-
Notifications
You must be signed in to change notification settings - Fork 3
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
ocpmodels unavailable? #19
Comments
Hello, you are correct that the repository has been renamed to FAIR-chem. The |
Hello, |
After some exploration I realized ocpmodels should be replaced with fairchem.core in the code. Is that right? Also, I am not able to find the base model from the base.py. Should it be renamed as well? Traceback (most recent call last): |
You can probably get rid of the BaseModel import in gemnet_charge.py. I don't think it's actually used in the file. If you manage to get the repository working with the current fairchem repository feel free to submit a PR! |
Thanks! Yes, that's what I did. However there are several other attributes that the code is importing but fairchem does not have them. such as: _Jd = torch.load(os.path.join(os.path.dirname(file), "Jd.pt")) I think I can not get rid of OCPDataParallel as this is being used in the code. Do you know how to fix it? Thanks |
I guess torch's DistributedDataParallel might work, following the changes that were made here: https://github.com/FAIR-Chem/fairchem/blob/e11e78e87f791241b3759a7218a1b0e307a49662/src/fairchem/core/trainers/base_trainer.py#L578 |
In the README file for this repository, it points to FAIR-Chem for the installation of ocpmodels. However, that package does not seem to exist in the FAIR-Chem repository. Is this deprecated, and if so, is there a replacement for ocpmodels that could be used to run the charge density prediction models?
The text was updated successfully, but these errors were encountered: