SRL-English
SRL-English copied to clipboard
separate_hyphens in predict_tokenized
Hi!
I wonder why do you have separate_hyphens in predict_tokenized for nominal_sense_srl. Will it affect the model if I remove the separate_hyphens step?