Facebook releases 300-dimensional pretrained Fasttext vectors for 90 languages

Re:Facebook releases 300-dimensional pretrained Fasttext vec... by ma2rten on Thursday March 02, 2017 @09:44AM (#13774267)

It doesn't say what data these were training on. That is kind of important information. I previously applied word vectors that were trained on news to social media posts and it didn't work well at all. Also I don't think there is a language called "Western".

Also interesting that this is hosted on S3.

[ Parent | on HN ]

Re:Facebook releases 300-dimensional pretrained Fasttext vec... by exgrv on Thursday March 02, 2017 @10:01AM (#13774424)

These models were trained on Wikipedia.

It should be "Western Frisian" instead of "Western" (https://en.wikipedia.org/wiki/West_Frisian_language). Thanks for the catch!

[ Parent | on HN ]

Re:Facebook releases 300-dimensional pretrained Fasttext vec... by ma2rten on Thursday March 02, 2017 @10:56AM (#13774978)

Thanks.

[ Parent | on HN ]

Re:Facebook releases 300-dimensional pretrained Fasttext vec... by coob on Thursday March 02, 2017 @10:03AM (#13774442)

> Also I don't think there is a language called "Western".

Going by the ISO 639 code I think that's supposed to be West Frisian

[ Parent | on HN ]