Failing featurizations.

[kjappelbaum]

Failed to featurize structures_qmof/qmof-5117bd4.cif because of You need at least one site to construct a <class 'pymatgen.core.structure.Structure'>
Failed to featurize structures_qmof/qmof-51d44f0.cif because of too many indices for array: array is 1-dimensional, but 2 were indexed
Failed to featurize structures_qmof/qmof-60d01a8.cif because of too many indices for array: array is 1-dimensional, but 2 were 
Failed to featurize structures_qmof/qmof-5117bd4.cif because of You need at least one site to construct a <class 'pymatgen.core.structure.Structure'>
Failed to featurize structures_qmof/qmof-51d44f0.cif because of too many indices for array: array is 1-dimensional, but 2 were indexed

[kjappelbaum]

for qmof-5117bd4.cif the RACs fail. The failure is already in the naive fragmentation (get_bb_from_indices). One reason for this is that there are already floating components in the structure graph we get (perhaps due to the cutoff s -> indeed, we currently use Ru-N of 2.2626 A, whereas the distance in the structure is about 2.5).

This one will be fixed with a new release of structuregraph_helpers. However, since we change the graph definition, it will require re-computing all features and hashes...

[kjappelbaum]

qmof-51d44f0.cif failes in PHHist. Was due to the fact that some persistence diagrams were empty. Seems now also fixed for qmof-51d44f0.cif

[kjappelbaum]

from what I recall, all of this has been fixed.