|
|
@@ -6,7 +6,7 @@
|
|
|
|
|
|
ga('create', 'UA-44373548-30', 'auto');
|
|
|
ga('send', 'pageview');
|
|
|
- </script><script src="/js/scrollSpy.js"></script><link rel="stylesheet" href="/css/main.css"/><script src="/js/codetabs.js"></script></head><body class="sideNavVisible"><div class="fixedHeaderContainer"><div class="headerWrapper wrapper"><header><a href="/"><img class="logo" src="/img/fasttext-icon-white-web.png" alt="fastText"/></a><div class="navigationWrapper navigationSlider"><nav class="slidingNav"><ul class="nav-site nav-site-internal"><li class=""><a href="/docs/en/support.html" target="_self">Docs</a></li><li class=""><a href="/docs/en/english-vectors.html" target="_self">Resources</a></li><li class="siteNavGroupActive"><a href="/blog/" target="_self">Blog</a></li><li class=""><a href="https://github.com/facebookresearch/fastText/" target="_blank">GitHub</a></li></ul></nav></div></header></div></div><div class="navPusher"><div class="docMainWrapper wrapper"><div class="container docsNavContainer" id="docsNav"><nav class="toc"><div class="toggleNav"><section class="navWrapper wrapper"><div class="navBreadcrumb wrapper"><div class="navToggle" id="navToggler"><div class="hamburger-menu"><div class="line1"></div><div class="line2"></div><div class="line3"></div></div></div><h2><i>›</i><span>Recent Posts</span></h2></div><div class="navGroups"><div class="navGroup"><h3 class="navGroupCategoryTitle">Recent Posts</h3><ul class=""><li class="navListItem"><a class="navItem" href="/blog/2019/06/25/blog-post.html">New release of python module</a></li><li class="navListItem"><a class="navItem" href="/blog/2017/10/02/blog-post.html">Language identification</a></li><li class="navListItem navListItemActive"><a class="navItem" href="/blog/2017/05/02/blog-post.html">fastText on mobile</a></li><li class="navListItem"><a class="navItem" href="/blog/2016/08/18/blog-post.html">Releasing fastText</a></li></ul></div></div></section></div><script>
|
|
|
+ </script><script src="/js/scrollSpy.js"></script><link rel="stylesheet" href="/css/main.css"/><script src="/js/codetabs.js"></script></head><body class="sideNavVisible"><div class="fixedHeaderContainer"><div class="headerWrapper wrapper"><header><a href="/"><img class="logo" src="/img/fasttext-icon-white-web.png" alt="fastText"/></a><div class="navigationWrapper navigationSlider"><nav class="slidingNav"><ul class="nav-site nav-site-internal"><li class=""><a href="/docs/en/support.html" target="_self">Docs</a></li><li class=""><a href="/docs/en/english-vectors.html" target="_self">Resources</a></li><li class="siteNavGroupActive"><a href="/blog/" target="_self">Blog</a></li><li class=""><a href="https://github.com/facebookresearch/fastText/" target="_blank">GitHub</a></li></ul></nav></div></header></div></div><div class="navPusher"><div class="docMainWrapper wrapper"><div class="docsNavContainer" id="docsNav"><nav class="toc"><div class="toggleNav"><section class="navWrapper wrapper"><div class="navBreadcrumb wrapper"><div class="navToggle" id="navToggler"><div class="hamburger-menu"><div class="line1"></div><div class="line2"></div><div class="line3"></div></div></div><h2><i>›</i><span>Recent Posts</span></h2></div><div class="navGroups"><div class="navGroup"><h3 class="navGroupCategoryTitle">Recent Posts</h3><ul class=""><li class="navListItem"><a class="navItem" href="/blog/2019/06/25/blog-post.html">New release of python module</a></li><li class="navListItem"><a class="navItem" href="/blog/2017/10/02/blog-post.html">Language identification</a></li><li class="navListItem navListItemActive"><a class="navItem" href="/blog/2017/05/02/blog-post.html">fastText on mobile</a></li><li class="navListItem"><a class="navItem" href="/blog/2016/08/18/blog-post.html">Releasing fastText</a></li></ul></div></div></section></div><script>
|
|
|
var coll = document.getElementsByClassName('collapsible');
|
|
|
