Joint Entity and Relation Extraction with Set Prediction Networks
Source code for Joint Entity and Relation Extraction with Set Prediction Networks. We would appreciate it if you cite our paper as following:
@article{sui2020joint,
title={Joint Entity and Relation Extraction with Set Prediction Networks},
author={Sui, Dianbo and Chen, Yubo and Liu, Kang and Zhao, Jun and Zeng, Xiangrong and Liu, Shengping},
journal={arXiv preprint arXiv:2011.01675},
year={2020}
}
Model Training
Requirement:
Python: 3.7
PyTorch: >= 1.5.0
Transformers: 2.6.0
NYT Partial Match
- Note That: Replacing BERT_DIR in the command line with the actual directory of BERT-base-cased in your machine!!!
- 注意：需将命令行中的BERT_DIR替换为你机器中实际存储BERT的目录！！！
python -m main --bert_directory BERT_DIR --num_generated_triples 15 --na_rel_coef 1 --max_grad_norm 1 --max_epoch 100 --max_span_length 10NYT Exact Match
python -m main --bert_directory BERT_DIR --num_generated_triples 15 --max_grad_norm 2.5 --na_rel_coef 0.25 --max_epoch 100 --max_span_length 10or
python -m main --bert_directory BERT_DIR --num_generated_triples 15 --max_grad_norm 1 --na_rel_coef 0.5 --max_epoch 100 --max_span_length 10WebNLG Partial Match
python -m main --bert_directory BERT_DIR --batch_size 4 --num_generated_triples 10 --na_rel_coef 0.25 --max_grad_norm 20 --max_epoch 100 --encoder_lr 0.00002 --decoder_lr 0.00005 --num_decoder_layers 4 --max_span_length 10 --weight_decay 0.000001 --lr_decay 0.02Trained Model Parameters
Model parameters can be download in Baidu Pan (key: SetP)
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent