| <!DOCTYPE html> |
| |
| <html lang="en"> |
| <head> |
| <meta charset="utf-8"/> |
| <meta content="IE=edge" http-equiv="X-UA-Compatible"/> |
| <meta content="width=device-width, initial-scale=1" name="viewport"/> |
| <title>Image API — mxnet documentation</title> |
| <link crossorigin="anonymous" href="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/css/bootstrap.min.css" integrity="sha384-1q8mTJOASx8j1Au+a5WDVnPi2lkFfwwEAa8hDDdjZlpLegxhjVME1fgjWPGmkzs7" rel="stylesheet"/> |
| <link href="https://maxcdn.bootstrapcdn.com/font-awesome/4.5.0/css/font-awesome.min.css" rel="stylesheet"/> |
| <link href="../../_static/basic.css" rel="stylesheet" type="text/css"/> |
| <link href="../../_static/pygments.css" rel="stylesheet" type="text/css"/> |
| <link href="../../_static/mxnet.css" rel="stylesheet" type="text/css"> |
| <script type="text/javascript"> |
| var DOCUMENTATION_OPTIONS = { |
| URL_ROOT: '../../', |
| VERSION: '', |
| COLLAPSE_INDEX: false, |
| FILE_SUFFIX: '.html', |
| HAS_SOURCE: true, |
| SOURCELINK_SUFFIX: '.txt' |
| }; |
| </script> |
| <script src="../../_static/jquery-1.11.1.js" type="text/javascript"></script> |
| <script src="../../_static/underscore.js" type="text/javascript"></script> |
| <script src="../../_static/searchtools_custom.js" type="text/javascript"></script> |
| <script src="../../_static/doctools.js" type="text/javascript"></script> |
| <script src="../../_static/selectlang.js" type="text/javascript"></script> |
| <script src="https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML" type="text/javascript"></script> |
| <script type="text/javascript"> jQuery(function() { Search.loadIndex("/searchindex.js"); Search.init();}); </script> |
| <script> |
| (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ |
| (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new |
| Date();a=s.createElement(o), |
| m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) |
| })(window,document,'script','https://www.google-analytics.com/analytics.js','ga'); |
| |
| ga('create', 'UA-96378503-1', 'auto'); |
| ga('send', 'pageview'); |
| |
| </script> |
| <!-- --> |
| <!-- <script type="text/javascript" src="../../_static/jquery.js"></script> --> |
| <!-- --> |
| <!-- <script type="text/javascript" src="../../_static/underscore.js"></script> --> |
| <!-- --> |
| <!-- <script type="text/javascript" src="../../_static/doctools.js"></script> --> |
| <!-- --> |
| <!-- <script type="text/javascript" src="https://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML"></script> --> |
| <!-- --> |
| <link href="../../genindex.html" rel="index" title="Index"/> |
| <link href="../../search.html" rel="search" title="Search"> |
| <link href="index.html" rel="up" title="MXNet - Python API"> |
| <link href="optimization.html" rel="next" title="Optimization: initialize and update weights"/> |
| <link href="io.html" rel="prev" title="Data Loading API"/> |
| <link href="https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-icon.png" rel="icon" type="image/png"/> |
| </link></link></link></head> |
| <body role="document"><!-- Previous Navbar Layout |
| <div class="navbar navbar-default navbar-fixed-top"> |
| <div class="container"> |
| <div class="navbar-header"> |
| <button type="button" class="navbar-toggle collapsed" data-toggle="collapse" data-target="#navbar" aria-expanded="false" aria-controls="navbar"> |
| <span class="sr-only">Toggle navigation</span> |
| <span class="icon-bar"></span> |
| <span class="icon-bar"></span> |
| <span class="icon-bar"></span> |
| </button> |
| <a href="../../" class="navbar-brand"> |
| <img src="http://data.mxnet.io/theme/mxnet.png"> |
| </a> |
| </div> |
| <div id="navbar" class="navbar-collapse collapse"> |
| <ul id="navbar" class="navbar navbar-left"> |
| |
| <li> <a href="../../get_started/index.html">Get Started</a> </li> |
| |
| <li> <a href="../../tutorials/index.html">Tutorials</a> </li> |
| |
| <li> <a href="../../how_to/index.html">How To</a> </li> |
| |
| |
| <li class="dropdown"> |
| <a href="#" class="dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Packages <span class="caret"></span></a> |
| <ul class="dropdown-menu"> |
| |
| <li><a href="../../packages/python/index.html"> |
| Python |
| </a></li> |
| |
| <li><a href="../../packages/r/index.html"> |
| R |
| </a></li> |
| |
| <li><a href="../../packages/julia/index.html"> |
| Julia |
| </a></li> |
| |
| <li><a href="../../packages/c++/index.html"> |
| C++ |
| </a></li> |
| |
| <li><a href="../../packages/scala/index.html"> |
| Scala |
| </a></li> |
| |
| <li><a href="../../packages/perl/index.html"> |
| Perl |
| </a></li> |
| |
| </ul> |
| </li> |
| |
| <li> <a href="../../system/index.html">System</a> </li> |
| <li> |
| <form class="" role="search" action="../../search.html" method="get" autocomplete="off"> |
| <div class="form-group inner-addon left-addon"> |
| <i class="glyphicon glyphicon-search"></i> |
| <input type="text" name="q" class="form-control" placeholder="Search"> |
| </div> |
| <input type="hidden" name="check_keywords" value="yes" /> |
| <input type="hidden" name="area" value="default" /> |
| |
| </form> </li> |
| </ul> |
| <ul id="navbar" class="navbar navbar-right"> |
| <li> <a href="../../index.html"><span class="flag-icon flag-icon-us"></span></a> </li> |
| <li> <a href="../..//zh/index.html"><span class="flag-icon flag-icon-cn"></span></a> </li> |
| </ul> |
| </div> |
| </div> |
| </div> |
| Previous Navbar Layout End --> |
| <div class="navbar navbar-fixed-top"> |
| <div class="container" id="navContainer"> |
| <div class="innder" id="header-inner"> |
| <h1 id="logo-wrap"> |
| <a href="../../" id="logo"><img src="http://data.mxnet.io/theme/mxnet.png"/></a> |
| </h1> |
| <nav class="nav-bar" id="main-nav"> |
| <a class="main-nav-link" href="../../get_started/install.html">Install</a> |
| <a class="main-nav-link" href="../../tutorials/index.html">Tutorials</a> |
| <a class="main-nav-link" href="../../how_to/index.html">How To</a> |
| <span id="dropdown-menu-position-anchor"> |
| <a aria-expanded="true" aria-haspopup="true" class="main-nav-link dropdown-toggle" data-toggle="dropdown" href="#" role="button">API <span class="caret"></span></a> |
| <ul class="dropdown-menu" id="package-dropdown-menu"> |
| <li><a class="main-nav-link" href="../../api/python/index.html">Python</a></li> |
| <li><a class="main-nav-link" href="../../api/scala/index.html">Scala</a></li> |
| <li><a class="main-nav-link" href="../../api/r/index.html">R</a></li> |
| <li><a class="main-nav-link" href="../../api/julia/index.html">Julia</a></li> |
| <li><a class="main-nav-link" href="../../api/c++/index.html">C++</a></li> |
| <li><a class="main-nav-link" href="../../api/perl/index.html">Perl</a></li> |
| </ul> |
| </span> |
| <a class="main-nav-link" href="../../architecture/index.html">Architecture</a> |
| <!-- <a class="main-nav-link" href="../../community/index.html">Community</a> --> |
| <a class="main-nav-link" href="https://github.com/dmlc/mxnet">Github</a> |
| <span id="dropdown-menu-position-anchor-version" style="position: relative"><a href="#" class="main-nav-link dropdown-toggle" data-toggle="dropdown" role="button" aria-haspopup="true" aria-expanded="true">Versions(master)<span class="caret"></span></a><ul id="package-dropdown-menu" class="dropdown-menu"><li><a class="main-nav-link" href=http://mxnet.incubator.apache.org/test/>v0.10.14</a></li><li><a class="main-nav-link" href=http://mxnet.incubator.apache.org/test/versions/0.10/index.html>0.10</a></li><li><a class="main-nav-link" href=http://mxnet.incubator.apache.org/test/versions/master/index.html>master</a></li></ul></span></nav> |
| <script> function getRootPath(){ return "../../" } </script> |
| <div class="burgerIcon dropdown"> |
| <a class="dropdown-toggle" data-toggle="dropdown" href="#" role="button">☰</a> |
| <ul class="dropdown-menu dropdown-menu-right" id="burgerMenu"> |
| <li><a href="../../get_started/install.html">Install</a></li> |
| <li><a href="../../tutorials/index.html">Tutorials</a></li> |
| <li><a href="../../how_to/index.html">How To</a></li> |
| <li class="dropdown-submenu"> |
| <a href="#" tabindex="-1">API</a> |
| <ul class="dropdown-menu"> |
| <li><a href="../../api/python/index.html" tabindex="-1">Python</a> |
| </li> |
| <li><a href="../../api/scala/index.html" tabindex="-1">Scala</a> |
| </li> |
| <li><a href="../../api/r/index.html" tabindex="-1">R</a> |
| </li> |
| <li><a href="../../api/julia/index.html" tabindex="-1">Julia</a> |
| </li> |
| <li><a href="../../api/c++/index.html" tabindex="-1">C++</a> |
| </li> |
| <li><a href="../../api/perl/index.html" tabindex="-1">Perl</a> |
| </li> |
| </ul> |
| </li> |
| <li><a href="../../architecture/index.html">Architecture</a></li> |
| <li><a class="main-nav-link" href="https://github.com/dmlc/mxnet">Github</a></li> |
| <li id="dropdown-menu-position-anchor-version-mobile" class="dropdown-submenu" style="position: relative"><a href="#" tabindex="-1">Versions(master)</a><ul class="dropdown-menu"><li><a tabindex="-1" href=http://mxnet.incubator.apache.org/test/>v0.10.14</a></li><li><a tabindex="-1" href=http://mxnet.incubator.apache.org/test/versions/0.10/index.html>0.10</a></li><li><a tabindex="-1" href=http://mxnet.incubator.apache.org/test/versions/master/index.html>master</a></li></ul></li></ul> |
| </div> |
| <div class="plusIcon dropdown"> |
| <a class="dropdown-toggle" data-toggle="dropdown" href="#" role="button"><span aria-hidden="true" class="glyphicon glyphicon-plus"></span></a> |
| <ul class="dropdown-menu dropdown-menu-right" id="plusMenu"></ul> |
| </div> |
| <div id="search-input-wrap"> |
| <form action="../../search.html" autocomplete="off" class="" method="get" role="search"> |
| <div class="form-group inner-addon left-addon"> |
| <i class="glyphicon glyphicon-search"></i> |
| <input class="form-control" name="q" placeholder="Search" type="text"/> |
| </div> |
| <input name="check_keywords" type="hidden" value="yes"/> |
| <input name="area" type="hidden" value="default"/> |
| </form> |
| <div id="search-preview"></div> |
| </div> |
| <div id="searchIcon"> |
| <span aria-hidden="true" class="glyphicon glyphicon-search"></span> |
| </div> |
| <!-- <div id="lang-select-wrap"> --> |
| <!-- <label id="lang-select-label"> --> |
| <!-- <\!-- <i class="fa fa-globe"></i> -\-> --> |
| <!-- <span></span> --> |
| <!-- </label> --> |
| <!-- <select id="lang-select"> --> |
| <!-- <option value="en">Eng</option> --> |
| <!-- <option value="zh">中文</option> --> |
| <!-- </select> --> |
| <!-- </div> --> |
| <!-- <a id="mobile-nav-toggle"> |
| <span class="mobile-nav-toggle-bar"></span> |
| <span class="mobile-nav-toggle-bar"></span> |
| <span class="mobile-nav-toggle-bar"></span> |
| </a> --> |
| </div> |
| </div> |
| </div> |
| <div class="container"> |
| <div class="row"> |
| <div aria-label="main navigation" class="sphinxsidebar leftsidebar" role="navigation"> |
| <div class="sphinxsidebarwrapper"> |
| <ul class="current"> |
| <li class="toctree-l1 current"><a class="reference internal" href="index.html">Python Documents</a><ul class="current"> |
| <li class="toctree-l2 current"><a class="reference internal" href="index.html#table-of-contents">Table of contents</a><ul class="current"> |
| <li class="toctree-l3"><a class="reference internal" href="ndarray.html">NDArray API</a></li> |
| <li class="toctree-l3"><a class="reference internal" href="symbol.html">Symbol API</a></li> |
| <li class="toctree-l3"><a class="reference internal" href="module.html">Module API</a></li> |
| <li class="toctree-l3"><a class="reference internal" href="gluon.html">Gluon Package</a></li> |
| <li class="toctree-l3"><a class="reference internal" href="rnn.html">RNN Cell API</a></li> |
| <li class="toctree-l3"><a class="reference internal" href="kvstore.html">KVStore API</a></li> |
| <li class="toctree-l3"><a class="reference internal" href="io.html">Data Loading API</a></li> |
| <li class="toctree-l3 current"><a class="current reference internal" href="#">Image API</a><ul> |
| <li class="toctree-l4"><a class="reference internal" href="#overview">Overview</a></li> |
| <li class="toctree-l4"><a class="reference internal" href="#image-processing-functions">Image processing functions</a></li> |
| <li class="toctree-l4"><a class="reference internal" href="#image-iterators">Image iterators</a></li> |
| <li class="toctree-l4"><a class="reference internal" href="#api-reference">API Reference</a></li> |
| </ul> |
| </li> |
| <li class="toctree-l3"><a class="reference internal" href="optimization.html">Optimization: initialize and update weights</a></li> |
| <li class="toctree-l3"><a class="reference internal" href="callback.html">Callback API</a></li> |
