OpenCompass/opencompass/datasets/calm/evaluation/accuracy/prob.py
Peng Bo edd0ffdf70
Calm dataset (#1287)
* add calm dataset

* modify config max_out_len

* update README

* Modify README

* update README

* update README

* update README

* update README

* update README

* add summarizer and modify readme

* delete summarizer config comment

* update summarizer

* modify same response to all questions

* update README
2024-07-26 11:48:16 +08:00

10 lines
373 B
Python

def compute_acc(gt_list, pred_list):
correct_num = 0
for pred, gold in zip(pred_list, gt_list):
kept_pred = round(pred, 4) if pred != None else pred
kept_gold = round(gold, 4)
if kept_pred == kept_gold:
correct_num += 1
acc = correct_num / len(gt_list)
return acc