rz-zhang / Active-NER

Bayesian Deep Active Learning for Named entity recognition (NER)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Active-NLP

Partly refer Active-NLP

For example to run a CNN_BiLSTM_CRF model on Conll dataset on full dataset, you can run

 $ python train_ner.py --usemodel CNN_BiLSTM_CRF --dataset conll

and to run active learning for CNN_BiLSTM_CRF model on Conll dataset with "MNLP" acquisition function, you can run

 $ python active_ner.py --usemodel CNN_BiLSTM_CRF --dataset conll --acquiremethod mnlp

For plots to evaluate the result, you can run

 $ python eval.py

About

Bayesian Deep Active Learning for Named entity recognition (NER)

License:GNU General Public License v3.0


Languages

Language:Python 61.9%Language:Jupyter Notebook 38.1%