Skip to content
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

Question: How are experiment results accessed when doing the paper writeup? #147

Open
neil-vqa opened this issue Nov 7, 2024 · 3 comments

Comments

@neil-vqa
Copy link

neil-vqa commented Nov 7, 2024

Thank you for your contribution to the LLM open source community.

I've been going through the code for several hours now, so from what I understand, the experiment.py file will save the experiment outputs in a final_info.json file. So, I've been looking for how this final_info.json file will be accessed in the perform_writeup function, but I can't seem to see where it is, so that I can access the experimental results. Please can you point out or enlighten me how this works? How can I access the experimental results when performing the writeup? My goal is to tweak the writeup prompt but I need the experiment results. Thank you!

@conglu1997
Copy link
Collaborator

Via the notes.txt file :) Also via the aider context.

@ibrahimptester
Copy link

ibrahimptester commented Dec 30, 2024

@conglu1997 @neil-vqa hello I'm trying to generate paper on a new topic but I'm confused about which files should I edit. can you help me out please? I know the custom templates says to edit the prompt.json and seed.json but is it enough to just change those two or we need to change the whole template, experiment.py altogether?

@ibrahimptester
Copy link

I've done
install miniconda
setup environment
git clone repository
pip install requirements.txt
export open_api key
python prepare scripts
now after these things should I edit the prompt.json then run the baseline and then launch experiment.py

please clear the confusion for me.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants