Open ukoethe opened 3 years ago
I updated the main routines of the program. Step debugging and error raising is now transparent (no more threading, errors are just raised unless suppressed).
The crate distribution was already homogeneous.
I added the missing call to game_events_occurred()
.
General
Environment
initialization at other than the standard starting state-> Add custom--scenario
and modifybuild_arena
possibility to pause an episode for inspection and later resume (instead of restart)-> use step debugginggame_events_occured()
also for the last step before the game ends.shorter thinking intervals or other speed-upsTFPyEnvironment
in https://github.com/tensorflow/agents, or use gym for compatibility with keras_rlitems_fast.py
,environment_fast.py
,agents_fast.py
,bomberman_adapter.py
) in https://gitlab.com/koetherminator/fml-projectUse IntEnums instead of strings to speed-up comparisons (?)Project instructions
Hardware