Open xiki-tempula opened 1 year ago
Sure, that should be trivial to implement. Do you or @saratk1 have time to open a PR adressing this?
I think this lies in simulation.context.setParameter("lambda_interpolate", lambda_)
so I'm not sure of how to best address this.
Currently when lambda=0 is used, the ML part is still being computed and it is just that it have zero weight. I wonder if it is possible that when lambda=0 is used, the openMM will be run without going through openmm-ml?