| <li class="toctree-l3"><a class="reference internal" href="metric.html">Evaluation Metric API</a></li> |
| </ul> |
| </li> |
| </ul> |
| </li> |
| <li class="toctree-l1"><a class="reference internal" href="../r/index.html">R Documents</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../julia/index.html">Julia Documents</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../c++/index.html">C++ Documents</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../scala/index.html">Scala Documents</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../perl/index.html">Perl Documents</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../how_to/index.html">HowTo Documents</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../architecture/index.html">System Documents</a></li> |
| <li class="toctree-l1"><a class="reference internal" href="../../tutorials/index.html">Tutorials</a></li> |
| </ul> |
| </div> |
| </div> |
| <div class="content"> |
| <div class="section" id="image-api"> |
| <span id="image-api"></span><h1>Image API<a class="headerlink" href="#image-api" title="Permalink to this headline">¶</a></h1> |
| <div class="section" id="overview"> |
| <span id="overview"></span><h2>Overview<a class="headerlink" href="#overview" title="Permalink to this headline">¶</a></h2> |
| <p>This document summarizes supporting functions and iterators to read and process |
| images provided in</p> |
| <table border="1" class="longtable docutils"> |
| <colgroup> |
| <col width="10%"/> |
| <col width="90%"/> |
| </colgroup> |
| <tbody valign="top"> |
| <tr class="row-odd"><td><a class="reference internal" href="#module-mxnet.image" title="mxnet.image"><code class="xref py py-obj docutils literal"><span class="pre">mxnet.image</span></code></a></td> |
| <td>Image Iterators and image augmentation functions</td> |
| </tr> |
| </tbody> |
| </table> |
| </div> |
| <div class="section" id="image-processing-functions"> |
| <span id="image-processing-functions"></span><h2>Image processing functions<a class="headerlink" href="#image-processing-functions" title="Permalink to this headline">¶</a></h2> |
| <table border="1" class="longtable docutils"> |
| <colgroup> |
| <col width="10%"/> |
| <col width="90%"/> |
| </colgroup> |
| <tbody valign="top"> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.imdecode" title="mxnet.image.imdecode"><code class="xref py py-obj docutils literal"><span class="pre">image.imdecode</span></code></a></td> |
| <td>Decode an image to an NDArray.</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.scale_down" title="mxnet.image.scale_down"><code class="xref py py-obj docutils literal"><span class="pre">image.scale_down</span></code></a></td> |
| <td>Scales down crop size if it’s larger than image size.</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.resize_short" title="mxnet.image.resize_short"><code class="xref py py-obj docutils literal"><span class="pre">image.resize_short</span></code></a></td> |
| <td>Resizes shorter edge to size.</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.fixed_crop" title="mxnet.image.fixed_crop"><code class="xref py py-obj docutils literal"><span class="pre">image.fixed_crop</span></code></a></td> |
| <td>Crop src at fixed location, and (optionally) resize it to size.</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.random_crop" title="mxnet.image.random_crop"><code class="xref py py-obj docutils literal"><span class="pre">image.random_crop</span></code></a></td> |
| <td>Randomly crop <cite>src</cite> with <cite>size</cite> (width, height).</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.center_crop" title="mxnet.image.center_crop"><code class="xref py py-obj docutils literal"><span class="pre">image.center_crop</span></code></a></td> |
| <td>Crops the image <cite>src</cite> to the given <cite>size</cite> by trimming on all four sides and preserving the center of the image.</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.color_normalize" title="mxnet.image.color_normalize"><code class="xref py py-obj docutils literal"><span class="pre">image.color_normalize</span></code></a></td> |
| <td>Normalize src with mean and std.</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.random_size_crop" title="mxnet.image.random_size_crop"><code class="xref py py-obj docutils literal"><span class="pre">image.random_size_crop</span></code></a></td> |
| <td>Randomly crop src with size.</td> |
| </tr> |
| </tbody> |
| </table> |
| </div> |
| <div class="section" id="image-iterators"> |
| <span id="image-iterators"></span><h2>Image iterators<a class="headerlink" href="#image-iterators" title="Permalink to this headline">¶</a></h2> |
| <p>Iterators support loading image from binary <code class="docutils literal"><span class="pre">Record</span> <span class="pre">IO</span></code> and raw image files.</p> |
| <table border="1" class="longtable docutils"> |
| <colgroup> |
| <col width="10%"/> |
| <col width="90%"/> |
| </colgroup> |
| <tbody valign="top"> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.ImageIter" title="mxnet.image.ImageIter"><code class="xref py py-obj docutils literal"><span class="pre">image.ImageIter</span></code></a></td> |
| <td>Image data iterator with a large number of augmentation choices.</td> |
| </tr> |
| </tbody> |
| </table> |
| <div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">data_iter</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">ImageIter</span><span class="p">(</span><span class="n">batch_size</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">data_shape</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span> <span class="n">label_width</span><span class="o">=</span><span class="mi">1</span><span class="p">,</span> |
| <span class="go"> path_imglist='data/custom.lst')</span> |
| <span class="gp">>>> </span><span class="n">data_iter</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span> |
| <span class="gp">>>> </span><span class="k">for</span> <span class="n">data</span> <span class="ow">in</span> <span class="n">data_iter</span><span class="p">:</span> |
| <span class="gp">... </span> <span class="n">d</span> <span class="o">=</span> <span class="n">data</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> |
| <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="n">d</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="c1"># we can apply lots of augmentations as well</span> |
| <span class="gp">>>> </span><span class="n">data_iter</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">ImageIter</span><span class="p">(</span><span class="mi">4</span><span class="p">,</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span> <span class="n">path_imglist</span><span class="o">=</span><span class="s1">'data/custom.lst'</span><span class="p">,</span> |
| <span class="go"> rand_crop=resize=True, rand_mirror=True, mean=True,</span> |
| <span class="go"> brightness=0.1, contrast=0.1, saturation=0.1, hue=0.1,</span> |
| <span class="go"> pca_noise=0.1, rand_gray=0.05)</span> |
| <span class="gp">>>> </span><span class="n">data</span> <span class="o">=</span> <span class="n">data_iter</span><span class="o">.</span><span class="n">next</span><span class="p">()</span> |
| <span class="gp">>>> </span><span class="c1"># specify augmenters manually is also supported</span> |
| <span class="gp">>>> </span><span class="n">data_iter</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">ImageIter</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span> <span class="n">path_rec</span><span class="o">=</span><span class="s1">'data/caltech.rec'</span><span class="p">,</span> |
| <span class="go"> path_imgidx='data/caltech.idx', shuffle=True,</span> |
| <span class="go"> aug_list=[mx.image.HorizontalFlipAug(0.5),</span> |
| <span class="go"> mx.image.ColorJitterAug(0.1, 0.1, 0.1)])</span> |
| </pre></div> |
| </div> |
| <p>We use helper function to initialize augmenters</p> |
| <table border="1" class="longtable docutils"> |
| <colgroup> |
| <col width="10%"/> |
| <col width="90%"/> |
| </colgroup> |
| <tbody valign="top"> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.CreateAugmenter" title="mxnet.image.CreateAugmenter"><code class="xref py py-obj docutils literal"><span class="pre">image.CreateAugmenter</span></code></a></td> |
| <td>Creates an augmenter list.</td> |
| </tr> |
| </tbody> |
| </table> |
| <p>A list of supporting augmenters</p> |
| <table border="1" class="longtable docutils"> |
| <colgroup> |
| <col width="10%"/> |
| <col width="90%"/> |
| </colgroup> |
| <tbody valign="top"> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.Augmenter" title="mxnet.image.Augmenter"><code class="xref py py-obj docutils literal"><span class="pre">image.Augmenter</span></code></a></td> |
| <td>Image Augmenter base class</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.ResizeAug" title="mxnet.image.ResizeAug"><code class="xref py py-obj docutils literal"><span class="pre">image.ResizeAug</span></code></a></td> |
| <td>Make resize shorter edge to size augmenter.</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.ForceResizeAug" title="mxnet.image.ForceResizeAug"><code class="xref py py-obj docutils literal"><span class="pre">image.ForceResizeAug</span></code></a></td> |
| <td>Force resize to size regardless of aspect ratio</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.RandomCropAug" title="mxnet.image.RandomCropAug"><code class="xref py py-obj docutils literal"><span class="pre">image.RandomCropAug</span></code></a></td> |
| <td>Make random crop augmenter</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.RandomSizedCropAug" title="mxnet.image.RandomSizedCropAug"><code class="xref py py-obj docutils literal"><span class="pre">image.RandomSizedCropAug</span></code></a></td> |
| <td>Make random crop with random resizing and random aspect ratio jitter augmenter.</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.CenterCropAug" title="mxnet.image.CenterCropAug"><code class="xref py py-obj docutils literal"><span class="pre">image.CenterCropAug</span></code></a></td> |
| <td>Make center crop augmenter.</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.RandomOrderAug" title="mxnet.image.RandomOrderAug"><code class="xref py py-obj docutils literal"><span class="pre">image.RandomOrderAug</span></code></a></td> |
| <td>Apply list of augmenters in random order</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.BrightnessJitterAug" title="mxnet.image.BrightnessJitterAug"><code class="xref py py-obj docutils literal"><span class="pre">image.BrightnessJitterAug</span></code></a></td> |
| <td>Random brightness jitter augmentation.</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.ContrastJitterAug" title="mxnet.image.ContrastJitterAug"><code class="xref py py-obj docutils literal"><span class="pre">image.ContrastJitterAug</span></code></a></td> |
| <td>Random contrast jitter augmentation.</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.SaturationJitterAug" title="mxnet.image.SaturationJitterAug"><code class="xref py py-obj docutils literal"><span class="pre">image.SaturationJitterAug</span></code></a></td> |
| <td>Random saturation jitter augmentation.</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.HueJitterAug" title="mxnet.image.HueJitterAug"><code class="xref py py-obj docutils literal"><span class="pre">image.HueJitterAug</span></code></a></td> |
| <td>Random hue jitter augmentation.</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.ColorJitterAug" title="mxnet.image.ColorJitterAug"><code class="xref py py-obj docutils literal"><span class="pre">image.ColorJitterAug</span></code></a></td> |
| <td>Apply random brightness, contrast and saturation jitter in random order.</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.LightingAug" title="mxnet.image.LightingAug"><code class="xref py py-obj docutils literal"><span class="pre">image.LightingAug</span></code></a></td> |
| <td>Add PCA based noise.</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.ColorNormalizeAug" title="mxnet.image.ColorNormalizeAug"><code class="xref py py-obj docutils literal"><span class="pre">image.ColorNormalizeAug</span></code></a></td> |
| <td>Mean and std normalization.</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.RandomGrayAug" title="mxnet.image.RandomGrayAug"><code class="xref py py-obj docutils literal"><span class="pre">image.RandomGrayAug</span></code></a></td> |
| <td>Randomly convert to gray image.</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.HorizontalFlipAug" title="mxnet.image.HorizontalFlipAug"><code class="xref py py-obj docutils literal"><span class="pre">image.HorizontalFlipAug</span></code></a></td> |
