jiangzhs/OpenCompass
1
0
Fork 0
mirror of https://github.com/open-compass/opencompass.git synced 2025-05-30 16:03:24 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Fix] Fix acc of IFEval (#849)

Browse Source 
* [Feature] Add IFEval

* [Fix] Changing the Score Rule.
This commit is contained in:
Jingming 2024-01-27 22:27:07 +08:00 committed by GitHub
parent 35aace776a
commit 2801883351
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
configs/datasets/IFEval/IFEval.md
View File

@ -36,10 +36,10 @@ Hark! Hearken to the tale of thy journey to the land of the rising sun, Japan. \
## Evaluation results ## Evaluation results
``` ```
dataset version metric mode internlm2-chat-20b-hf dataset version metric mode qwen-72b-chat-hf mistral-7b-instruct-v0.2-hf mixtral-8x7b-instruct-v0.1 chatglm3-6b-hf
--------- --------- ---------- ------ ----------------------- --------- --------- ---------- ------ ------------------ ----------------------------- ---------------------------- ----------------
IFEval 3321a3 strict_acc gen 0.33 IFEval 27a9cc strict_acc gen 43.62 49.17 48.98 29.76
IFEval 3321a3 loose_acc gen 0.4 IFEval 27a9cc loose_acc gen 45.47 53.97 54.71 32.16
``` ```
## Reference ## Reference

2
opencompass/datasets/IFEval/ifeval.py
View File

@ -50,4 +50,4 @@ class IFEvaluator(BaseEvaluator):
loose = sum( loose = sum(
[result['loose'].follow_all_instructions [result['loose'].follow_all_instructions
for result in results]) / len(results) for result in results]) / len(results)
return dict(strict_acc=strict, loose_acc=loose) return dict(strict_acc=strict * 100, loose_acc=loose * 100)