After running the training for millions of timesteps, this error occasionally occurs (roughly in about one fifth of the runs).
I would like to know whether anyone has encountered the same issue and how it can be resolved.
AssertionError: Agent 5 cannot perform action 14
After running the training for millions of timesteps, this error occasionally occurs (roughly in about one fifth of the runs).
I would like to know whether anyone has encountered the same issue and how it can be resolved.
AssertionError: Agent 5 cannot perform action 14