| <td>Random horizontal flip.</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.CastAug" title="mxnet.image.CastAug"><code class="xref py py-obj docutils literal"><span class="pre">image.CastAug</span></code></a></td> |
| <td>Cast to float32</td> |
| </tr> |
| </tbody> |
| </table> |
| <p>Similar to <code class="docutils literal"><span class="pre">ImageIter</span></code>, <code class="docutils literal"><span class="pre">ImageDetIter</span></code> is designed for <code class="docutils literal"><span class="pre">Object</span> <span class="pre">Detection</span></code> tasks.</p> |
| <table border="1" class="longtable docutils"> |
| <colgroup> |
| <col width="10%"/> |
| <col width="90%"/> |
| </colgroup> |
| <tbody valign="top"> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.ImageDetIter" title="mxnet.image.ImageDetIter"><code class="xref py py-obj docutils literal"><span class="pre">image.ImageDetIter</span></code></a></td> |
| <td>Image iterator with a large number of augmentation choices for detection.</td> |
| </tr> |
| </tbody> |
| </table> |
| <div class="highlight-python"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">data_iter</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">ImageDetIter</span><span class="p">(</span><span class="n">batch_size</span><span class="o">=</span><span class="mi">4</span><span class="p">,</span> <span class="n">data_shape</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">224</span><span class="p">,</span> <span class="mi">224</span><span class="p">),</span> |
| <span class="go"> path_imglist='data/train.lst')</span> |
| <span class="gp">>>> </span><span class="n">data_iter</span><span class="o">.</span><span class="n">reset</span><span class="p">()</span> |
| <span class="gp">>>> </span><span class="k">for</span> <span class="n">data</span> <span class="ow">in</span> <span class="n">data_iter</span><span class="p">:</span> |
| <span class="gp">... </span> <span class="n">d</span> <span class="o">=</span> <span class="n">data</span><span class="o">.</span><span class="n">data</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> |
| <span class="gp">... </span> <span class="n">l</span> <span class="o">=</span> <span class="n">data</span><span class="o">.</span><span class="n">label</span><span class="p">[</span><span class="mi">0</span><span class="p">]</span> |
| <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="n">d</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> |
| <span class="gp">... </span> <span class="k">print</span><span class="p">(</span><span class="n">l</span><span class="o">.</span><span class="n">shape</span><span class="p">)</span> |
| </pre></div> |
| </div> |
| <p>Unlike object classification with fixed label_width, object count may vary from |
| image to image. Thus we have special format for object detection labels. |
| Usually the <code class="docutils literal"><span class="pre">lst</span></code> file generated by <code class="docutils literal"><span class="pre">tools/im2rec.py</span></code> is a list of</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">index_0</span> <span class="n">label_0</span> <span class="n">image_path_0</span> |
| <span class="n">index_1</span> <span class="n">label_1</span> <span class="n">image_path_1</span> |
| </pre></div> |
| </div> |
| <p>Where <code class="docutils literal"><span class="pre">label_N</span></code> is a number a of fixed-width vector. |
| The format of label used in object detection is a variable length vector</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="n">A</span> <span class="n">B</span> <span class="p">[</span><span class="n">header</span><span class="p">]</span> <span class="p">[(</span><span class="n">object0</span><span class="p">),</span> <span class="p">(</span><span class="n">object1</span><span class="p">),</span> <span class="o">...</span> <span class="p">(</span><span class="n">objectN</span><span class="p">)]</span> |
| </pre></div> |
| </div> |
| <p>Where A is the width of header, B is the width of each object. |
| Header is optional and used for inserting helper information such as (width, height). |
| Each object is usually 5 or 6 numbers describing the object properties, for example: |
| [id, xmin, ymin, xmax, ymax, difficulty] |
| Putting all together, we have a <code class="docutils literal"><span class="pre">lst</span></code> file for object detection:</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="mi">0</span> <span class="mi">2</span> <span class="mi">5</span> <span class="mi">640</span> <span class="mi">480</span> <span class="mi">1</span> <span class="mf">0.1</span> <span class="mf">0.2</span> <span class="mf">0.8</span> <span class="mf">0.9</span> <span class="mi">2</span> <span class="mf">0.5</span> <span class="mf">0.3</span> <span class="mf">0.6</span> <span class="mf">0.8</span> <span class="n">data</span><span class="o">/</span><span class="n">xxx</span><span class="o">.</span><span class="n">jpg</span> |
| <span class="mi">1</span> <span class="mi">2</span> <span class="mi">5</span> <span class="mi">480</span> <span class="mi">640</span> <span class="mi">3</span> <span class="mf">0.05</span> <span class="mf">0.16</span> <span class="mf">0.75</span> <span class="mf">0.9</span> <span class="n">data</span><span class="o">/</span><span class="n">xxx</span><span class="o">.</span><span class="n">jpg</span> |
| <span class="mi">2</span> <span class="mi">2</span> <span class="mi">5</span> <span class="mi">500</span> <span class="mi">600</span> <span class="mi">2</span> <span class="mf">0.6</span> <span class="mf">0.1</span> <span class="mf">0.7</span> <span class="mf">0.5</span> <span class="mi">0</span> <span class="mf">0.1</span> <span class="mf">0.3</span> <span class="mf">0.2</span> <span class="mf">0.4</span> <span class="mi">3</span> <span class="mf">0.25</span> <span class="mf">0.25</span> <span class="mf">0.3</span> <span class="mf">0.3</span> <span class="n">data</span><span class="o">/</span><span class="n">xxx</span><span class="o">.</span><span class="n">jpg</span> |
| <span class="o">...</span> |
| </pre></div> |
| </div> |
| <p>A helper function to initialize Augmenters for <code class="docutils literal"><span class="pre">Object</span> <span class="pre">detection</span></code> task</p> |
| <table border="1" class="longtable docutils"> |
| <colgroup> |
| <col width="10%"/> |
| <col width="90%"/> |
| </colgroup> |
| <tbody valign="top"> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.CreateDetAugmenter" title="mxnet.image.CreateDetAugmenter"><code class="xref py py-obj docutils literal"><span class="pre">image.CreateDetAugmenter</span></code></a></td> |
| <td>Create augmenters for detection.</td> |
| </tr> |
| </tbody> |
| </table> |
| <p>Since <code class="docutils literal"><span class="pre">Detection</span></code> task is sensitive to object localization, any modification |
| to image that introduced localization shift will require correction to label, |
| and a list of augmenters specific for <code class="docutils literal"><span class="pre">Object</span> <span class="pre">detection</span></code> is provided</p> |
| <table border="1" class="longtable docutils"> |
| <colgroup> |
| <col width="10%"/> |
| <col width="90%"/> |
| </colgroup> |
| <tbody valign="top"> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.DetBorrowAug" title="mxnet.image.DetBorrowAug"><code class="xref py py-obj docutils literal"><span class="pre">image.DetBorrowAug</span></code></a></td> |
| <td>Borrow standard augmenter from image classification.</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.DetRandomSelectAug" title="mxnet.image.DetRandomSelectAug"><code class="xref py py-obj docutils literal"><span class="pre">image.DetRandomSelectAug</span></code></a></td> |
| <td>Randomly select one augmenter to apply, with chance to skip all.</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.DetHorizontalFlipAug" title="mxnet.image.DetHorizontalFlipAug"><code class="xref py py-obj docutils literal"><span class="pre">image.DetHorizontalFlipAug</span></code></a></td> |
| <td>Random horizontal flipping.</td> |
| </tr> |
| <tr class="row-even"><td><a class="reference internal" href="#mxnet.image.DetRandomCropAug" title="mxnet.image.DetRandomCropAug"><code class="xref py py-obj docutils literal"><span class="pre">image.DetRandomCropAug</span></code></a></td> |
| <td>Random cropping with constraints</td> |
| </tr> |
| <tr class="row-odd"><td><a class="reference internal" href="#mxnet.image.DetRandomPadAug" title="mxnet.image.DetRandomPadAug"><code class="xref py py-obj docutils literal"><span class="pre">image.DetRandomPadAug</span></code></a></td> |
| <td>Random padding augmenter.</td> |
| </tr> |
| </tbody> |
| </table> |
| </div> |
| <div class="section" id="api-reference"> |
| <span id="api-reference"></span><h2>API Reference<a class="headerlink" href="#api-reference" title="Permalink to this headline">¶</a></h2> |
| <script src="../../_static/js/auto_module_index.js" type="text/javascript"></script><span class="target" id="module-mxnet.image"></span><p>Image Iterators and image augmentation functions</p> |
| <dl class="class"> |
| <dt id="mxnet.image.ImageIter"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">ImageIter</code><span class="sig-paren">(</span><em>batch_size</em>, <em>data_shape</em>, <em>label_width=1</em>, <em>path_imgrec=None</em>, <em>path_imglist=None</em>, <em>path_root=None</em>, <em>path_imgidx=None</em>, <em>shuffle=False</em>, <em>part_index=0</em>, <em>num_parts=1</em>, <em>aug_list=None</em>, <em>imglist=None</em>, <em>data_name='data'</em>, <em>label_name='softmax_label'</em>, <em>**kwargs</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageIter" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Image data iterator with a large number of augmentation choices. |
| This iterator supports reading from both .rec files and raw image files.</p> |
| <p>To load input images from .rec files, use <cite>path_imgrec</cite> parameter and to load from raw image |
| files, use <cite>path_imglist</cite> and <cite>path_root</cite> parameters.</p> |
| <p>To use data partition (for distributed training) or shuffling, specify <cite>path_imgidx</cite> parameter.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>batch_size</strong> (<em>int</em>) – Number of examples per batch.</li> |
| <li><strong>data_shape</strong> (<em>tuple</em>) – Data shape in (channels, height, width) format. |
| For now, only RGB image with 3 channels is supported.</li> |
| <li><strong>label_width</strong> (<em>int</em><em>, </em><em>optional</em>) – Number of labels per example. The default label width is 1.</li> |
| <li><strong>path_imgrec</strong> (<em>str</em>) – Path to image record file (.rec). |
| Created with tools/im2rec.py or bin/im2rec.</li> |
| <li><strong>path_imglist</strong> (<em>str</em>) – Path to image list (.lst). |
| Created with tools/im2rec.py or with custom script. |
| Format: Tab separated record of index, one or more labels and relative_path_from_root.</li> |
| <li><strong>imglist</strong> (<em>list</em>) – A list of images with the label(s). |
| Each item is a list [imagelabel: float or list of float, imgpath].</li> |
| <li><strong>path_root</strong> (<em>str</em>) – Root folder of image files.</li> |
| <li><strong>path_imgidx</strong> (<em>str</em>) – Path to image index file. Needed for partition and shuffling when using .rec source.</li> |
| <li><strong>shuffle</strong> (<em>bool</em>) – Whether to shuffle all images at the start of each iteration or not. |
| Can be slow for HDD.</li> |
| <li><strong>part_index</strong> (<em>int</em>) – Partition index.</li> |
| <li><strong>num_parts</strong> (<em>int</em>) – Total number of partitions.</li> |
| <li><strong>data_name</strong> (<em>str</em>) – Data name for provided symbols.</li> |
| <li><strong>label_name</strong> (<em>str</em>) – Label name for provided symbols.</li> |
| <li><strong>kwargs</strong> – More arguments for creating augmenter. See mx.image.CreateAugmenter.</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageIter.reset"> |
| <code class="descname">reset</code><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageIter.reset" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Resets the iterator to the beginning of the data.</p> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageIter.next_sample"> |
| <code class="descname">next_sample</code><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageIter.next_sample" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Helper function for reading in next sample.</p> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageIter.next"> |
