blob: 0b6e3c1d2eb2c25e81c8405c859464ea6e5cbcb8 [file] [log] [blame]
<!DOCTYPE html>
<!--[if lt IE 7]> <html class="no-js lt-ie9 lt-ie8 lt-ie7"> <![endif]-->
<!--[if IE 7]> <html class="no-js lt-ie9 lt-ie8"> <![endif]-->
<!--[if IE 8]> <html class="no-js lt-ie9"> <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js"> <!--<![endif]--><head>
<meta charset='utf-8'/><meta http-equiv='X-UA-Compatible' content='IE=edge'/><meta name='viewport' content='width=device-width, initial-scale=1'/><meta name='keywords' content='deep learning, eclipse collections, groovy, emoji, records, eclipse deeplearning4j'/><meta name='description' content='This post uses Emojis and Eclipse Collections to process pet records. We then use deep learning to predict whether each emoji looks like a dog or cat.'/><title>The Apache Groovy programming language - Blogs - Deep Learning and Eclipse Collections</title><link href='../img/favicon.ico' type='image/x-ico' rel='icon'/><link rel='stylesheet' type='text/css' href='../css/bootstrap.css'/><link rel='stylesheet' type='text/css' href='../css/font-awesome.min.css'/><link rel='stylesheet' type='text/css' href='../css/style.css'/><link rel='stylesheet' type='text/css' href='https://cdnjs.cloudflare.com/ajax/libs/prettify/r298/prettify.min.css'/>
</head><body>
<div id='fork-me'>
<a href='https://github.com/apache/groovy'>
<img style='position: fixed; top: 20px; right: -58px; border: 0; z-index: 100; transform: rotate(45deg);' src='/img/horizontal-github-ribbon.png'/>
</a>
</div><div id='st-container' class='st-container st-effect-9'>
<nav class='st-menu st-effect-9' id='menu-12'>
<h2 class='icon icon-lab'>Socialize</h2><ul>
<li>
<a href='http://groovy-lang.org/mailing-lists.html' class='icon'><span class='fa fa-envelope'></span> Discuss on the mailing-list</a>
</li><li>
<a href='https://twitter.com/ApacheGroovy' class='icon'><span class='fa fa-twitter'></span> Groovy on Twitter</a>
</li><li>
<a href='http://groovy-lang.org/events.html' class='icon'><span class='fa fa-calendar'></span> Events and conferences</a>
</li><li>
<a href='https://github.com/apache/groovy' class='icon'><span class='fa fa-github'></span> Source code on GitHub</a>
</li><li>
<a href='http://groovy-lang.org/reporting-issues.html' class='icon'><span class='fa fa-bug'></span> Report issues in Jira</a>
</li><li>
<a href='http://stackoverflow.com/questions/tagged/groovy' class='icon'><span class='fa fa-stack-overflow'></span> Stack Overflow questions</a>
</li><li>
<a href='http://groovycommunity.com/' class='icon'><span class='fa fa-slack'></span> Slack Community</a>
</li>
</ul>
</nav><div class='st-pusher'>
<div class='st-content'>
<div class='st-content-inner'>
<!--[if lt IE 7]>
<p class="browsehappy">You are using an <strong>outdated</strong> browser. Please <a href="http://browsehappy.com/">upgrade your browser</a> to improve your experience.</p>
<![endif]--><div><div class='navbar navbar-default navbar-static-top' role='navigation'>
<div class='container'>
<div class='navbar-header'>
<button type='button' class='navbar-toggle' data-toggle='collapse' data-target='.navbar-collapse'>
<span class='sr-only'></span><span class='icon-bar'></span><span class='icon-bar'></span><span class='icon-bar'></span>
</button><a class='navbar-brand' href='../index.html'>
<i class='fa fa-star'></i> Apache Groovy
</a>
</div><div class='navbar-collapse collapse'>
<ul class='nav navbar-nav navbar-right'>
<li class=''><a href='http://groovy-lang.org/learn.html'>Learn</a></li><li class=''><a href='http://groovy-lang.org/documentation.html'>Documentation</a></li><li class=''><a href='/download.html'>Download</a></li><li class=''><a href='http://groovy-lang.org/support.html'>Support</a></li><li class=''><a href='/'>Contribute</a></li><li class=''><a href='http://groovy-lang.org/ecosystem.html'>Ecosystem</a></li><li class=''><a href='https://groovy.apache.org/events.html'></a></li><li>
