In current trainers/tester flow there are several loops that need to be broken if a given condition is true. In particular, some conditions need to be checked few times (e.g. episode_limit in both inner and outer loops of offline trainer).
I guess this might be simplified with the right raising and handling of exceptions
In current trainers/tester flow there are several loops that need to be broken if a given condition is true. In particular, some conditions need to be checked few times (e.g. episode_limit in both inner and outer loops of offline trainer).
I guess this might be simplified with the right raising and handling of exceptions