Tian Xie

@yuyouyu32 The definition of cif files can be found in https://en.wikipedia.org/wiki/Crystallographic_Information_File This function from pymatgen can convert the cif string to a `Structure` object: https://pymatgen.org/pymatgen.core.structure.html?highlight=structure#pymatgen.core.structure.IStructure.from_str From the object you can...

Thanks for your contribution. I think to solve this problem, you need to use an older version of torchmetrics. Try: ``` conda install torchmetrics=0.7.3 ```

Great question. Yes, it loads the test materials from `eval_recon.py`. The test materials are used to compute some metrics like coverage. The reconstructed materials are not used. This dependency is...

I think this is due to training instability, which we also see during our experiments. My experience is that you can restart the training by loading the latest checkpoint. The...

Thanks Luis! This is a strange error. ```File "scripts/evaluate.py", line 69, in reconstructon input_data_list = input_data_list + batch.to_data_list() ``` This line simply reads test data and save them into a...

Thanks for sharing the additional information! I was able to reproduce the error and looked little into it. I now suspect this is because pytorch geometric had some [breaking changes](https://github.com/pyg-team/pytorch_geometric/releases/tag/2.0.0)...

This is an interesting observation. Are you training your model to predict the total energy per atom? What is your overall MAE? It looks as if the errors for both...

Why do you think that the model cannot predict the difference between graphene and diamond? The prediction values are different. (-1.7248, -1.7681) The problem seems to be more related to...

Another potential approach is to copy the code in `main.py` into a Jupyter notebook. This would allow to modify the code in the notebook as well.

Thanks for reaching out. Can you provide more details? What is your train/test data? What are other hyperparameters?