1 million English-language fastText word embeddings, w/subword information

A matrix of roughly 1 million word vectors (rows) and 300 dimensions (columns). Embeddings were trained trained with subword infomation on Wikipedia 2017, UMBC webbase corpus 2007 and statmt.org news dataset

Usage

data("vecs_fasttext300_wiki_news_subword")

Format

A matrix of 999,993 rows and 300 columns

Source

https://fasttext.cc/docs/en/english-vectors.html