franknoe
commented
7 years ago You mean generate_trajectories? OK, but you can call generate_trajectory multiple times and that would probably not be a performance issue.
Am 30/11/16 um 21:54 schrieb Tara Athan:
In the generation api there is a method for generating M trajectories
trajectories(self, M, N, start=None, stop=None):
As now implemented, start is only allowed to be None or int, so it is only possible to specify the same starting state for multiple trajectories. This causes problems when we try to use this for non-stationary MM because we want to continue the generation with a new transition matrix, but with the (different) states obtained from the previous transition matrix. It doesn't look like the performance would be affected by allowing different states, since each trajectory is generated independently.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/markovmodel/msmtools/issues/95, or mute the thread https://github.com/notifications/unsubscribe-auth/AGMeQp44DvMv-TuGNpR3TXGWQnlfnjzpks5rDeKDgaJpZM4LAq65.
--
Prof. Dr. Frank Noe Head of Computational Molecular Biology group Freie Universitaet Berlin
In the generation api there is a method for generating M trajectories
trajectories(self, M, N, start=None, stop=None):
As now implemented, start is only allowed to be None or int, so it is only possible to specify the same starting state for multiple trajectories. This causes problems when we try to use this for non-stationary MM because we want to continue the generation with a new transition matrix, but with the (different) states obtained from the previous transition matrix. It doesn't look like the performance would be affected by allowing different states, since each trajectory is generated independently.