner">RobustScanner</a></li>
- </ul>
- </li>
- <li>2023/07/05
- 1. 增加新模型<ul>
- <li>文本识别 <a href="../../configs/rec/visionlan">VISIONLAN</a></li>
- </ul>
- </li>
- <li>2023/06/29
- 1. 新增2个SoTA模型<ul>
- <li>文本检测 <a href="../../configs/det/fcenet">FCENet</a></li>
- <li>文本识别 <a href="../../configs/rec/master">MASTER</a></li>
- </ul>
- </li>
- <li>
- <p>2023/06/07
- 1. 增加新模型</p>
- <ul>
- <li>文本检测 <a href="../../configs/det/psenet">PSENet</a></li>
- <li>文本检测 <a href="../../configs/det/east">EAST</a></li>
- <li>文本识别 <a href="../../configs/rec/svtr">SVTR</a>
- 2. 添加更多基准数据集及其结果</li>
- <li><a href="datasets/totaltext/">totaltext</a></li>
- <li><a href="datasets/mlt2017/">mlt2017</a></li>
- <li><a href="datasets/chinese_text_recognition/">chinese_text_recognition</a>
- 3. 增加断点续训(resume training)功能,可在训练意外中断时使用。如需使用,请在配置文件中<code>model</code>字段下增加<code>resume</code>参数,允许传入具体路径<code>resume: /path/to/train_resume.ckpt</code>或者通过设置<code>resume: True</code>来加载在ckpt_save_dir下保存的trian_resume.ckpt
- 4. 改进检测模块的后处理部分:默认情况下,将检测到的文本多边形重新缩放到原始图像空间,可以通过在<code>eval.dataset.output_columns</code>列表中增加"shape_list"实现。
- 5. 重构在线推理以支持更多模型,详情请参见<a href="../../tools/infer/text/README.md">README.md</a> 。</li>
- </ul>
- </li>
- <li>
- <p>2023/05/15
- 1. 增加新模型</p>
- <ul>
- <li>文本检测 <a href="../../configs/det/dbnet">DBNet++</a></li>
- <li>文本识别 <a href="../../configs/rec/rare">CRNN-Seq2Seq</a></li>
- <li>在SynthText数据集上预训练的<a href="https://download.mindspore.cn/toolkits/mindocr/dbnet/dbnet_resnet50_synthtext-40655acb.ckpt">DBNet</a>
- 2. 添加更多基准数据集及其结果</li>
- <li><a href="datasets/synthtext/">SynthText</a>, <a href="datasets/td500/">MSRA-TD500</a>, <a href="datasets/ctw1500/">CTW1500</a></li>
- <li>DBNet的更多基准结果可以<a href="../../configs/det/dbnet/README_CN.md">在此找到</a>.
- 3. 添加用于保存前k个checkpoint的checkpoint manager并改进日志。
- 4. Python推理代码重构。
- 5. Bug修复:对大型数据集使用平均损失meter,在AMP训练中对ctcloss禁用<code>pred_cast_fp32</code>,修复存在无效多边形的错误。</li>
- </ul>
- </li>
- <li>
- <p>2023/05/04
- 1. 支持加载自定义的预训练checkpoint, 通过在yaml配置中将<code>model-pretrained</code>设置为checkpoint url或本地路径来使用。
- 2. 支持设置执行包括旋转和翻转在内的数据增强操作的概率。
- 3. 为模型训练添加EMA功能,可以通过在yaml配置中设置<code>train-ema</code>(默认值:False)和<code>train-ema_decay</code>来启用。
- 4. 参数修改:<code>num_columns_to_net</code> -> <code>net_input_column_index</code>: 输入网络的columns数量改为输入网络的columns索引
- 5. 参数修改:<code>num_columns_of_labels</code> -> <code>label_column_index</code>: 用索引替换数量,以表示label的位置。</p>
- </li>
- <li>
- <p>2023/04/21
- 1. 添加参数分组以支持训练中的正则化。用法:在yaml config中添加<code>grouping_strategy</code>参数以选择预定义的分组策略,或使用<code>no_weight_decay_params</code>参数选择要从权重衰减中排除的层(例如,bias、norm)。示例可参考<code>configs/rec/crn/crnn_icdar15.yaml</code>
- 2. 添加梯度累积,支持大批量训练。用法:在yaml配置中添加<code>gradient_accumulation_steps</code>,全局批量大小=batch_size * devices * gradient_aaccumulation_steps。示例可参考<code>configs/rec/crn/crnn_icdar15.yaml</code>
- 3. 添加梯度裁剪,支持训练稳定。通过在yaml配置中将<code>grad_clip</code>设置为True来启用。</p>
- </li>
- <li>
- <p>2023/03/23