| <code class="descname">next</code><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageIter.next" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Returns the next batch of data.</p> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageIter.check_data_shape"> |
| <code class="descname">check_data_shape</code><span class="sig-paren">(</span><em>data_shape</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageIter.check_data_shape" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Checks if the input data shape is valid</p> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageIter.check_valid_image"> |
| <code class="descname">check_valid_image</code><span class="sig-paren">(</span><em>data</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageIter.check_valid_image" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Checks if the input data is valid</p> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageIter.imdecode"> |
| <code class="descname">imdecode</code><span class="sig-paren">(</span><em>s</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageIter.imdecode" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Decodes a string or byte string to an NDArray. |
| See mx.img.imdecode for more details.</p> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageIter.read_image"> |
| <code class="descname">read_image</code><span class="sig-paren">(</span><em>fname</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageIter.read_image" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Reads an input image <cite>fname</cite> and returns the decoded raw bytes.</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">dataIter</span><span class="o">.</span><span class="n">read_image</span><span class="p">(</span><span class="s1">'Face.jpg'</span><span class="p">)</span> <span class="c1"># returns decoded raw bytes.</span> |
| </pre></div> |
| </div> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageIter.augmentation_transform"> |
| <code class="descname">augmentation_transform</code><span class="sig-paren">(</span><em>data</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageIter.augmentation_transform" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Transforms input data with specified augmentation.</p> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageIter.postprocess_data"> |
| <code class="descname">postprocess_data</code><span class="sig-paren">(</span><em>datum</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageIter.postprocess_data" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Final postprocessing step before image is loaded into the batch.</p> |
| </dd></dl> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.imdecode"> |
| <code class="descclassname">image.</code><code class="descname">imdecode</code><span class="sig-paren">(</span><em>buf</em>, <em>*args</em>, <em>**kwargs</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.imdecode" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Decode an image to an NDArray.</p> |
| <p>Note: <cite>imdecode</cite> uses OpenCV (not the CV2 Python library). |
| MXNet must have been built with OpenCV for <cite>imdecode</cite> to work.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple"> |
| <li><strong>buf</strong> (<em>str/bytes</em><em> or </em><em>numpy.ndarray</em>) – Binary image data as string or numpy ndarray.</li> |
| <li><strong>flag</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=1</em>) – 1 for three channel color output. 0 for grayscale output.</li> |
| <li><strong>to_rgb</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=1</em>) – 1 for RGB formatted output (MXNet default). 0 for BGR formatted output (OpenCV default).</li> |
| <li><strong>out</strong> (<a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray"><em>NDArray</em></a><em>, </em><em>optional</em>) – Output buffer. Use <cite>None</cite> for automatic allocation.</li> |
| </ul> |
| </td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">An <cite>NDArray</cite> containing the image.</p> |
| </td> |
| </tr> |
| <tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last"><a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray">NDArray</a></p> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| <p class="rubric">Example</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s2">"flower.jpg"</span><span class="p">,</span> <span class="s1">'rb'</span><span class="p">)</span> <span class="k">as</span> <span class="n">fp</span><span class="p">:</span> |
| <span class="gp">... </span> <span class="n">str_image</span> <span class="o">=</span> <span class="n">fp</span><span class="o">.</span><span class="n">read</span><span class="p">()</span> |
| <span class="gp">...</span> |
| <span class="gp">>>> </span><span class="n">image</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">img</span><span class="o">.</span><span class="n">imdecode</span><span class="p">(</span><span class="n">str_image</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">image</span> |
| <span class="go"><ndarray 224x224x3="" @cpu(0)=""></ndarray></span> |
| </pre></div> |
| </div> |
| <p>Set <cite>flag</cite> parameter to 0 to get grayscale output</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s2">"flower.jpg"</span><span class="p">,</span> <span class="s1">'rb'</span><span class="p">)</span> <span class="k">as</span> <span class="n">fp</span><span class="p">:</span> |
| <span class="gp">... </span> <span class="n">str_image</span> <span class="o">=</span> <span class="n">fp</span><span class="o">.</span><span class="n">read</span><span class="p">()</span> |
| <span class="gp">...</span> |
| <span class="gp">>>> </span><span class="n">image</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">img</span><span class="o">.</span><span class="n">imdecode</span><span class="p">(</span><span class="n">str_image</span><span class="p">,</span> <span class="n">flag</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">image</span> |
| <span class="go"><ndarray 224x224x1="" @cpu(0)=""></ndarray></span> |
| </pre></div> |
| </div> |
| <p>Set <cite>to_rgb</cite> parameter to 0 to get output in OpenCV format (BGR)</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s2">"flower.jpg"</span><span class="p">,</span> <span class="s1">'rb'</span><span class="p">)</span> <span class="k">as</span> <span class="n">fp</span><span class="p">:</span> |
| <span class="gp">... </span> <span class="n">str_image</span> <span class="o">=</span> <span class="n">fp</span><span class="o">.</span><span class="n">read</span><span class="p">()</span> |
| <span class="gp">...</span> |
| <span class="gp">>>> </span><span class="n">image</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">img</span><span class="o">.</span><span class="n">imdecode</span><span class="p">(</span><span class="n">str_image</span><span class="p">,</span> <span class="n">to_rgb</span><span class="o">=</span><span class="mi">0</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">image</span> |
| <span class="go"><ndarray 224x224x3="" @cpu(0)=""></ndarray></span> |
| </pre></div> |
| </div> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.scale_down"> |
| <code class="descclassname">image.</code><code class="descname">scale_down</code><span class="sig-paren">(</span><em>src_size</em>, <em>size</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.scale_down" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Scales down crop size if it’s larger than image size.</p> |
| <p>If width/height of the crop is larger than the width/height of the image, |
| sets the width/height to the width/height of the image.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple"> |
| <li><strong>src_size</strong> (<em>tuple of int</em>) – Size of the image in (width, height) format.</li> |
| <li><strong>size</strong> (<em>tuple of int</em>) – Size of the crop in (width, height) format.</li> |
| </ul> |
| </td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">A tuple containing the scaled crop size in (width, height) format.</p> |
| </td> |
| </tr> |
| <tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">tuple of int</p> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| <p class="rubric">Example</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">src_size</span> <span class="o">=</span> <span class="p">(</span><span class="mi">640</span><span class="p">,</span><span class="mi">480</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">size</span> <span class="o">=</span> <span class="p">(</span><span class="mi">720</span><span class="p">,</span><span class="mi">120</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">new_size</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">img</span><span class="o">.</span><span class="n">scale_down</span><span class="p">(</span><span class="n">src_size</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">new_size</span> |
| <span class="go">(640,106)</span> |
| </pre></div> |
| </div> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.resize_short"> |
| <code class="descclassname">image.</code><code class="descname">resize_short</code><span class="sig-paren">(</span><em>src</em>, <em>size</em>, <em>interp=2</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.resize_short" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Resizes shorter edge to size.</p> |
| <p>Note: <cite>resize_short</cite> uses OpenCV (not the CV2 Python library). |
| MXNet must have been built with OpenCV for <cite>resize_short</cite> to work.</p> |
| <p>Resizes the original image by setting the shorter edge to size |
| and setting the longer edge accordingly. |
| Resizing function is called from OpenCV.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple"> |
| <li><strong>src</strong> (<a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – The original image.</li> |
| <li><strong>size</strong> (<em>int</em>) – The length to be set for the shorter edge.</li> |
| <li><strong>interp</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=2</em>) – Interpolation method used for resizing the image. |
| Possible values: |
| 0: Nearest Neighbors Interpolation. |
| 1: Bilinear interpolation. |
| 2: Area-based (resampling using pixel area relation). It may be a |
| preferred method for image decimation, as it gives moire-free |
| results. But when the image is zoomed, it is similar to the Nearest |
| Neighbors method. (used by default). |
| 3: Bicubic interpolation over 4x4 pixel neighborhood. |
| 4: Lanczos interpolation over 8x8 pixel neighborhood. |
| 9: Cubic for enlarge, area for shrink, bilinear for others |
| 10: Random select from interpolation method metioned above. |
| Note: |
| When shrinking an image, it will generally look best with AREA-based |
| interpolation, whereas, when enlarging an image, it will generally look best |
| with Bicubic (slow) or Bilinear (faster but still looks OK). |
| More details can be found in the documentation of OpenCV, please refer to |
| <a class="reference external" href="http://docs.opencv.org/master/da/d54/group__imgproc__transform.html">http://docs.opencv.org/master/da/d54/group__imgproc__transform.html</a>.</li> |
| </ul> |
| </td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">An ‘NDArray’ containing the resized image.</p> |
| </td> |
| </tr> |
| <tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last"><a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray">NDArray</a></p> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| <p class="rubric">Example</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s2">"flower.jpeg"</span><span class="p">,</span> <span class="s1">'rb'</span><span class="p">)</span> <span class="k">as</span> <span class="n">fp</span><span class="p">:</span> |
| <span class="gp">... </span> <span class="n">str_image</span> <span class="o">=</span> <span class="n">fp</span><span class="o">.</span><span class="n">read</span><span class="p">()</span> |
| <span class="gp">...</span> |
| <span class="gp">>>> </span><span class="n">image</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">img</span><span class="o">.</span><span class="n">imdecode</span><span class="p">(</span><span class="n">str_image</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">image</span> |
| <span class="go"><ndarray 2321x3482x3="" @cpu(0)=""></ndarray></span> |
| <span class="gp">>>> </span><span class="n">size</span> <span class="o">=</span> <span class="mi">640</span> |
| <span class="gp">>>> </span><span class="n">new_image</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">img</span><span class="o">.</span><span class="n">resize_short</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">size</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">new_image</span> |
