qipeng
commented
5 years ago The evaluation script in the repository contains all evaluation metrics we use. See: https://github.com/hotpotqa/hotpot/blob/master/hotpot_evaluate_v1.py
Closed tomtang110 closed 5 years ago
qipeng
commented
5 years ago The evaluation script in the repository contains all evaluation metrics we use. See: https://github.com/hotpotqa/hotpot/blob/master/hotpot_evaluate_v1.py
Hi
I found you only provide the evaluating method for answer in your code. Could you provide the evaluating code for support facts? or joint evaluating method between ans and sup in order to make experiment better.
Kind regards Tang