Pretrained embeddings

The pretrained embeddings are trained using finalfrontier.

Language Model Corpus Hyperparameters
Dutch skipgram Lassy Large Metadata
Dutch structured skipgram Lassy Large Metadata
English skipgram CoNLL 2017 raw Metadata
German skipgram TüBa-D/DP Metadata
German structured skipgram (2019-11-15) TüBa-D/DP Metadata
German structured skipgram explicit 30 (2019-11-16) TüBa-D/DP Metadata
German structured skipgram explicit 100 (2019-11-20) TüBa-D/DP Metadata
Arabic structured skipgram (2020-06-12) CoNLL 2017 raw Metadata
Arabic skipgram (2020-06-13) CoNLL 2017 raw Metadata
Basque structured skipgram (2020-06-12) CoNLL 2017 raw Metadata
Basque skipgram (2020-06-13) CoNLL 2017 raw Metadata
Bulgarian structured skipgram (2020-06-12) CoNLL 2017 raw Metadata
Bulgarian skipgram (2020-06-13) CoNLL 2017 raw Metadata
Catalan structured skipgram (2020-06-12) CoNLL 2017 raw Metadata
Catalan skipgram (2020-06-14) CoNLL 2017 raw Metadata
Chinese (Traditional) structured skipgram (2020-06-13) CoNLL 2017 raw Metadata
Chinese (Traditional) skipgram (2020-06-15) CoNLL 2017 raw Metadata
Croatian structured skipgram (2020-06-14) CoNLL 2017 raw Metadata
Croatian skipgram (2020-06-15) CoNLL 2017 raw Metadata
Czech structured skipgram (2020-06-15) CoNLL 2017 raw Metadata
Czech skipgram (2020-06-16) CoNLL 2017 raw Metadata
Danish structured skipgram (2020-06-16) CoNLL 2017 raw Metadata
Danish skipgram (2020-06-17) CoNLL 2017 raw Metadata
Dutch structured skipgram (2020-06-18) CoNLL 2017 raw Metadata
Dutch skipgram (2020-06-19) CoNLL 2017 raw Metadata
English structured skipgram (2020-06-23) CoNLL 2017 raw Metadata
English skipgram (2020-06-25) CoNLL 2017 raw Metadata
Estonian structured skipgram (2020-06-24) CoNLL 2017 raw Metadata
Estonian skipgram (2020-06-26) CoNLL 2017 raw Metadata
Finnish structured skipgram (2020-06-25) CoNLL 2017 raw Metadata
Finnish skipgram (2020-06-26) CoNLL 2017 raw Metadata
French structured skipgram (2020-06-30) CoNLL 2017 raw Metadata
French skipgram (2020-06-30) CoNLL 2017 raw Metadata
Galician structured skipgram (2020-07-01) CoNLL 2017 raw Metadata
Galician skipgram (2020-06-30) CoNLL 2017 raw Metadata
German structured skipgram (2020-07-11) CoNLL 2017 raw Metadata
German skipgram (2020-07-05) CoNLL 2017 raw Metadata
Greek structured skigpram (2020-07-11) CoNLL 2017 raw Metadata
Greek skipgram (2020-07-06) CoNLL 2017 raw Metadata
Hebrew structured skipgram (2020-07-12) CoNLL 2017 raw Metadata
Hebrew skipgram (2020-07-06) CoNLL 2017 raw Metadata
Hungarian structured skipgram (2020-07-13) CoNLL 2017 raw Metadata
Hungarian skipgram (2020-07-07) CoNLL 2017 raw Metadata
Indonesian structured skipgram (2020-07-17) CoNLL 2017 raw Metadata
Indonesian skipgram (2020-07-11) CoNLL 2017 raw Metadata
Italian structured skipgram (2020-07-20) CoNLL 2017 raw Metadata
Italian skipgram (2020-07-15) CoNLL 2017 raw Metadata
Japanese structured skipgram (2020-07-23) CoNLL 2017 raw Metadata
Japanese skipgram (2020-07-18) CoNLL 2017 raw Metadata
Korean structured skipgram (2020-07-23) CoNLL 2017 raw Metadata
Korean skipgram (2020-07-18) CoNLL 2017 raw Metadata
Latin structured skipgram (2020-07-23) CoNLL 2017 raw Metadata
Latin skipgram (2020-07-18) CoNLL 2017 raw Metadata
Latvian structured skipgram (2020-07-24) CoNLL 2017 raw Metadata
Latvian skipgram (2020-07-18) CoNLL 2017 raw Metadata
Norwegian Bokmaal structured skipgram (2020-07-24) CoNLL 2017 raw Metadata
Norwegian Bokmaal skipgram (2020-07-19) CoNLL 2017 raw Metadata
Persian structured skipgram (2020-07-25) CoNLL 2017 raw Metadata
Persian skipgram (2020-07-20) CoNLL 2017 raw Metadata
Polish structured skipgram (2020-07-29) CoNLL 2017 raw Metadata
Polish skipgram (2020-07-24) CoNLL 2017 raw Metadata
Portuguese skipgram (2020-07-28) CoNLL 2017 raw Metadata
Romanian skipgram (2020-07-30) CoNLL 2017 raw Metadata

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Converted Embeddings

We also provide conversions of the fastText Wikipedia and Common Crawl embeddings.