- 1. 增加dynamic loss scaler支持, 且与drop overflow update兼容。如需使用, 请在配置文件中增加<code>loss_scale</code>字段并将<code>type</code>参数设为<code>dynamic</code>,参考例子请见<code>configs/rec/crnn/crnn_icdar15.yaml</code></p>
- </li>
- <li>
- <p>2023/03/20
- 1. 参数名修改:<code>output_keys</code> -> <code>output_columns</code>;<code>num_keys_to_net</code> -> <code>num_columns_to_net</code>;
- 2. 更新数据流程。</p>
- </li>
- <li>
- <p>2023/03/13
- 1. 增加系统测试和CI工作流;
- 2. 增加modelarts平台适配器,使得支持在OpenI平台上训练,在OpenI平台上训练需要以下步骤:
- <div class="highlight"><pre><span></span><code> i) 在OpenI云平台上创建一个训练任务;
- ii) 在网页上关联数据集,如ic15_mindocr;
- iii) 增加 `config` 参数,在网页的UI界面配置yaml文件路径,如'/home/work/user-job-dir/V0001/configs/rec/test.yaml';
- iv) 在网页的UI界面增加运行参数`enable_modelarts`并将其设置为True;
- v) 填写其他项并启动训练任务。
- </code></pre></div></p>
- </li>
- </ul>
- </details>
- <h3 id="_12">如何贡献<a class="headerlink" href="#_12" title="Permanent link">¶</a></h3>
- <p>我们欢迎包括问题单和PR在内的所有贡献,来让MindOCR变得更好。</p>
- <p>请参考<a href="../../CONTRIBUTING.md">CONTRIBUTING.md</a>作为贡献指南,请按照<a href="../../mindocr/models/README.md">Model Template and Guideline</a>的指引贡献一个适配所有接口的模型,多谢合作。</p>
- <h3 id="_13">许可<a class="headerlink" href="#_13" title="Permanent link">¶</a></h3>
- <p>本项目遵从<a href="../../LICENSE">Apache License 2.0</a>开源许可。</p>
- <h3 id="_14">引用<a class="headerlink" href="#_14" title="Permanent link">¶</a></h3>
- <p>如果本项目对您的研究有帮助,请考虑引用:</p>
- <div class="highlight"><pre><span></span><code>@misc<span class="nb">{</span>MindSpore OCR 2023,
- title=<span class="nb">{{</span>MindSpore OCR <span class="nb">}</span>:MindSpore OCR Toolbox<span class="nb">}</span>,
- author=<span class="nb">{</span>MindSpore Team<span class="nb">}</span>,
- howpublished = <span class="nb">{</span><span class="k">\url</span><span class="nb">{</span>https://github.com/mindspore-lab/mindocr/<span class="nb">}}</span>,
- year=<span class="nb">{</span>2023<span class="nb">}</span>
- <span class="nb">}</span>
- </code></pre></div>
- <!-- END INCLUDE -->
-
-
-
-
-
-
-
-
-
-
-
-
-
- </article>
- </div>
-
-
- <script>var tabs=__md_get("__tabs");if(Array.isArray(tabs))e:for(var set of document.querySelectorAll(".tabbed-set")){var tab,labels=set.querySelector(".tabbed-labels");for(tab of tabs)for(var label of labels.getElementsByTagName("label"))if(label.innerText.trim()===tab){var input=document.getElementById(label.htmlFor);input.checked=!0;continue e}}</script>
-
- <script>var target=document.getElementById(location.hash.slice(1));target&&target.name&&(target.checked=target.name.startsWith("__tabbed_"))</script>
- </div>
-
- <button type="button" class="md-top md-icon" data-md-component="top" hidden>
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M13 20h-2V8l-5.5 5.5-1.42-1.42L12 4.16l7.92 7.92-1.42 1.42L13 8v12Z"/></svg>
- Back to top
- </button>
-
- </main>
-
- <footer class="md-footer">
-
-
-
- <nav class="md-footer__inner md-grid" aria-label="Footer" >
-
-
-
- <a href="mkdocs/modelzoo_training/" class="md-footer__link md-footer__link--next" aria-label="Next: 1. Training">
- <div class="md-footer__title">