var checkActiveCategory = true;
|
|
|
for (var i = 0; i < coll.length; i++) {
|
|
|
@@ -72,7 +72,7 @@
|
|
|
<p>fastText is designed to be simple to use for developers, domain experts, and students. Its speed allows you to iterate quickly and refine your models without specialized hardware. fastText models can be trained on more than a billion words on any multicore CPU in less than few minutes and can classify half a million sentences with hundreds of thousands of classes in less than a minute.</p>
|
|
|
<p>fastText classification compares favorably with more complex neural network architectures implemented for specialized GPU hardware. The performance comparisons were reported in another paper authored by the fastText team, “<a href="https://arxiv.org/pdf/1607.01759.pdf">Bag of Tricks for Efficient Text Classification</a>.” Little or no accuracy is lost with fastText compared with more complex neural network models. For example, fastText performed competitively on sentiment analysis problems when compared to the results of convolutional neural networks (Zhang et al. 2015).</p>
|
|
|
<p><img src="../../../../img/blog/2017-05-02-blog-post-img1.jpg" alt="fastText performance"></p>
|
|
|
-<h2><a class="anchor" aria-hidden="true" id="how-facebook-ai-research-engineered-fasttext-s-performance"></a><a href="#how-facebook-ai-research-engineered-fasttext-s-performance" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>How Facebook AI Research engineered fastText’s performance</h2>
|
|
|
+<h2><a class="anchor" aria-hidden="true" id="how-facebook-ai-research-engineered-fasttexts-performance"></a><a href="#how-facebook-ai-research-engineered-fasttexts-performance" aria-hidden="true" class="hash-link"><svg class="hash-link-icon" aria-hidden="true" height="16" version="1.1" viewBox="0 0 16 16" width="16"><path fill-rule="evenodd" d="M4 9h1v1H4c-1.5 0-3-1.69-3-3.5S2.55 3 4 3h4c1.45 0 3 1.69 3 3.5 0 1.41-.91 2.72-2 3.25V8.59c.58-.45 1-1.27 1-2.09C10 5.22 8.98 4 8 4H4c-.98 0-2 1.22-2 2.5S3 9 4 9zm9-3h-1v1h1c1 0 2 1.22 2 2.5S13.98 12 13 12H9c-.98 0-2-1.22-2-2.5 0-.83.42-1.64 1-2.09V6.25c-1.09.53-2 1.84-2 3.25C6 11.31 7.55 13 9 13h4c1.45 0 3-1.69 3-3.5S14.5 6 13 6z"></path></svg></a>How Facebook AI Research engineered fastText’s performance</h2>
|
|
|
<p>Throughout the history of machine learning, research developments have often outpaced hardware performance, and researchers have worked to optimize for practical applications by maximizing accuracy while minimizing computational complexity. Facebook's research teams have developed unique expertise in fitting the best possible model to the hardware available. With fastText, however, one additional constraint is extending machine learning capabilities to everyone who has a multicore CPU computer with a C++ compiler — pretty much everyone developing software or in an engineering role.</p>
|
|
|
<p>Given this expertise, we were able to build a simple yet powerful library to solve important text classification problems tailored for generic, less powerful hardware. fastText is both impactful as a library for learning text classification and for adding accurate text classification features to applications. fastText also enables developers to add text classification features such as ranking comments with hashtags and ranking reviews based on sentiment analysis without a formal machine learning education.</p>
|
|
|
<p>Low dimensional vectors were used to improve performance. Large vectors improve accuracy because of the larger number of features in the word vector but are computationally expensive at training time. State of the art performance is possible with low dimensional vectors if the right features are present and the models can scale to a very large corpus. During encoding, vector size is reduced by presenting examples of low dimensional vectors obtained through conventional optimization methods.</p>
|
Website Deployment Script