<a data-effect='st-effect-9' class='st-trigger' href='#'>Socialize</a>
</li><li class=''>
<a href='../search.html'>
<i class='fa fa-search'></i>
</a>
</li>
</ul>
</div>
</div>
</div><div id='content' class='page-1'><div class='row'><div class='row-fluid'><div class='col-lg-3'><ul class='nav-sidebar'><li><a href='./'>Blog index</a></li><li class='active'><a href='#doc'>Deep Learning and Eclipse Collections</a></li></ul><br/><ul class='nav-sidebar'><li style='padding: 0.35em 0.625em; background-color: #eee'><span>Related posts</span></li><li><a href='./deck-of-cards-with-groovy'>Deck of cards with Groovy, JDK collections and Eclipse Collections</a></li><li><a href='./zipping-collections-with-groovy'>Zipping Collections with Groovy</a></li><li><a href='./fruity-eclipse-collections'>Fruity Eclipse Collections</a></li><li><a href='./classifying-iris-flowers-with-deep'>Classifying Iris Flowers with Deep Learning, Groovy and GraalVM</a></li><li><a href='./groovy-null-processing'>Groovy Processing Nulls In Lists</a></li><li><a href='./reading-and-writing-csv-files'>Reading and Writing CSV files with Groovy</a></li><li><a href='./matrix-calculations-with-groovy-apache'>Matrix calculations with Groovy, Apache Commons Math, ojAlgo, Nd4j and EJML</a></li><li><a href='./detecting-objects-with-groovy-the'>Detecting objects with Groovy, the Deep Java Library (DJL), and Apache MXNet</a></li><li><a href='./comparators-and-sorting-in-groovy'>Comparators and Sorting in Groovy</a></li><li><a href='./wordle-checker'>Checking Wordle with Groovy</a></li><li><a href='./groovy-list-processing-cheat-sheet'>Groovy List Processing Cheat Sheet</a></li><li><a href='./using-groovy-with-apache-wayang'>Using Groovy with Apache Wayang and Apache Spark</a></li></ul></div><div class='col-lg-8 col-lg-pull-0'><a name='doc'></a><h1>Deep Learning and Eclipse Collections</h1><p><span>Author: <i>Paul King</i></span><br/><span>Published: 2022-10-11 10:41AM</span></p><hr/><div class="paragraph">
<p>In previous blogs, we&#8217;ve covered
<a href="https://groovy.apache.org/blog/deck-of-cards-with-groovy">Eclipse Collections</a> and
<a href="https://groovy.apache.org/blog/detecting-objects-with-groovy-the">Deep Learning</a>.
Recently, a couple of the highly recommended katas for Eclipse Collections have been
revamped to include "pet" and "fruit" emojis for a bit of extra fun. What could be better
than <em>Learning</em> Eclipse Collections?<em>Deep Learning</em> and Eclipse Collections of course!</p>
</div>
<div class="paragraph">
<p>First, we create a <code>PetType</code> enum with the emoji as <code>toString</code>:</p>
</div>
<div class="listingblock">
<div class="content">
<pre class="prettyprint highlight"><code data-lang="groovy">enum PetType {
CAT("🐱"),
DOG("🐶"),
HAMSTER("🐹"),
TURTLE("🐢"),
BIRD("🐦"),
SNAKE("🐍")
private final String emoji
PetType(String emoji) { this.emoji = emoji }
@Override
String toString() { emoji }
}</code></pre>
</div>
</div>
<div class="paragraph">
<p>Then a <code>Pet</code> record (with the <code>type</code> as the <code>toString</code>):</p>
</div>
<div class="listingblock">
<div class="content">
<pre class="prettyprint highlight"><code data-lang="groovy">record Pet(PetType type, String name, int age) {
String toString() { type.toString() }
}</code></pre>
</div>
</div>
<div class="paragraph">
<p>Similarly, we&#8217;ll create a <code>Person</code> record. We&#8217;ll also populate a <code>people</code> list as is done in the kata. The full details are in the <a href="https://github.com/paulk-asert/deep-learning-eclipse-collections">repo</a>.</p>
</div>
<div class="paragraph">
<p>Let&#8217;s use a GQuery expression to explore the pre-populated list:</p>
</div>
<div class="listingblock">
<div class="content">
<pre class="prettyprint highlight"><code data-lang="groovy">println GQ {
from p in people
select p.fullName, p.pets
}</code></pre>
</div>
</div>
<div class="paragraph">
<p>The result is:</p>
