Adding Lambo tokenizer

Łukasz Pszenny authored 2 years ago and Łukasz Pszenny committed 2 years ago

adding scripts to train new UD with instruction, fixing numpy problem on Mac(M1), fixing link to download models

Łukasz Pszenny authored 3 years ago and Łukasz Pszenny committed 2 years ago

Update README.md

small fix in performance.md

small fix in performance.md - got rid of models that were trained on data with sentences of form "____"

Error while too many tokens are processed by BERT

Add models license information, make dataset_reader public attribute, exclude treebanks without data from script.

instructions.

Extend installation and pre-trained models documentation, fix installation, update trainer to be in sync with AllenNLP 1.2.1.