Andrew S. Rosen
Andrew S. Rosen
Yeah I don't know too much about how `torch` is built myself. We just happened to stumble upon this in the `Pymatgen` repo because both dependencies happened to be installed...
Tagging @samblau.
Looks like this is somewhat of a "me" problem but also a `periodictable` issue that can be resolved. See https://github.com/pypa/packaging-problems/issues/573. By default, the Python environment does not have `setuptools` yet...
> Currently, the python packages for DFT-D3 and [DFT-D4](https://github.com/dftd4/dftd4) dispersion models from the Grimme group are not compatible with torch, so the conventional method of adding the dispersion does not...
Makes sense! Definitely in the meantime, `torch-dftd` seems like the way to go here.
That's wonderful! Thanks for the update!
It looks like this issue can be closed or migrated, as it does not seem to be a Custodian issue.
The bug is in the following line: https://github.com/cclib/cclib/blob/ee91131e9c278081d1ff88fa763d3c2b5e501da6/cclib/parser/xtbparser.py#L437 ```python a="Br 1.750781 0.001502 0.030011" a.split()[0].isupper() >>> False ```
@rkingsbury: FYI that monty is updated, although I don't know if it's related to your issue or not.
That's understandable. Thanks for the reply! Feel free to close this issue if you would like.