| <span class="go"><ndarray 2321x3482x3="" @cpu(0)=""></ndarray></span> |
| </pre></div> |
| </div> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.fixed_crop"> |
| <code class="descclassname">image.</code><code class="descname">fixed_crop</code><span class="sig-paren">(</span><em>src</em>, <em>x0</em>, <em>y0</em>, <em>w</em>, <em>h</em>, <em>size=None</em>, <em>interp=2</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.fixed_crop" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Crop src at fixed location, and (optionally) resize it to size.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple"> |
| <li><strong>src</strong> (<a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – Input image</li> |
| <li><strong>x0</strong> (<em>int</em>) – Left boundary of the cropping area</li> |
| <li><strong>y0</strong> (<em>int</em>) – Top boundary of the cropping area</li> |
| <li><strong>w</strong> (<em>int</em>) – Width of the cropping area</li> |
| <li><strong>h</strong> (<em>int</em>) – Height of the cropping area</li> |
| <li><strong>size</strong> (<em>tuple of</em><em> (</em><em>w</em><em>, </em><em>h</em><em>)</em><em></em>) – Optional, resize to new size after cropping</li> |
| <li><strong>interp</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=2</em>) – Interpolation method. See resize_short for details.</li> |
| </ul> |
| </td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">An <cite>NDArray</cite> containing the cropped image.</p> |
| </td> |
| </tr> |
| <tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last"><a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray">NDArray</a></p> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.random_crop"> |
| <code class="descclassname">image.</code><code class="descname">random_crop</code><span class="sig-paren">(</span><em>src</em>, <em>size</em>, <em>interp=2</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.random_crop" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Randomly crop <cite>src</cite> with <cite>size</cite> (width, height). |
| Upsample result if <cite>src</cite> is smaller than <cite>size</cite>.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple"> |
| <li><strong>src</strong> (Source image <cite>NDArray</cite>) – </li> |
| <li><strong>size</strong> (Size of the crop formatted as (width, height). If the <cite>size</cite> is larger) – than the image, then the source image is upsampled to <cite>size</cite> and returned.</li> |
| <li><strong>interp</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=2</em>) – Interpolation method. See resize_short for details.</li> |
| </ul> |
| </td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last"><ul class="simple"> |
| <li><em>NDArray</em> – An <cite>NDArray</cite> containing the cropped image.</li> |
| <li><em>Tuple</em> – A tuple (x, y, width, height) where (x, y) is top-left position of the crop in the |
| original image and (width, height) are the dimensions of the cropped image.</li> |
| </ul> |
| </p> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| <p class="rubric">Example</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">im</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">nd</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="n">cv2</span><span class="o">.</span><span class="n">imread</span><span class="p">(</span><span class="s2">"flower.jpg"</span><span class="p">))</span> |
| <span class="gp">>>> </span><span class="n">cropped_im</span><span class="p">,</span> <span class="n">rect</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">random_crop</span><span class="p">(</span><span class="n">im</span><span class="p">,</span> <span class="p">(</span><span class="mi">100</span><span class="p">,</span> <span class="mi">100</span><span class="p">))</span> |
| <span class="gp">>>> </span><span class="nb">print</span> <span class="n">cropped_im</span> |
| <span class="go"><ndarray 100x100x1="" @cpu(0)=""></ndarray></span> |
| <span class="gp">>>> </span><span class="nb">print</span> <span class="n">rect</span> |
| <span class="go">(20, 21, 100, 100)</span> |
| </pre></div> |
| </div> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.center_crop"> |
| <code class="descclassname">image.</code><code class="descname">center_crop</code><span class="sig-paren">(</span><em>src</em>, <em>size</em>, <em>interp=2</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.center_crop" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Crops the image <cite>src</cite> to the given <cite>size</cite> by trimming on all four |
| sides and preserving the center of the image. Upsamples if <cite>src</cite> is smaller |
| than <cite>size</cite>.</p> |
| <div class="admonition note"> |
| <p class="first admonition-title">Note</p> |
| <p class="last">This requires MXNet to be compiled with USE_OPENCV.</p> |
| </div> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple"> |
| <li><strong>src</strong> (<a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – Binary source image data.</li> |
| <li><strong>size</strong> (<em>list</em><em> or </em><em>tuple of int</em>) – The desired output image size.</li> |
| <li><strong>interp</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=2</em>) – Interpolation method. See resize_short for details.</li> |
| </ul> |
| </td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last"><ul class="simple"> |
| <li><em>NDArray</em> – The cropped image.</li> |
| <li><em>Tuple</em> – (x, y, width, height) where x, y are the positions of the crop in the |
| original image and width, height the dimensions of the crop.</li> |
| </ul> |
| </p> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| <p class="rubric">Example</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="k">with</span> <span class="nb">open</span><span class="p">(</span><span class="s2">"flower.jpg"</span><span class="p">,</span> <span class="s1">'rb'</span><span class="p">)</span> <span class="k">as</span> <span class="n">fp</span><span class="p">:</span> |
| <span class="gp">... </span> <span class="n">str_image</span> <span class="o">=</span> <span class="n">fp</span><span class="o">.</span><span class="n">read</span><span class="p">()</span> |
| <span class="gp">...</span> |
| <span class="gp">>>> </span><span class="n">image</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">imdecode</span><span class="p">(</span><span class="n">str_image</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">image</span> |
| <span class="go"><ndarray 2321x3482x3="" @cpu(0)=""></ndarray></span> |
| <span class="gp">>>> </span><span class="n">cropped_image</span><span class="p">,</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">width</span><span class="p">,</span> <span class="n">height</span><span class="p">)</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">center_crop</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="p">(</span><span class="mi">1000</span><span class="p">,</span> <span class="mi">500</span><span class="p">))</span> |
| <span class="gp">>>> </span><span class="n">cropped_image</span> |
| <span class="go"><ndarray 500x1000x3="" @cpu(0)=""></ndarray></span> |
| <span class="gp">>>> </span><span class="n">x</span><span class="p">,</span> <span class="n">y</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">height</span> |
| <span class="go">(1241, 910, 1000, 500)</span> |
| </pre></div> |
| </div> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.color_normalize"> |
| <code class="descclassname">image.</code><code class="descname">color_normalize</code><span class="sig-paren">(</span><em>src</em>, <em>mean</em>, <em>std=None</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.color_normalize" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Normalize src with mean and std.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple"> |
| <li><strong>src</strong> (<a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – Input image</li> |
| <li><strong>mean</strong> (<a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – RGB mean to be subtracted</li> |
| <li><strong>std</strong> (<a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – RGB standard deviation to be divided</li> |
| </ul> |
| </td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">An <cite>NDArray</cite> containing the normalized image.</p> |
| </td> |
| </tr> |
| <tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last"><a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray">NDArray</a></p> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.random_size_crop"> |
| <code class="descclassname">image.</code><code class="descname">random_size_crop</code><span class="sig-paren">(</span><em>src</em>, <em>size</em>, <em>min_area</em>, <em>ratio</em>, <em>interp=2</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.random_size_crop" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Randomly crop src with size. Randomize area and aspect ratio.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple"> |
| <li><strong>src</strong> (<a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – Input image</li> |
| <li><strong>size</strong> (<em>tuple of</em><em> (</em><em>int</em><em>, </em><em>int</em><em>)</em><em></em>) – Size of the crop formatted as (width, height).</li> |
| <li><strong>min_area</strong> (<em>int</em>) – Minimum area to be maintained after cropping</li> |
| <li><strong>ratio</strong> (<em>tuple of</em><em> (</em><em>float</em><em>, </em><em>float</em><em>)</em><em></em>) – Aspect ratio range as (min_aspect_ratio, max_aspect_ratio)</li> |
| <li><strong>interp</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=2</em>) – Interpolation method. See resize_short for details.</li> |
| </ul> |
| </td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first last"><ul class="simple"> |
| <li><em>NDArray</em> – An <cite>NDArray</cite> containing the cropped image.</li> |
| <li><em>Tuple</em> – A tuple (x, y, width, height) where (x, y) is top-left position of the crop in the |
| original image and (width, height) are the dimensions of the cropped image.</li> |
| </ul> |
| </p> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.Augmenter"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">Augmenter</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.Augmenter" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Image Augmenter base class</p> |
| <dl class="method"> |
| <dt id="mxnet.image.Augmenter.dumps"> |
| <code class="descname">dumps</code><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.Augmenter.dumps" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Saves the Augmenter to string</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Returns:</th><td class="field-body">JSON formatted string that describes the Augmenter.</td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Return type:</th><td class="field-body">str</td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.ResizeAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">ResizeAug</code><span class="sig-paren">(</span><em>size</em>, <em>interp=2</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ResizeAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Make resize shorter edge to size augmenter.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>size</strong> (<em>int</em>) – The length to be set for the shorter edge.</li> |
| <li><strong>interp</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=2</em>) – Interpolation method. See resize_short for details.</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.ForceResizeAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">ForceResizeAug</code><span class="sig-paren">(</span><em>size</em>, <em>interp=2</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ForceResizeAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Force resize to size regardless of aspect ratio</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>size</strong> (<em>tuple of</em><em> (</em><em>int</em><em>, </em><em>int</em><em>)</em><em></em>) – The desired size as in (width, height)</li> |
| <li><strong>interp</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=2</em>) – Interpolation method. See resize_short for details.</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.RandomCropAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">RandomCropAug</code><span class="sig-paren">(</span><em>size</em>, <em>interp=2</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.RandomCropAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Make random crop augmenter</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>size</strong> (<em>int</em>) – The length to be set for the shorter edge.</li> |
