Michal Krompiec

I think this is a Python 2->3 bug. I changed line 309 in Periodic.py: m = np.array(map(lambda x: ATOMICMASSES[x-1], self.mol0.atoms))*1000.0 to: m = np.array(list(map(lambda x: ATOMICMASSES[x-1], self.mol0.atoms)))*1000.0 and it seems...

Thanks a lot for pushing the updated code and for the references.

On my test case, the new code is 2x faster (when using the same model, i.e. model 3).

Isn’t a 2-body kernel supposed to give good performance only for some special cases?

You can try restarting from a different initial guess of hyperparameters. But the most likely reasons are that you don’t have enough training data, or the system can’t be adequately...

Thanks @leeping ! Do you have any tips for speeding up torsion fitting? We are fitting all rotatable torsions in polyaromatic molecules, 2-5 torsions each.

Is it related to https://github.com/choderalab/perses-barnase-barstar-paper/blob/main/scripts/05_analyze/run_make_traj_per_replica.py and https://github.com/choderalab/openmmtools/issues/678 ?

Thank you!

Solution 1 would be ideal :)

I just hacked, refactored and wrapped portions of GammCor. All I’ve done is a half-automatic translation from F77 to F90, pruning unused pieces of code and making it work with...