</div>
<div class="listingblock">
<div class="content">
<pre class="prettyprint highlight"><code>+---------------+----------+
| fullName | pets |
+---------------+----------+
| Mary Smith | [🐱] |
| Bob Smith | [🐱, 🐶] |
| Ted Smith | [🐶] |
| Jake Snake | [🐍] |
| Barry Bird | [🐦] |
| Terry Turtle | [🐢] |
| Harry Hamster | [🐹, 🐹] |
| John Doe | [] |
+---------------+----------+</code></pre>
</div>
</div>
<div class="paragraph">
<p>Now let&#8217;s duplicate the assertion from the <code>getCountsByPetType</code> test in the original kata&#8217;s exercise3 which checks pet counts:</p>
</div>
<div class="listingblock">
<div class="content">
<pre class="prettyprint highlight"><code data-lang="groovy">var counts = people.countByEach(person -&gt; person.petTypes).collect(Object::toString)
var expected = Bags.mutable.withOccurrences("🐱", 2, "🐶", 2, "🐹", 2).with("🐍").with("🐢").with("🐦")
assert counts == expected</code></pre>
</div>
</div>
<div class="paragraph">
<p>As we expect, it passes.</p>
</div>
<div class="paragraph">
<p>Now, for a bit of fun, we will use a neural network trained to detect cat and dog images and apply it to our emojis. We&#8217;ll follow the process described <a href="http://ramok.tech/2018/01/03/java-image-cat-vs-dog-recognizer-with-deep-neural-networks/">here</a>. It uses DeepLearning4J to train and then use a model. The images used to train the model were real cat and dog images, not emojis, so we aren&#8217;t expecting our model to be super accurate.</p>
</div>
<div class="paragraph">
<p>The first attempt was to write the emojis into swing JLabel components and then save using a buffered image. This lead to poor looking images:</p>
</div>
<div class="paragraph">
<p><span class="image"><img src="img/pet_emoji_fonts.jpg" alt="PetAsFonts"></span></p>
</div>
<div class="paragraph">
<p>And consequently, poor image inference. Recent JDK versions on some platforms might do better, but we gave up on this approach.</p>
</div>
<div class="paragraph">
<p>Instead, emoji image files from the <a href="https://fonts.google.com/noto/specimen/Noto+Color+Emoji?preview.text=%F0%9F%98%BB%F0%9F%90%B6%F0%9F%90%B9%F0%9F%90%A2%F0%9F%90%A6%F0%9F%90%8D&amp;preview.text_type=custom">Noto Color Emoji</a> font were used and saved under the pet type in the <code>resources</code> folder. These look much nicer:</p>
</div>
<div class="paragraph">
<p><span class="image"><img src="img/pet_emoji.png" alt="Noto Color Emoji"></span></p>
</div>
<div class="paragraph">
<p>Here is the code which makes use of those saved images to detect the animal types (note the use of type aliasing since we have two <code>PetType</code> classes; we rename one to <code>PT</code>):</p>
</div>
<div class="listingblock">
<div class="content">
<pre class="prettyprint highlight"><code data-lang="groovy">import ramo.klevis.ml.vg16.PetType as PT
import ramo.klevis.ml.vg16.VG16ForCat
var vg16ForCat = new VG16ForCat().tap{ loadModel() }
var results = []
people.each{ p -&gt;
results &lt;&lt; p.pets.collect { pet -&gt;
var file = new File("resources/${pet.type.name()}.png")
PT petType = vg16ForCat.detectCat(file, 0.675d)
var desc = switch(petType) {
case PT.CAT -&gt; 'is a cat'
case PT.DOG -&gt; 'is a dog'
default -&gt; 'is unknown'
}
"$pet.name $desc"
}
}
println results.flatten().join('\n')</code></pre>
</div>
</div>
<div class="paragraph">
<p>Note that the model exceeds the maximum allowable size for normal GitHub repos, so you should create it following the original repo <a href="https://github.com/klevis/CatAndDogRecognizer">instructions</a> and then store the resulting <code>model.zip</code> in the <code>resources</code> folder.</p>
</div>
<div class="paragraph">
<p>When we run the script, we get the following output:</p>
</div>
<div class="listingblock">
<div class="content">
<pre class="prettyprint highlight"><code>[main] INFO org.nd4j.linalg.factory.Nd4jBackend - Loaded [CpuBackend] backend
...
