TurkuNLP/bert-base-finnish-cased-v1

A version of Google's BERT deep transfer learning model for Finnish . FinBERT has been pre-trained for 1 million steps on over 3 billion tokens (24B characters) of Finnish text drawn from news, online discussion, and internet crawls . The model can be fine-tuned to achieve state-of-the-art results for various Finnish natural language processing tasks . It features a custom 50,000 wordpiece vocabulary that has much better coverage of Finnish words than e.g. the previously released multilingual BERT models from Google: . The Finnish BERT model outperforms multilingual . BERT (M-BERT) on document classification over a range of training set,