| <li><strong>interp</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=2</em>) – Interpolation method. See resize_short for details.</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.RandomSizedCropAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">RandomSizedCropAug</code><span class="sig-paren">(</span><em>size</em>, <em>min_area</em>, <em>ratio</em>, <em>interp=2</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.RandomSizedCropAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Make random crop with random resizing and random aspect ratio jitter augmenter.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>size</strong> (<em>tuple of</em><em> (</em><em>int</em><em>, </em><em>int</em><em>)</em><em></em>) – Size of the crop formatted as (width, height).</li> |
| <li><strong>min_area</strong> (<em>int</em>) – Minimum area to be maintained after cropping</li> |
| <li><strong>ratio</strong> (<em>tuple of</em><em> (</em><em>float</em><em>, </em><em>float</em><em>)</em><em></em>) – Aspect ratio range as (min_aspect_ratio, max_aspect_ratio)</li> |
| <li><strong>interp</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=2</em>) – Interpolation method. See resize_short for details.</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.CenterCropAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">CenterCropAug</code><span class="sig-paren">(</span><em>size</em>, <em>interp=2</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.CenterCropAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Make center crop augmenter.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>size</strong> (<em>list</em><em> or </em><em>tuple of int</em>) – The desired output image size.</li> |
| <li><strong>interp</strong> (<em>int</em><em>, </em><em>optional</em><em>, </em><em>default=2</em>) – Interpolation method. See resize_short for details.</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.RandomOrderAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">RandomOrderAug</code><span class="sig-paren">(</span><em>ts</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.RandomOrderAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Apply list of augmenters in random order</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>ts</strong> (<em>list of augmenters</em>) – A series of augmenters to be applied in random order</td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.BrightnessJitterAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">BrightnessJitterAug</code><span class="sig-paren">(</span><em>brightness</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.BrightnessJitterAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Random brightness jitter augmentation.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>brightness</strong> (<em>float</em>) – The brightness jitter ratio range, [0, 1]</td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.ContrastJitterAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">ContrastJitterAug</code><span class="sig-paren">(</span><em>contrast</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ContrastJitterAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Random contrast jitter augmentation.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>contrast</strong> (<em>float</em>) – The contrast jitter ratio range, [0, 1]</td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.SaturationJitterAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">SaturationJitterAug</code><span class="sig-paren">(</span><em>saturation</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.SaturationJitterAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Random saturation jitter augmentation.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>saturation</strong> (<em>float</em>) – The saturation jitter ratio range, [0, 1]</td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.HueJitterAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">HueJitterAug</code><span class="sig-paren">(</span><em>hue</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.HueJitterAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Random hue jitter augmentation.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>hue</strong> (<em>float</em>) – The hue jitter ratio range, [0, 1]</td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.ColorJitterAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">ColorJitterAug</code><span class="sig-paren">(</span><em>brightness</em>, <em>contrast</em>, <em>saturation</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ColorJitterAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Apply random brightness, contrast and saturation jitter in random order.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>brightness</strong> (<em>float</em>) – The brightness jitter ratio range, [0, 1]</li> |
| <li><strong>contrast</strong> (<em>float</em>) – The contrast jitter ratio range, [0, 1]</li> |
| <li><strong>saturation</strong> (<em>float</em>) – The saturation jitter ratio range, [0, 1]</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.LightingAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">LightingAug</code><span class="sig-paren">(</span><em>alphastd</em>, <em>eigval</em>, <em>eigvec</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.LightingAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Add PCA based noise.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>alphastd</strong> (<em>float</em>) – Noise level</li> |
| <li><strong>eigval</strong> (<em>3x1 np.array</em>) – Eigen values</li> |
| <li><strong>eigvec</strong> (<em>3x3 np.array</em>) – Eigen vectors</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.ColorNormalizeAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">ColorNormalizeAug</code><span class="sig-paren">(</span><em>mean</em>, <em>std</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ColorNormalizeAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Mean and std normalization.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>mean</strong> (<a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – RGB mean to be subtracted</li> |
| <li><strong>std</strong> (<a class="reference internal" href="ndarray.html#mxnet.ndarray.NDArray" title="mxnet.ndarray.NDArray"><em>NDArray</em></a>) – RGB standard deviation to be divided</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.RandomGrayAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">RandomGrayAug</code><span class="sig-paren">(</span><em>p</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.RandomGrayAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Randomly convert to gray image.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>p</strong> (<em>float</em>) – Probability to convert to grayscale</td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.HorizontalFlipAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">HorizontalFlipAug</code><span class="sig-paren">(</span><em>p</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.HorizontalFlipAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Random horizontal flip.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>p</strong> (<em>float</em>) – Probability to flip image horizontally</td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.CastAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">CastAug</code><a class="headerlink" href="#mxnet.image.CastAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Cast to float32</p> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.CreateAugmenter"> |
| <code class="descclassname">image.</code><code class="descname">CreateAugmenter</code><span class="sig-paren">(</span><em>data_shape</em>, <em>resize=0</em>, <em>rand_crop=False</em>, <em>rand_resize=False</em>, <em>rand_mirror=False</em>, <em>mean=None</em>, <em>std=None</em>, <em>brightness=0</em>, <em>contrast=0</em>, <em>saturation=0</em>, <em>hue=0</em>, <em>pca_noise=0</em>, <em>rand_gray=0</em>, <em>inter_method=2</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.CreateAugmenter" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Creates an augmenter list.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>data_shape</strong> (<em>tuple of int</em>) – shape for output data</li> |
| <li><strong>resize</strong> (<em>int</em>) – resize shorter edge if larger than 0 at the begining</li> |
| <li><strong>rand_resize</strong> (<em>float</em>) – [0, 1], probability to apply random resizing</li> |
| <li><strong>rand_gray</strong> (<em>float</em>) – [0, 1], probability to convert to grayscale for all channels</li> |
| <li><strong>rand_mirror</strong> (<em>bool</em>) – whether apply horizontal flip to image with probability 0.5</li> |
| <li><strong>mean</strong> (<em>np.ndarray</em><em> or </em><em>None</em>) – mean pixel values for [r, g, b]</li> |
| <li><strong>std</strong> (<em>np.ndarray</em><em> or </em><em>None</em>) – standard deviations for [r, g, b]</li> |
| <li><strong>brightness</strong> (<em>float</em>) – brightness jittering range (percent)</li> |
| <li><strong>contrast</strong> (<em>float</em>) – contrast jittering range</li> |
| <li><strong>saturation</strong> (<em>float</em>) – saturation jittering range</li> |
| <li><strong>hue</strong> (<em>float</em>) – hue jittering range</li> |
| <li><strong>pca_noise</strong> (<em>float</em>) – pca noise level</li> |
| <li><strong>inter_method</strong> (<em>int</em><em>, </em><em>default=2</em><em>(</em><em>Area-based</em><em>)</em><em></em>) – <p>interpolation method for all resizing operations</p> |
| <p>Possible values: |
| 0: Nearest Neighbors Interpolation. |
| 1: Bilinear interpolation. |
| 2: Area-based (resampling using pixel area relation). It may be a |
| preferred method for image decimation, as it gives moire-free |
| results. But when the image is zoomed, it is similar to the Nearest |
| Neighbors method. (used by default). |
| 3: Bicubic interpolation over 4x4 pixel neighborhood. |
| 4: Lanczos interpolation over 8x8 pixel neighborhood. |
| 9: Cubic for enlarge, area for shrink, bilinear for others |
| 10: Random select from interpolation method metioned above. |
| Note: |
| When shrinking an image, it will generally look best with AREA-based |
| interpolation, whereas, when enlarging an image, it will generally look best |
| with Bicubic (slow) or Bilinear (faster but still looks OK).</p> |
| </li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| <p class="rubric">Examples</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="c1"># An example of creating multiple augmenters</span> |
| <span class="gp">>>> </span><span class="n">augs</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">CreateAugmenter</span><span class="p">(</span><span class="n">data_shape</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">300</span><span class="p">),</span> <span class="n">rand_mirror</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> |
| <span class="gp">... </span> <span class="n">mean</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">brightness</span><span class="o">=</span><span class="mf">0.125</span><span class="p">,</span> <span class="n">contrast</span><span class="o">=</span><span class="mf">0.125</span><span class="p">,</span> <span class="n">rand_gray</span><span class="o">=</span><span class="mf">0.05</span><span class="p">,</span> |
| <span class="gp">... </span> <span class="n">saturation</span><span class="o">=</span><span class="mf">0.125</span><span class="p">,</span> <span class="n">pca_noise</span><span class="o">=</span><span class="mf">0.05</span><span class="p">,</span> <span class="n">inter_method</span><span class="o">=</span><span class="mi">10</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="c1"># dump the details</span> |
| <span class="gp">>>> </span><span class="k">for</span> <span class="n">aug</span> <span class="ow">in</span> <span class="n">augs</span><span class="p">:</span> |
| <span class="gp">... </span> <span class="n">aug</span><span class="o">.</span><span class="n">dumps</span><span class="p">()</span> |
| </pre></div> |
