New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add MATH evaluation #135
base: main
Are you sure you want to change the base?
Add MATH evaluation #135
Conversation
@@ -24,6 +24,9 @@ wget -P data/eval/tydiqa/ https://storage.googleapis.com/tydiqa/v1.1/tydiqa-gold | |||
# GSM dataset | |||
wget -P data/eval/gsm/ https://github.com/openai/grade-school-math/raw/master/grade_school_math/data/test.jsonl | |||
|
|||
# MATH dataset | |||
mkdir -p data/eval/MATH |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This mkdir
might not be necessary.
if args.no_cot: | ||
prompts = [prompt_prefix + "Question: " + example["question"].strip() + "\nAnswer:" for example in test_data] | ||
else: | ||
prompts = [prompt_prefix + "Question: " + "\n" + example["question"].strip() + "\nSolution: " + "\n" for example in test_data] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Interesting. So, for CoT, they use "Solution" in the prefix not "Answer"?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Oh I saw it above. Nvm.
Looks good. Thanks @danieljkim0118! Have you tested the performance of some vanilla pretrained models and tulu models? I am planning to run some tests. It would be great if you have some numbers that I can compare to. |
It would be good to merge this soon! |
Added evaluation scripts for the MATH (Hendrycks et al., 2021) dataset.