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

Add EScAIP Model #191

Open
wants to merge 9 commits into
base: main
Choose a base branch
from
Open

Add EScAIP Model #191

wants to merge 9 commits into from

Conversation

EricZQu
Copy link

@EricZQu EricZQu commented Jan 15, 2025

Hey maintainers,

Thanks for all the awesome work on this benchmark! I’m submitting our new model with this PR. Let me know if anything else is needed.

@janosh
Copy link
Owner

janosh commented Jan 15, 2025

thanks a lot @EricZQu for submitting EScAIP! could you also share

  • the test script you used to generate the energy predictions and relaxed structures
  • the EScAIP-relaxed WBM structures
  • which model variant was used to generate these predictions (number of trainable parameters, any other important hyperparams worth sharing?)

@janosh janosh added the new model Model submission label Jan 15, 2025
requirements:
fairchem-core: 1.2.1

openness: OSCD
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

labelling the model OSCD [Open source, closed data] is incompatible with it just being trained on MPtrj below

Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

good catch, this was a copy paste error on my part (i remember copying this from MACE-MPA-0). will add a test to catch this in future. probably each dataset in training-sets.yml should have an openness key as well

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

Successfully merging this pull request may close these issues.

3 participants