dbmdz/flair-historic-ner-lft

We use BPEmbeddings instead of the combination of Wikipedia, Common Crawl and character embeddings (as used in the paper) to save space and training/inferencing time . Paper reported an averaged F1-score of 77.51. The model was trained on the LFT dataset, with an average F1 score of 76.32.76.3276.13.36.36-76.07.07-77.35.2077.21-20.20-87.51-77-51-51.39-57.39.37-39.38-38.40-39-36-36.38.39,