A third unrelated (and lowest priority) bug in 0050.perovskitedata.csv: SMILES and InChI representations for ECL reactions are not handled correctly. (they are output as zeros)
[This actually goes back to an unresolved issue earlier in this channel from 19 Nov 2019]
It looks bad if we are releasing datasets with type errors in the columns. It would be best if we actually fix these (the data is available). It would satisfice if we just exclude ECL data from exported data products (given that the amount of data is small, has uncertaintainties about performance, and won't be generated in the future).
Perhaps this one gets triaged as a summer undergrad project?
A third unrelated (and lowest priority) bug in 0050.perovskitedata.csv: SMILES and InChI representations for ECL reactions are not handled correctly. (they are output as zeros) [This actually goes back to an unresolved issue earlier in this channel from 19 Nov 2019] It looks bad if we are releasing datasets with type errors in the columns. It would be best if we actually fix these (the data is available). It would satisfice if we just exclude ECL data from exported data products (given that the amount of data is small, has uncertaintainties about performance, and won't be generated in the future). Perhaps this one gets triaged as a summer undergrad project?