parser
parser copied to clipboard
yzhangcs
:rocket: State-of-the-art parsers for natural language.
Using predict(['This', 'is', '(not)', 'parsable', '!']) with the crf-con-en model doesn't work. I suspect you work with a parenthesis representation of the string which probably causes this conflict.
python -u -m supar.cmds.dep.biaffine train -b -d 0 -c dep-biaffine-xlmr -p model --train train.conllx \ --dev dev.conllx \ --test test.conllx \ -encoder=bert \ --bert=xlm-roberta-large \ --lr=5e-5 \ --lr-rate=20 \ --batch-size=500...
I'm having issue loading multilingual dependency parsing model, named biaffine-dep-xlmr. **System Specifications:** **_PyTorch_**: 2.3.0+cu121 **_Transformers_**: 4.41.1 **_Python_**: 3.10.14 **_Supar_**: 1.1.4 **Code to Reproduce:** from supar import Parser parser = Parser.load('biaffine-dep-xlmr')...
Hello! First, I want to extend a BIG thank you for your amazing work, clean code, and clear guidance! As someone new to this field, I have a question regarding...
After installing newers versions of torch and transformers (2.2.2 and 4.43.4, respectively), I've been receiving the error `AttributeError: 'BertTokenizerFast' object has no attribute 'split_special_tokens'` I tried installing from source, no...
