![](\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/05\/fasttext.jpg?w=738\")
<\/p>\n<\/p>\n
This morning Facebook\u2019s AI Research (FAIR) lab released an update to fastText<\/a>, its super speedy open source text classification\u00a0library. When it was initially released, fastText shipped with pre-trained word vectors for 90 languages, but today it\u2019s getting a boost to\u00a0294 languages. The release also brings enhancements to reduce model size and ultimately memory demand.<\/p>\n Text classifiers like fastText make it easy for developers to ship tools\u00a0that depend on underlying language analysis. Flagging clickbait headlines or\u00a0filtering spam\u00a0both require an underlying model that can interpret and categorize language.<\/p>\n From the start, fastText was designed to be implemented on a wide variety of hardware. Unfortunately, in its original state, it still required a few gigabytes of memory to run. This isn\u2019t a problem if you\u2019re working in a state of the art lab, but it\u2019s a deal killer\u00a0if you\u2019re trying to make things work\u00a0on mobile.<\/p>\n By collaborating with the team that produced\u00a0another Facebook open source project, similarity search (FAISS)<\/a>, the company was able to reduce down the memory requirement to just a few hundred kilobytes. FAISS addresses some of the inherent bottlenecks that developers face when\u00a0dealing with huge amounts of data.<\/p>\n A massive corpus\u00a0of\u00a0information\u00a0is often best\u00a0represented in\u00a0a multi-dimensional vector space. For Facebook and many other companies, optimizing the comparison\u00a0of these vectors for\u00a0comparing content with user preferences and comparing content with other content is critical. The approach of the FAISS team ended up playing a big role in reducing the memory\u00a0demands of fastText.<\/p>\n \u201cA few key ingredients, namely feature pruning, quantization, hashing, and re-training, allow us to produce text classification models with tiny size, often less than 100kB when trained on several popular datasets, without noticeably sacrificing accuracy or speed,\u201d said\u00a0the Facebook authors of a December 2016 paper entitled fastText.zip: Compressing Text Classification Models.<\/p>\n The authors went on to hypothesize that additional model size reduction might\u00a0be possible in the future. The challenge isn\u2019t so much shrinking the models as it is maintaining accuracy. But until then, engineers can access the updated library on GitHub<\/a> and begin tinkering today.<\/p>\n","protected":false},"excerpt":{"rendered":"![\"\"](\"https:\/\/tctechcrunch2011.files.wordpress.com\/2017\/05\/fasttext.jpg?w=680\")
This morning Facebook’s AI Research (FAIR) lab released an update to fastText, its super speedy open source text classification\u00a0library. When it was initially released, fastText shipped with pre-trained word vectors for 90 languages, but today it’s getting a boost to\u00a0294 languages. The release also brings enhancements to reduce model size and ultimately memory demand.
\nText… Read More<\/a><\/p>\n\n<\/img><\/a>
<\/img><\/a>
<\/img><\/a>
<\/img><\/a>
<\/img><\/a>
<\/img><\/a>\n<\/div>\n
<\/p>\n","protected":false},"author":5817,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_et_pb_use_builder":"","_et_pb_old_content":"","_et_gb_content_width":"","footnotes":""},"categories":[25],"tags":[59,66,26],"class_list":["post-339857","post","type-post","status-publish","format-standard","hentry","category-technology","tag-media","tag-techcrunch","tag-technology"],"_links":{"self":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts\/339857","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/users\/5817"}],"replies":[{"embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/comments?post=339857"}],"version-history":[{"count":0,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/posts\/339857\/revisions"}],"wp:attachment":[{"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/media?parent=339857"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/categories?post=339857"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/people.utm.my\/asmawisham\/wp-json\/wp\/v2\/tags?post=339857"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}