Facebook AI Research Open Sources fastText (research.facebook.com) 121 points by Nikhil_teja 10y ago ↗ HN
[–] Radim 10y ago ↗ Jupyter notebook with some numbers on performance and accuracy of fastText:https://github.com/jayantj/gensim/blob/fast_text_notebook/do...This compares fastText against the word2vec implementation in gensim, on the syntactic/semantic word analogy task.(raw pull request with comments: https://github.com/RaRe-Technologies/gensim/pull/815) [–] nl 10y ago ↗ Note that this is only looking at the embeddings side of FastText (which looks quite impressive).The classification side is equally interesting, and is better compared to things like VW, eg http://nlpers.blogspot.com.au/2016/08/fast-easy-baseline-tex...VW is a pretty strong baseline here, so equaling it is a good accomplishment. [–] wodenokoto 10y ago ↗ That notebook is great! thanks for sharing.
[–] nl 10y ago ↗ Note that this is only looking at the embeddings side of FastText (which looks quite impressive).The classification side is equally interesting, and is better compared to things like VW, eg http://nlpers.blogspot.com.au/2016/08/fast-easy-baseline-tex...VW is a pretty strong baseline here, so equaling it is a good accomplishment.
[–] minimaxir 10y ago ↗ Previous discussion on fasttext: https://news.ycombinator.com/item?id=12226988This submission is apparently the official announcement.
6 comments
[ 2.6 ms ] story [ 36.7 ms ] threadhttps://github.com/jayantj/gensim/blob/fast_text_notebook/do...
This compares fastText against the word2vec implementation in gensim, on the syntactic/semantic word analogy task.
(raw pull request with comments: https://github.com/RaRe-Technologies/gensim/pull/815)
The classification side is equally interesting, and is better compared to things like VW, eg http://nlpers.blogspot.com.au/2016/08/fast-easy-baseline-tex...
VW is a pretty strong baseline here, so equaling it is a good accomplishment.
This submission is apparently the official announcement.