Maxime Chevalier-Boisvert e1a6afbf81 Added code to train from one hot encoding 5 gadi atpakaļ
..
envs 97bca6e172 Fixed issues with dynamic obstacles env 5 gadi atpakaļ
__init__.py e7e870ce2d Fixed issues with wrappers 6 gadi atpakaļ
minigrid.py 9c57465a0a Implemented one-hot observation wrapper 5 gadi atpakaļ
register.py 5d9e8cab8a Fixed issue with default reward_threshold 5 gadi atpakaļ
rendering.py 6e77fbef44 Fixup 5 gadi atpakaļ
roomgrid.py 13d4651e03 Refactored to eliminate start_pos, simplify code. 5 gadi atpakaļ
wrappers.py e1a6afbf81 Added code to train from one hot encoding 5 gadi atpakaļ