| </div> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.ImageDetIter"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">ImageDetIter</code><span class="sig-paren">(</span><em>batch_size</em>, <em>data_shape</em>, <em>path_imgrec=None</em>, <em>path_imglist=None</em>, <em>path_root=None</em>, <em>path_imgidx=None</em>, <em>shuffle=False</em>, <em>part_index=0</em>, <em>num_parts=1</em>, <em>aug_list=None</em>, <em>imglist=None</em>, <em>data_name='data'</em>, <em>label_name='label'</em>, <em>**kwargs</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageDetIter" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Image iterator with a large number of augmentation choices for detection.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>aug_list</strong> (<em>list</em><em> or </em><em>None</em>) – augmenter list for generating distorted images</li> |
| <li><strong>batch_size</strong> (<em>int</em>) – Number of examples per batch.</li> |
| <li><strong>data_shape</strong> (<em>tuple</em>) – Data shape in (channels, height, width) format. |
| For now, only RGB image with 3 channels is supported.</li> |
| <li><strong>path_imgrec</strong> (<em>str</em>) – Path to image record file (.rec). |
| Created with tools/im2rec.py or bin/im2rec.</li> |
| <li><strong>path_imglist</strong> (<em>str</em>) – Path to image list (.lst). |
| Created with tools/im2rec.py or with custom script. |
| Format: Tab separated record of index, one or more labels and relative_path_from_root.</li> |
| <li><strong>imglist</strong> (<em>list</em>) – A list of images with the label(s). |
| Each item is a list [imagelabel: float or list of float, imgpath].</li> |
| <li><strong>path_root</strong> (<em>str</em>) – Root folder of image files.</li> |
| <li><strong>path_imgidx</strong> (<em>str</em>) – Path to image index file. Needed for partition and shuffling when using .rec source.</li> |
| <li><strong>shuffle</strong> (<em>bool</em>) – Whether to shuffle all images at the start of each iteration or not. |
| Can be slow for HDD.</li> |
| <li><strong>part_index</strong> (<em>int</em>) – Partition index.</li> |
| <li><strong>num_parts</strong> (<em>int</em>) – Total number of partitions.</li> |
| <li><strong>data_name</strong> (<em>str</em>) – Data name for provided symbols.</li> |
| <li><strong>label_name</strong> (<em>str</em>) – name for detection labels</li> |
| <li><strong>kwargs</strong> – More arguments for creating augmenter. See mx.image.CreateDetAugmenter.</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageDetIter.reshape"> |
| <code class="descname">reshape</code><span class="sig-paren">(</span><em>data_shape=None</em>, <em>label_shape=None</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageDetIter.reshape" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Reshape iterator for data_shape or label_shape.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>data_shape</strong> (<em>tuple</em><em> or </em><em>None</em>) – reshape the data_shape to the new shape if not None</li> |
| <li><strong>label_shape</strong> (<em>tuple</em><em> or </em><em>None</em>) – reshape label shape to new shape if not None</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageDetIter.next"> |
| <code class="descname">next</code><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageDetIter.next" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Override the function for returning next batch.</p> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageDetIter.augmentation_transform"> |
| <code class="descname">augmentation_transform</code><span class="sig-paren">(</span><em>data</em>, <em>label</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageDetIter.augmentation_transform" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Override Transforms input data with specified augmentations.</p> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageDetIter.check_label_shape"> |
| <code class="descname">check_label_shape</code><span class="sig-paren">(</span><em>label_shape</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageDetIter.check_label_shape" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Checks if the new label shape is valid</p> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageDetIter.draw_next"> |
| <code class="descname">draw_next</code><span class="sig-paren">(</span><em>color=None</em>, <em>thickness=2</em>, <em>mean=None</em>, <em>std=None</em>, <em>clip=True</em>, <em>waitKey=None</em>, <em>window_name='draw_next'</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageDetIter.draw_next" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Display next image with bounding boxes drawn.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple"> |
| <li><strong>color</strong> (<em>tuple</em>) – Bounding box color in RGB, use None for random color</li> |
| <li><strong>thickness</strong> (<em>int</em>) – Bounding box border thickness</li> |
| <li><strong>mean</strong> (<em>True</em><em> or </em><em>numpy.ndarray</em>) – Compensate for the mean to have better visual effect</li> |
| <li><strong>std</strong> (<em>True</em><em> or </em><em>numpy.ndarray</em>) – Revert standard deviations</li> |
| <li><strong>clip</strong> (<em>bool</em>) – If true, clip to [0, 255] for better visual effect</li> |
| <li><strong>waitKey</strong> (<em>None</em><em> or </em><em>int</em>) – Hold the window for waitKey milliseconds if set, skip ploting if None</li> |
| <li><strong>window_name</strong> (<em>str</em>) – Plot window name if waitKey is set.</li> |
| </ul> |
| </td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first"></p> |
| </td> |
| </tr> |
| <tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last">numpy.ndarray</p> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| <p class="rubric">Examples</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="c1"># use draw_next to get images with bounding boxes drawn</span> |
| <span class="gp">>>> </span><span class="n">iterator</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">ImageDetIter</span><span class="p">(</span><span class="mi">1</span><span class="p">,</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">600</span><span class="p">,</span> <span class="mi">600</span><span class="p">),</span> <span class="n">path_imgrec</span><span class="o">=</span><span class="s1">'train.rec'</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="k">for</span> <span class="n">image</span> <span class="ow">in</span> <span class="n">iterator</span><span class="o">.</span><span class="n">draw_next</span><span class="p">(</span><span class="n">waitKey</span><span class="o">=</span><span class="kc">None</span><span class="p">):</span> |
| <span class="gp">... </span> <span class="c1"># display image</span> |
| <span class="gp">>>> </span><span class="c1"># or let draw_next display using cv2 module</span> |
| <span class="gp">>>> </span><span class="k">for</span> <span class="n">image</span> <span class="ow">in</span> <span class="n">iterator</span><span class="o">.</span><span class="n">draw_next</span><span class="p">(</span><span class="n">waitKey</span><span class="o">=</span><span class="mi">0</span><span class="p">,</span> <span class="n">window_name</span><span class="o">=</span><span class="s1">'disp'</span><span class="p">):</span> |
| <span class="gp">... </span> <span class="k">pass</span> |
| </pre></div> |
| </div> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.ImageDetIter.sync_label_shape"> |
| <code class="descname">sync_label_shape</code><span class="sig-paren">(</span><em>it</em>, <em>verbose=False</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.ImageDetIter.sync_label_shape" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Synchronize label shape with the input iterator. This is useful when |
| train/validation iterators have different label padding.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first simple"> |
| <li><strong>it</strong> (<a class="reference internal" href="#mxnet.image.ImageDetIter" title="mxnet.image.ImageDetIter"><em>ImageDetIter</em></a>) – The other iterator to synchronize</li> |
| <li><strong>verbose</strong> (<em>bool</em>) – Print verbose log if true</li> |
| </ul> |
| </td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Returns:</th><td class="field-body"><p class="first">The synchronized other iterator, the internal label shape is updated as well.</p> |
| </td> |
| </tr> |
| <tr class="field-odd field"><th class="field-name">Return type:</th><td class="field-body"><p class="first last"><a class="reference internal" href="#mxnet.image.ImageDetIter" title="mxnet.image.ImageDetIter">ImageDetIter</a></p> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| <p class="rubric">Examples</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="n">train_iter</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">ImageDetIter</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">300</span><span class="p">),</span> <span class="n">path_imgrec</span><span class="o">=</span><span class="s1">'train.rec'</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">val_iter</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">ImageDetIter</span><span class="p">(</span><span class="mi">32</span><span class="p">,</span> <span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">300</span><span class="p">),</span> <span class="n">path</span><span class="o">.</span><span class="n">imgrec</span><span class="o">=</span><span class="s1">'val.rec'</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">train_iter</span><span class="o">.</span><span class="n">label_shape</span> |
| <span class="go">(30, 6)</span> |
| <span class="gp">>>> </span><span class="n">val_iter</span><span class="o">.</span><span class="n">label_shape</span> |
| <span class="go">(25, 6)</span> |
| <span class="gp">>>> </span><span class="n">val_iter</span> <span class="o">=</span> <span class="n">train_iter</span><span class="o">.</span><span class="n">sync_label_shape</span><span class="p">(</span><span class="n">val_iter</span><span class="p">,</span> <span class="n">verbose</span><span class="o">=</span><span class="kc">False</span><span class="p">)</span> |
| <span class="gp">>>> </span><span class="n">train_iter</span><span class="o">.</span><span class="n">label_shape</span> |
| <span class="go">(30, 6)</span> |
| <span class="gp">>>> </span><span class="n">val_iter</span><span class="o">.</span><span class="n">label_shape</span> |
| <span class="go">(30, 6)</span> |
| </pre></div> |
| </div> |
| </dd></dl> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.DetAugmenter"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">DetAugmenter</code><span class="sig-paren">(</span><em>**kwargs</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.DetAugmenter" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Detection base augmenter</p> |
| <dl class="method"> |
| <dt id="mxnet.image.DetAugmenter.dumps"> |
| <code class="descname">dumps</code><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.DetAugmenter.dumps" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Saves the Augmenter to string</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Returns:</th><td class="field-body">JSON formatted string that describes the Augmenter.</td> |
| </tr> |
| <tr class="field-even field"><th class="field-name">Return type:</th><td class="field-body">str</td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.DetBorrowAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">DetBorrowAug</code><span class="sig-paren">(</span><em>augmenter</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.DetBorrowAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Borrow standard augmenter from image classification. |
| Which is good once you know label won’t be affected after this augmenter.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>augmenter</strong> (<em>mx.image.Augmenter</em>) – The borrowed standard augmenter which has no effect on label</td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.DetRandomSelectAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">DetRandomSelectAug</code><span class="sig-paren">(</span><em>aug_list</em>, <em>skip_prob=0</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.DetRandomSelectAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Randomly select one augmenter to apply, with chance to skip all.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>aug_list</strong> (<em>list of DetAugmenter</em>) – The random selection will be applied to one of the augmenters</li> |