[main] INFO org.nd4j.linalg.api.ops.executioner.DefaultOpExecutioner - Blas vendor: [OPENBLAS]
...
==============================================================================================================
VertexName (VertexType) nIn,nOut TotalParams ParamsShape Vertex Inputs
==============================================================================================================
input_1 (InputVertex) -,- - - -
block1_conv1 (Frozen ConvolutionLayer) 3,64 1792 b:{1,64}, W:{64,3,3,3} [input_1]
block1_conv2 (Frozen ConvolutionLayer) 64,64 36928 b:{1,64}, W:{64,64,3,3} [block1_conv1]
block1_pool (Frozen SubsamplingLayer) -,- 0 - [block1_conv2]
block2_conv1 (Frozen ConvolutionLayer) 64,128 73856 b:{1,128}, W:{128,64,3,3} [block1_pool]
block2_conv2 (Frozen ConvolutionLayer) 128,128 147584 b:{1,128}, W:{128,128,3,3} [block2_conv1]
block2_pool (Frozen SubsamplingLayer) -,- 0 - [block2_conv2]
block3_conv1 (Frozen ConvolutionLayer) 128,256 295168 b:{1,256}, W:{256,128,3,3} [block2_pool]
block3_conv2 (Frozen ConvolutionLayer) 256,256 590080 b:{1,256}, W:{256,256,3,3} [block3_conv1]
block3_conv3 (Frozen ConvolutionLayer) 256,256 590080 b:{1,256}, W:{256,256,3,3} [block3_conv2]
block3_pool (Frozen SubsamplingLayer) -,- 0 - [block3_conv3]
block4_conv1 (Frozen ConvolutionLayer) 256,512 1180160 b:{1,512}, W:{512,256,3,3} [block3_pool]
block4_conv2 (Frozen ConvolutionLayer) 512,512 2359808 b:{1,512}, W:{512,512,3,3} [block4_conv1]
block4_conv3 (Frozen ConvolutionLayer) 512,512 2359808 b:{1,512}, W:{512,512,3,3} [block4_conv2]
block4_pool (Frozen SubsamplingLayer) -,- 0 - [block4_conv3]
block5_conv1 (Frozen ConvolutionLayer) 512,512 2359808 b:{1,512}, W:{512,512,3,3} [block4_pool]
block5_conv2 (Frozen ConvolutionLayer) 512,512 2359808 b:{1,512}, W:{512,512,3,3} [block5_conv1]
block5_conv3 (Frozen ConvolutionLayer) 512,512 2359808 b:{1,512}, W:{512,512,3,3} [block5_conv2]
block5_pool (Frozen SubsamplingLayer) -,- 0 - [block5_conv3]
flatten (PreprocessorVertex) -,- - - [block5_pool]
fc1 (Frozen DenseLayer) 25088,4096 102764544 b:{1,4096}, W:{25088,4096} [flatten]
fc2 (Frozen DenseLayer) 4096,4096 16781312 b:{1,4096}, W:{4096,4096} [fc1]
predictions (OutputLayer) 4096,2 8194 b:{1,2}, W:{4096,2} [fc2]
--------------------------------------------------------------------------------------------------------------
Total Parameters: 134268738
Trainable Parameters: 8194
Frozen Parameters: 134260544
==============================================================================================================
...
