added embedding checks - Githubissues

coarse-graining / cgnet

learning coarse-grained force fields

BSD 3-Clause "New" or "Revised" License

57 stars 26 forks source link

added embedding checks #130

Closed brookehus closed 5 years ago

brookehus commented 5 years ago

Development:

[x] Implement feature / fix bug
[x] Add documentation
[ ] Add tests

Checks:
[x] Run nosetests
[ ] Check pep8 compliance

MoleculeDataset now returns an error if any embeddings are given that less than and/or that are not integers. I updated the test_embedding_shape test in test_molecule_dataset appropriately.

nec4 commented 5 years ago

Great - taking a look now.

nec4 commented 5 years ago

Things look great and tests pass. LGTM!

nec4 commented 5 years ago

Ahh - I just remembered, there are a few other places in the tests that still possibly use zeros in embeddings, eg) _get_random_schnet_feature() in test_feature_combiner.py. I think these should be changed too.

brookehus commented 5 years ago

@nec4 noticed that some of our tests use embeddings starting with zero, so in 1af2784 I tried to fix them all.

nec4 commented 5 years ago

LGTM - all tests pass!