An unofficial implementation of QANER
The repo was tested with Python 3.8
# cuda 11.3
conda install pytorch==1.11.0 cudatoolkit=11.3 -c pytorch
pip install -r requirements
Only supports running on CONLL03 (conll) and MIT restaurant (mit) datasets for now. Datasets are under data folder.
See utils/arguments.py for detailed argument instructions.
Train and test
python train.py --dataset mit --model-name "deepset/bert-large-uncased-whole-word-masking-squad2" --n 10 --num-negatives 2python train.py --dataset conll --model-name "deepset/bert-large-uncased-whole-word-masking-squad2" --n 10 --num-negatives 0Simply test/inference with a specified model
python inference.py --dataset mit --model-name "deepset/bert-large-uncased-whole-word-masking-squad2"