Open bismuthadams1 opened 2 weeks ago
Two quick things here -
I feel strongly on point 1 but @ntBre can feel free override me on point 2. Other than that I'll leave this to him to help finalize, review, and merge (or otherwise get these workers going).
@bismuthadams1 I reapplied the rocket
compute tag on #397, and it seems to be making very good progress on NRP without any of these errors. I'm going to hold off on this PR hoping that that continues to go smoothly. At the current rate it might even finish overnight.
Previously running with the psi4 ddx env, all molecules failed validation qcengine:
ValidationError: Following atoms are too close
. This was encountered while spinning up the workers for this submission https://github.com/openforcefield/qca-dataset-submission/tree/master/submissions/2024-10-11-MLPepper-RECAP-Optimized-Fragments-Add-Iodines-v1.0.Adding openff-toolkit fixed the issue, but specifically this seemed to downgrade qcfractal-compute and qcportal from 0.56 to 0.54.1