| <li><strong>skip_prob</strong> (<em>float</em>) – The probability to skip all augmenters and return input directly</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.DetHorizontalFlipAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">DetHorizontalFlipAug</code><span class="sig-paren">(</span><em>p</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.DetHorizontalFlipAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Random horizontal flipping.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><strong>p</strong> (<em>float</em>) – chance [0, 1] to flip</td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.DetRandomCropAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">DetRandomCropAug</code><span class="sig-paren">(</span><em>min_object_covered=0.1</em>, <em>aspect_ratio_range=(0.75</em>, <em>1.33)</em>, <em>area_range=(0.05</em>, <em>1.0)</em>, <em>min_eject_coverage=0.3</em>, <em>max_attempts=50</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.DetRandomCropAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Random cropping with constraints</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>min_object_covered</strong> (<em>float</em><em>, </em><em>default=0.1</em>) – The cropped area of the image must contain at least this fraction of |
| any bounding box supplied. The value of this parameter should be non-negative. |
| In the case of 0, the cropped area does not need to overlap any of the |
| bounding boxes supplied.</li> |
| <li><strong>min_eject_coverage</strong> (<em>float</em><em>, </em><em>default=0.3</em>) – The minimum coverage of cropped sample w.r.t its original size. With this |
| constraint, objects that have marginal area after crop will be discarded.</li> |
| <li><strong>aspect_ratio_range</strong> (<em>tuple of floats</em><em>, </em><em>default=</em><em>(</em><em>0.75</em><em>, </em><em>1.33</em><em>)</em><em></em>) – The cropped area of the image must have an aspect ratio = width / height |
| within this range.</li> |
| <li><strong>area_range</strong> (<em>tuple of floats</em><em>, </em><em>default=</em><em>(</em><em>0.05</em><em>, </em><em>1.0</em><em>)</em><em></em>) – The cropped area of the image must contain a fraction of the supplied |
| image within in this range.</li> |
| <li><strong>max_attempts</strong> (<em>int</em><em>, </em><em>default=50</em>) – Number of attempts at generating a cropped/padded region of the image of the |
| specified constraints. After max_attempts failures, return the original image.</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="class"> |
| <dt id="mxnet.image.DetRandomPadAug"> |
| <em class="property">class </em><code class="descclassname">mxnet.image.</code><code class="descname">DetRandomPadAug</code><span class="sig-paren">(</span><em>aspect_ratio_range=(0.75</em>, <em>1.33)</em>, <em>area_range=(1.0</em>, <em>3.0)</em>, <em>max_attempts=50</em>, <em>pad_val=(128</em>, <em>128</em>, <em>128)</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.DetRandomPadAug" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Random padding augmenter.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>aspect_ratio_range</strong> (<em>tuple of floats</em><em>, </em><em>default=</em><em>(</em><em>0.75</em><em>, </em><em>1.33</em><em>)</em><em></em>) – The padded area of the image must have an aspect ratio = width / height |
| within this range.</li> |
| <li><strong>area_range</strong> (<em>tuple of floats</em><em>, </em><em>default=</em><em>(</em><em>1.0</em><em>, </em><em>3.0</em><em>)</em><em></em>) – The padded area of the image must be larger than the original area</li> |
| <li><strong>max_attempts</strong> (<em>int</em><em>, </em><em>default=50</em>) – Number of attempts at generating a padded region of the image of the |
| specified constraints. After max_attempts failures, return the original image.</li> |
| <li><strong>pad_val</strong> (<em>float</em><em> or </em><em>tuple of float</em><em>, </em><em>default=</em><em>(</em><em>128</em><em>, </em><em>128</em><em>, </em><em>128</em><em>)</em><em></em>) – pixel value to be filled when padding is enabled.</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| </dd></dl> |
| <dl class="method"> |
| <dt id="mxnet.image.CreateDetAugmenter"> |
| <code class="descclassname">image.</code><code class="descname">CreateDetAugmenter</code><span class="sig-paren">(</span><em>data_shape</em>, <em>resize=0</em>, <em>rand_crop=0</em>, <em>rand_pad=0</em>, <em>rand_gray=0</em>, <em>rand_mirror=False</em>, <em>mean=None</em>, <em>std=None</em>, <em>brightness=0</em>, <em>contrast=0</em>, <em>saturation=0</em>, <em>pca_noise=0</em>, <em>hue=0</em>, <em>inter_method=2</em>, <em>min_object_covered=0.1</em>, <em>aspect_ratio_range=(0.75</em>, <em>1.33)</em>, <em>area_range=(0.05</em>, <em>3.0)</em>, <em>min_eject_coverage=0.3</em>, <em>max_attempts=50</em>, <em>pad_val=(128</em>, <em>128</em>, <em>128)</em><span class="sig-paren">)</span><a class="headerlink" href="#mxnet.image.CreateDetAugmenter" title="Permalink to this definition">¶</a></dt> |
| <dd><p>Create augmenters for detection.</p> |
| <table class="docutils field-list" frame="void" rules="none"> |
| <col class="field-name"/> |
| <col class="field-body"/> |
| <tbody valign="top"> |
| <tr class="field-odd field"><th class="field-name">Parameters:</th><td class="field-body"><ul class="first last simple"> |
| <li><strong>data_shape</strong> (<em>tuple of int</em>) – shape for output data</li> |
| <li><strong>resize</strong> (<em>int</em>) – resize shorter edge if larger than 0 at the begining</li> |
| <li><strong>rand_crop</strong> (<em>float</em>) – [0, 1], probability to apply random cropping</li> |
| <li><strong>rand_pad</strong> (<em>float</em>) – [0, 1], probability to apply random padding</li> |
| <li><strong>rand_gray</strong> (<em>float</em>) – [0, 1], probability to convert to grayscale for all channels</li> |
| <li><strong>rand_mirror</strong> (<em>bool</em>) – whether apply horizontal flip to image with probability 0.5</li> |
| <li><strong>mean</strong> (<em>np.ndarray</em><em> or </em><em>None</em>) – mean pixel values for [r, g, b]</li> |
| <li><strong>std</strong> (<em>np.ndarray</em><em> or </em><em>None</em>) – standard deviations for [r, g, b]</li> |
| <li><strong>brightness</strong> (<em>float</em>) – brightness jittering range (percent)</li> |
| <li><strong>contrast</strong> (<em>float</em>) – contrast jittering range</li> |
| <li><strong>saturation</strong> (<em>float</em>) – saturation jittering range</li> |
| <li><strong>hue</strong> (<em>float</em>) – hue jittering range</li> |
| <li><strong>pca_noise</strong> (<em>float</em>) – pca noise level</li> |
| <li><strong>inter_method</strong> (<em>int</em><em>, </em><em>default=2</em><em>(</em><em>Area-based</em><em>)</em><em></em>) – <p>interpolation method for all resizing operations</p> |
| <p>Possible values: |
| 0: Nearest Neighbors Interpolation. |
| 1: Bilinear interpolation. |
| 2: Area-based (resampling using pixel area relation). It may be a |
| preferred method for image decimation, as it gives moire-free |
| results. But when the image is zoomed, it is similar to the Nearest |
| Neighbors method. (used by default). |
| 3: Bicubic interpolation over 4x4 pixel neighborhood. |
| 4: Lanczos interpolation over 8x8 pixel neighborhood. |
| 9: Cubic for enlarge, area for shrink, bilinear for others |
| 10: Random select from interpolation method metioned above. |
| Note: |
| When shrinking an image, it will generally look best with AREA-based |
| interpolation, whereas, when enlarging an image, it will generally look best |
| with Bicubic (slow) or Bilinear (faster but still looks OK).</p> |
| </li> |
| <li><strong>min_object_covered</strong> (<em>float</em>) – The cropped area of the image must contain at least this fraction of |
| any bounding box supplied. The value of this parameter should be non-negative. |
| In the case of 0, the cropped area does not need to overlap any of the |
| bounding boxes supplied.</li> |
| <li><strong>min_eject_coverage</strong> (<em>float</em>) – The minimum coverage of cropped sample w.r.t its original size. With this |
| constraint, objects that have marginal area after crop will be discarded.</li> |
| <li><strong>aspect_ratio_range</strong> (<em>tuple of floats</em>) – The cropped area of the image must have an aspect ratio = width / height |
| within this range.</li> |
| <li><strong>area_range</strong> (<em>tuple of floats</em>) – The cropped area of the image must contain a fraction of the supplied |
| image within in this range.</li> |
| <li><strong>max_attempts</strong> (<em>int</em>) – Number of attempts at generating a cropped/padded region of the image of the |
| specified constraints. After max_attempts failures, return the original image.</li> |
| <li><strong>pad_val</strong> (<em>float</em>) – pixel value to be filled when padding is enabled. pad_val will automatically |
| be subtracted by mean and divided by std if applicable.</li> |
| </ul> |
| </td> |
| </tr> |
| </tbody> |
| </table> |
| <p class="rubric">Examples</p> |
| <div class="highlight-default"><div class="highlight"><pre><span></span><span class="gp">>>> </span><span class="c1"># An example of creating multiple augmenters</span> |
| <span class="gp">>>> </span><span class="n">augs</span> <span class="o">=</span> <span class="n">mx</span><span class="o">.</span><span class="n">image</span><span class="o">.</span><span class="n">CreateDetAugmenter</span><span class="p">(</span><span class="n">data_shape</span><span class="o">=</span><span class="p">(</span><span class="mi">3</span><span class="p">,</span> <span class="mi">300</span><span class="p">,</span> <span class="mi">300</span><span class="p">),</span> <span class="n">rand_crop</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span> |
| <span class="gp">... </span> <span class="n">rand_pad</span><span class="o">=</span><span class="mf">0.5</span><span class="p">,</span> <span class="n">rand_mirror</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">mean</span><span class="o">=</span><span class="kc">True</span><span class="p">,</span> <span class="n">brightness</span><span class="o">=</span><span class="mf">0.125</span><span class="p">,</span> <span class="n">contrast</span><span class="o">=</span><span class="mf">0.125</span><span class="p">,</span> |
| <span class="gp">... </span> <span class="n">saturation</span><span class="o">=</span><span class="mf">0.125</span><span class="p">,</span> <span class="n">pca_noise</span><span class="o">=</span><span class="mf">0.05</span><span class="p">,</span> <span class="n">inter_method</span><span class="o">=</span><span class="mi">10</span><span class="p">,</span> <span class="n">min_object_covered</span><span class="o">=</span><span class="p">[</span><span class="mf">0.3</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.9</span><span class="p">],</span> |
| <span class="gp">... </span> <span class="n">area_range</span><span class="o">=</span><span class="p">(</span><span class="mf">0.3</span><span class="p">,</span> <span class="mf">3.0</span><span class="p">))</span> |
| <span class="gp">>>> </span><span class="c1"># dump the details</span> |
| <span class="gp">>>> </span><span class="k">for</span> <span class="n">aug</span> <span class="ow">in</span> <span class="n">augs</span><span class="p">:</span> |
| <span class="gp">... </span> <span class="n">aug</span><span class="o">.</span><span class="n">dumps</span><span class="p">()</span> |
| </pre></div> |
| </div> |
| </dd></dl> |
| <script>auto_index("api-reference");</script></div> |
| </div> |
| <div class="container"> |
| <div class="footer"> |
| <p> © 2015-2017 DMLC. All rights reserved. </p> |
| </div> |
| </div> |
| </div> |
| <div aria-label="main navigation" class="sphinxsidebar rightsidebar" role="navigation"> |
| <div class="sphinxsidebarwrapper"> |
| <h3><a href="../../index.html">Table Of Contents</a></h3> |
| <ul> |
| <li><a class="reference internal" href="#">Image API</a><ul> |
| <li><a class="reference internal" href="#overview">Overview</a></li> |
| <li><a class="reference internal" href="#image-processing-functions">Image processing functions</a></li> |
| <li><a class="reference internal" href="#image-iterators">Image iterators</a></li> |
| <li><a class="reference internal" href="#api-reference">API Reference</a></li> |
| </ul> |
| </li> |
| </ul> |
| </div> |
| </div> |
| </div> <!-- pagename != index --> |
| <script crossorigin="anonymous" integrity="sha384-0mSbJDEHialfmuBBQP6A4Qrprq5OVfW37PRR3j5ELqxss1yVqOtnepnHVP9aJ7xS" src="https://maxcdn.bootstrapcdn.com/bootstrap/3.3.6/js/bootstrap.min.js"></script> |
| <script src="../../_static/js/sidebar.js" type="text/javascript"></script> |
| <script src="../../_static/js/search.js" type="text/javascript"></script> |
| <script src="../../_static/js/navbar.js" type="text/javascript"></script> |
| <script src="../../_static/js/clipboard.min.js" type="text/javascript"></script> |
| <script src="../../_static/js/copycode.js" type="text/javascript"></script> |
| <script type="text/javascript"> |
| $('body').ready(function () { |
| $('body').css('visibility', 'visible'); |
| }); |
| </script> |
| </div></body> |
| </html> |
