dkleczek/bert-base-polish-cased-v1

Polish version of BERT language model is now available in two variants: cased and uncased . Both can be downloaded and used via HuggingFace transformers library . I recommend using the cased model, more info on the differences and benchmark results below . Cased model improves on the uncased model in the following ways: All Polish characters and accents should now be tokenized correctly . Uncased model follows bert-base-uncased model architecture (12-layer, 768-hidden, 12-heads, 110M parameters) The model was trained on a single Google Cloud TPU v3-8 with the following differences:Whole Word Masking . The model is trained,