- <span class="md-footer__direction">
- Next
- </span>
- <div class="md-ellipsis">
- 1. Training
- </div>
- </div>
- <div class="md-footer__button md-icon">
-
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path d="M4 11v2h12l-5.5 5.5 1.42 1.42L19.84 12l-7.92-7.92L10.5 5.5 16 11H4Z"/></svg>
- </div>
- </a>
-
- </nav>
-
-
- <div class="md-footer-meta md-typeset">
- <div class="md-footer-meta__inner md-grid">
- <div class="md-copyright">
-
- <div class="md-copyright__highlight">
- Copyright © 2022 - 2024 MindSpore Lab
- </div>
-
-
- </div>
-
- <div class="md-social">
-
-
-
-
-
- <a href="mailto:mindspore-lab@huawei.com" target="_blank" rel="noopener" title="" class="md-social__link">
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><!--! Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License) Copyright 2024 Fonticons, Inc.--><path d="M498.1 5.6c10.1 7 15.4 19.1 13.5 31.2l-64 416c-1.5 9.7-7.4 18.2-16 23s-18.9 5.4-28 1.6L284 427.7l-68.5 74.1c-8.9 9.7-22.9 12.9-35.2 8.1S160 493.2 160 480v-83.6c0-4 1.5-7.8 4.2-10.7l167.6-182.9c5.8-6.3 5.6-16-.4-22s-15.7-6.4-22-.7L106 360.8l-88.3-44.2C7.1 311.3.3 300.7 0 288.9s5.9-22.8 16.1-28.7l448-256c10.7-6.1 23.9-5.5 34 1.4z"/></svg>
- </a>
-
-
-
-
-
-
-
-
- <a href="https://github.com/mindspore-lab/mindocr" target="_blank" rel="noopener" title="github.com" class="md-social__link">
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 496 512"><!--! Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License) Copyright 2024 Fonticons, Inc.--><path d="M165.9 397.4c0 2-2.3 3.6-5.2 3.6-3.3.3-5.6-1.3-5.6-3.6 0-2 2.3-3.6 5.2-3.6 3-.3 5.6 1.3 5.6 3.6zm-31.1-4.5c-.7 2 1.3 4.3 4.3 4.9 2.6 1 5.6 0 6.2-2s-1.3-4.3-4.3-5.2c-2.6-.7-5.5.3-6.2 2.3zm44.2-1.7c-2.9.7-4.9 2.6-4.6 4.9.3 2 2.9 3.3 5.9 2.6 2.9-.7 4.9-2.6 4.6-4.6-.3-1.9-3-3.2-5.9-2.9zM244.8 8C106.1 8 0 113.3 0 252c0 110.9 69.8 205.8 169.5 239.2 12.8 2.3 17.3-5.6 17.3-12.1 0-6.2-.3-40.4-.3-61.4 0 0-70 15-84.7-29.8 0 0-11.4-29.1-27.8-36.6 0 0-22.9-15.7 1.6-15.4 0 0 24.9 2 38.6 25.8 21.9 38.6 58.6 27.5 72.9 20.9 2.3-16 8.8-27.1 16-33.7-55.9-6.2-112.3-14.3-112.3-110.5 0-27.5 7.6-41.3 23.6-58.9-2.6-6.5-11.1-33.3 2.6-67.9 20.9-6.5 69 27 69 27 20-5.6 41.5-8.5 62.8-8.5s42.8 2.9 62.8 8.5c0 0 48.1-33.6 69-27 13.7 34.7 5.2 61.4 2.6 67.9 16 17.7 25.8 31.5 25.8 58.9 0 96.5-58.9 104.2-114.8 110.5 9.2 7.9 17 22.9 17 46.4 0 33.7-.3 75.4-.3 83.6 0 6.5 4.6 14.4 17.3 12.1C428.2 457.8 496 362.9 496 252 496 113.3 383.5 8 244.8 8zM97.2 352.9c-1.3 1-1 3.3.7 5.2 1.6 1.6 3.9 2.3 5.2 1 1.3-1 1-3.3-.7-5.2-1.6-1.6-3.9-2.3-5.2-1zm-10.8-8.1c-.7 1.3.3 2.9 2.3 3.9 1.6 1 3.6.7 4.3-.7.7-1.3-.3-2.9-2.3-3.9-2-.6-3.6-.3-4.3.7zm32.4 35.6c-1.6 1.3-1 4.3 1.3 6.2 2.3 2.3 5.2 2.6 6.5 1 1.3-1.3.7-4.3-1.3-6.2-2.2-2.3-5.2-2.6-6.5-1zm-11.4-14.7c-1.6 1-1.6 3.6 0 5.9 1.6 2.3 4.3 3.3 5.6 2.3 1.6-1.3 1.6-3.9 0-6.2-1.4-2.3-4-3.3-5.6-2z"/></svg>
- </a>
-
-
-
-
-
-
-
-
- <a href="https://www.zhihu.com/people/mindsporelab" target="_blank" rel="noopener" title="www.zhihu.com" class="md-social__link">
