Skip to content

Latest commit

 

History

History
15 lines (8 loc) · 426 Bytes

README.md

File metadata and controls

15 lines (8 loc) · 426 Bytes
Raw

GAHs

Guided Attention Heads

steps

  1. Put your datset in the folder datasets/your_DataSet_name/your_data_file, then update the load in preprocessor/raw_data_loader.py

  2. Preprocess your data (semantic annotation): python main.py run_mode preprocess --dataset TREC --splits train,test

-- the processed data will be in datasets/your_DataSet_name/** -- splits means how man files you have

  1. Train the model: train.py