Closed ckae95 closed 5 years ago
hi @ckae95
Could you update to the latest megnet and pymatgen on github and see what happens?
These problems seem to be a version issue.
Thanks. Updating pymatgen helped. But in the crystal_example notebook I now get a new error related to pymatgen:
MPRestError Traceback (most recent call last) ~/.local/lib/python3.6/site-packages/pymatgen/ext/matproj.py in _make_request(self, sub_url, payload, method, mp_decode) 170 raise MPRestError("REST query returned with error status code {}" --> 171 .format(response.status_code)) 172
MPRestError: REST query returned with error status code 403
During handling of the above exception, another exception occurred:
MPRestError Traceback (most recent call last)
As the error message implies, you need to set your Materials Project API key to get structures from MP. Pls look at the pymatgen documentation on how to set the API key.
Thanks. But still: in the transfer_learning notebook: When I try to run
model_form = MEGNetModel.from_file('../mvl_models/mp-2019.4.1/formation_energy.hdf5')
I still get an error: TypeError: init() got an unexpected keyword argument 'atom_converter'
@ckae95 This error is caused by the older API and has been fixed (convertor to converter in all names). Again, please update to the latest megnet version and model files (the one in github, not pypi) and set the correct path for the model hdf5. It should work.
PS: Just updated the correct path for the model. It should be
model_form = MEGNetModel.from_file('../mvl_models/mp-2018.6.1/formation_energy.hdf5')
.
When I run the test_models.py script, I get a "TypeError: init() got an unexpected keyword argument 'bond_converter'", which can be fixed by changing "converter" to "convertor".
However, then I get another error further down, namely: RuntimeError: get_all_neighbors() got an unexpected keyword argument 'include_site'. The same happens by the way when I run the crystal_example notebook.
Thanks for your help.