WORD RADAR DERIVED WORD-VECTOR DATASETS — CC BY-SA 3.0

Source: fastText word vectors trained on Common Crawl and Wikipedia.
https://fasttext.cc/docs/en/crawl-vectors.html
Original text vectors: https://dl.fbaipublicfiles.com/fasttext/vectors-crawl/cc.{en,ru,uk}.300.vec.gz

Attribution:
Edouard Grave, Piotr Bojanowski, Prakhar Gupta, Armand Joulin, Tomas Mikolov.
Learning Word Vectors for 157 Languages. LREC 2018.
https://arxiv.org/abs/1802.06893

The original word vectors and these modified subsets are distributed under
Creative Commons Attribution-ShareAlike 3.0 Unported (CC BY-SA 3.0).
License: https://creativecommons.org/licenses/by-sa/3.0/
Legal code: https://creativecommons.org/licenses/by-sa/3.0/legalcode

Modifications by Awesome Games: frequent-vocabulary selection, language and
noise filtering, canonical text normalization, unit normalization, per-row Int8
quantization, reviewed target subsets and form aliases, compact binary packaging.
No endorsement by the original authors is implied. Preserve attribution, indicate
changes and distribute adaptations of these datasets under the same license.
This notice applies to the derived vector datasets, not the application's code
or separately generated game icon.