Tabby is a cat
Dolly is a cat
Spot is a dog
Spike is a dog
Serpy is a cat
Tweety is unknown
Speedy is a dog
Fuzzy is unknown
Wuzzy is unknown</code></pre>
</div>
</div>
<div class="paragraph">
<p>As we can see, it correctly predicted the cats (Tabby and Dolly) and dogs
(Spot and Spike) but incorrectly thought a snake (Serpy) was a cat and a
turtle (Speedy) was a dog. Given the lack of detail in the emoji images
compared to the training images, this lack of accuracy isn&#8217;t unexpected.
We could certainly use better images or train our model differently if
we wanted better results, but it is fun to see our model not doing too
badly even with emojis!</p>
</div></div></div></div></div><footer id='footer'>
<div class='row'>
<div class='colset-3-footer'>
<div class='col-1'>
<h1>Groovy</h1><ul>
<li><a href='http://groovy-lang.org/learn.html'>Learn</a></li><li><a href='http://groovy-lang.org/documentation.html'>Documentation</a></li><li><a href='/download.html'>Download</a></li><li><a href='http://groovy-lang.org/support.html'>Support</a></li><li><a href='/'>Contribute</a></li><li><a href='http://groovy-lang.org/ecosystem.html'>Ecosystem</a></li><li><a href='https://groovy.apache.org/events.html'></a></li>
</ul>
</div><div class='col-2'>
<h1>About</h1><ul>
<li><a href='https://github.com/apache/groovy'>Source code</a></li><li><a href='http://groovy-lang.org/security.html'>Security</a></li><li><a href='http://groovy-lang.org/learn.html#books'>Books</a></li><li><a href='http://groovy-lang.org/thanks.html'>Thanks</a></li><li><a href='http://www.apache.org/foundation/sponsorship.html'>Sponsorship</a></li><li><a href='http://groovy-lang.org/faq.html'>FAQ</a></li><li><a href='http://groovy-lang.org/search.html'>Search</a></li>
</ul>
</div><div class='col-3'>
<h1>Socialize</h1><ul>
<li><a href='http://groovy-lang.org/mailing-lists.html'>Discuss on the mailing-list</a></li><li><a href='https://twitter.com/ApacheGroovy'>Groovy on Twitter</a></li><li><a href='http://groovy-lang.org/events.html'>Events and conferences</a></li><li><a href='https://github.com/apache/groovy'>Source code on GitHub</a></li><li><a href='http://groovy-lang.org/reporting-issues.html'>Report issues in Jira</a></li><li><a href='http://stackoverflow.com/questions/tagged/groovy'>Stack Overflow questions</a></li><li><a href='http://groovycommunity.com/'>Slack Community</a></li>
</ul>
</div><div class='col-right'>
<p>
The Groovy programming language is supported by the <a href='http://www.apache.org'>Apache Software Foundation</a> and the Groovy community.
</p><div text-align='right'>
<img src='../img/asf_logo.png' title='The Apache Software Foundation' alt='The Apache Software Foundation' style='width:60%'/>
</div><p>Apache&reg; and the Apache feather logo are either registered trademarks or trademarks of The Apache Software Foundation.</p>
</div>
</div><div class='clearfix'>&copy; 2003-2023 the Apache Groovy project &mdash; Groovy is Open Source: <a href='http://www.apache.org/licenses/LICENSE-2.0.html' alt='Apache 2 License'>license</a>, <a href='https://privacy.apache.org/policies/privacy-policy-public.html'>privacy policy</a>.</div>
</div>
</footer></div>
</div>
</div>
</div>
</div><script src='../js/vendor/jquery-1.10.2.min.js' defer></script><script src='../js/vendor/classie.js' defer></script><script src='../js/vendor/bootstrap.js' defer></script><script src='../js/vendor/sidebarEffects.js' defer></script><script src='../js/vendor/modernizr-2.6.2.min.js' defer></script><script src='../js/plugins.js' defer></script><script src='https://cdnjs.cloudflare.com/ajax/libs/prettify/r298/prettify.min.js'></script><script>document.addEventListener('DOMContentLoaded',prettyPrint)</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','//www.google-analytics.com/analytics.js','ga');
ga('create', 'UA-257558-10', 'auto');
ga('send', 'pageview');
</script>
</body></html>