- <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 640 512"><!--! Font Awesome Free 6.5.2 by @fontawesome - https://fontawesome.com License - https://fontawesome.com/license/free (Icons: CC BY 4.0, Fonts: SIL OFL 1.1, Code: MIT License) Copyright 2024 Fonticons, Inc.--><path d="M170.54 148.13v217.54l23.43.01 7.71 26.37 42.01-26.37h49.53V148.13H170.54zm97.75 193.93h-27.94l-27.9 17.51-5.08-17.47-11.9-.04V171.75h72.82v170.31zm-118.46-94.39H97.5c1.74-27.1 2.2-51.59 2.2-73.46h51.16s1.97-22.56-8.58-22.31h-88.5c3.49-13.12 7.87-26.66 13.12-40.67 0 0-24.07 0-32.27 21.57-3.39 8.9-13.21 43.14-30.7 78.12 5.89-.64 25.37-1.18 36.84-22.21 2.11-5.89 2.51-6.66 5.14-14.53h28.87c0 10.5-1.2 66.88-1.68 73.44H20.83c-11.74 0-15.56 23.62-15.56 23.62h65.58C66.45 321.1 42.83 363.12 0 396.34c20.49 5.85 40.91-.93 51-9.9 0 0 22.98-20.9 35.59-69.25l53.96 64.94s7.91-26.89-1.24-39.99c-7.58-8.92-28.06-33.06-36.79-41.81L87.9 311.95c4.36-13.98 6.99-27.55 7.87-40.67h61.65s-.09-23.62-7.59-23.62v.01zm412.02-1.6c20.83-25.64 44.98-58.57 44.98-58.57s-18.65-14.8-27.38-4.06c-6 8.15-36.83 48.2-36.83 48.2l19.23 14.43zm-150.09-59.09c-9.01-8.25-25.91 2.13-25.91 2.13s39.52 55.04 41.12 57.45l19.46-13.73s-25.67-37.61-34.66-45.86h-.01zM640 258.35c-19.78 0-130.91.93-131.06.93v-101c4.81 0 12.42-.4 22.85-1.2 40.88-2.41 70.13-4 87.77-4.81 0 0 12.22-27.19-.59-33.44-3.07-1.18-23.17 4.58-23.17 4.58s-165.22 16.49-232.36 18.05c1.6 8.82 7.62 17.08 15.78 19.55 13.31 3.48 22.69 1.7 49.15.89 24.83-1.6 43.68-2.43 56.51-2.43v99.81H351.41s2.82 22.31 25.51 22.85h107.94v70.92c0 13.97-11.19 21.99-24.48 21.12-14.08.11-26.08-1.15-41.69-1.81 1.99 3.97 6.33 14.39 19.31 21.84 9.88 4.81 16.17 6.57 26.02 6.57 29.56 0 45.67-17.28 44.89-45.31v-73.32h122.36c9.68 0 8.7-23.78 8.7-23.78l.03-.01z"/></svg>
- </a>
-
- </div>
-
- </div>
- </div>
- </footer>
-
- </div>
- <div class="md-dialog" data-md-component="dialog">
- <div class="md-dialog__inner md-typeset"></div>
- </div>
-
-
- <script id="__config" type="application/json">{"base": "..", "features": ["navigation.tracking", "navigation.tabs", "navigation.sections", "navigation.indexes", "navigation.top", "navigation.footer", "toc.follow", "search.highlight", "search.share", "search.suggest", "content.action.view", "content.action.edit", "content.tabs.link", "content.code.copy", "content.code.select", "content.code.annotations"], "search": "../assets/javascripts/workers/search.b8dbb3d2.min.js", "translations": {"clipboard.copied": "Copied to clipboard", "clipboard.copy": "Copy to clipboard", "search.result.more.one": "1 more on this page", "search.result.more.other": "# more on this page", "search.result.none": "No matching documents", "search.result.one": "1 matching document", "search.result.other": "# matching documents", "search.result.placeholder": "Type to start searching", "search.result.term.missing": "Missing", "select.version": "Select version"}}</script>
-
-
- <script src="../assets/javascripts/bundle.3220b9d7.min.js"></script>
-
-
- </body>
- </html>
|