no code implementations • 3 Apr 2019 • Jan Kocoń, Michał Gawor
The article introduces a new set of Polish word embeddings, built using KGR10 corpus, which contains more than 4 billion words.
Word Embeddings