Codes and Dataset for AAAI2019 paper ‘‘Exploring Answer Stance Detection with Recurrent Conditional Attention’’. (pdf)
The Chinese CommunityQA dataset has been processed into the /data/processed/, '3k' refers to the test set and '10k' refers to the training set.
Run the vectorization code before training.
python3 vectorize.py
cd to the nnet/ and run the following cmd:
python3 main_batch.py --nhops=3
cd to the nnet/ and run the following cmd:
python3 main_batch.py --is_test --nhops=3
- python 3.6
- pytorch 0.4.1
- numpy 1.15
- pyltp 0.2.1