Python 3.8
Pytorch 1.9.0
CUDA 10.1
BERT - our BERT model is adapted from this implementation:
https://github.com/huggingface/pytorch-pretrained-BERT

• divide_fold.py: used to get 20 files, which will be named as “foldx_train.txt” and “foldx_test.txt”, where “x” should be from 1 to 10.

data_combine_ECPE - A dir where contains data splits for ECPE task. The test dataset are named as fold*_test.txt, while the train datasets are named as fold*_train.txt.

data_combine_ECE - A dir where contains data splits for ECE task. The test dataset are named as fold*_test.txt, while the train datasets are named as fold*_train.txt.

data_combine_CCRC - A dir where contains data splits for CCRC task. The test dataset are named as fold*_test.txt, while the train datasets are named as fold*_train.txt.

• preprocess.py: used to get the manually labeled datase.

• gen_nega_samples.py: used to generate the constructed conditional-ECPE dataset.

data_combine_ECE_balance - A dir where contains data splits for de-bias dataset for ECE task. The test dataset are named as fold*_test.txt, while the train datasets are named as fold*_train.txt.

data_combine_ECPE_balance - A dir where contains data splits for de-bias dataset for ECPE task. The test dataset are named as fold*_test.txt, while the train datasets are named as fold*_train.txt.

Download checkpoint from https://www.dropbox.com/sh/45jj8dcenhbuzvn/AABbXSxccgyi1AMGA5yi4DBUa?dl=0 and save in the fold checkpoint

Download pretraind model from https://huggingface.co/bert-base-chinese and save it as bert-base-chinese.

• run ECE.py for ECE task.

• run ECPE.py for ECPE task.

• run CCRC.py for CCRC task.

• run ECPE_M2M.py for M2M variant method in ECPE task.

If you find our work useful, please consider citing UECA-Prompt:

@inproceedings{zheng2022ueca,
  title={UECA-Prompt: Universal Prompt for Emotion Cause Analysis},
  author={Zheng, Xiaopeng and Liu, Zhiyue and Zhang, Zizhen and Wang, Zhaoyang and Wang, Jiahai},
  booktitle={Proceedings of the 29th International Conference on Computational Linguistics},
  pages={7031--7041},
  year={2022}
}