The Arcade Learning Environment (ALE) -- a platform for AI research.
copied from cf-post-staging / ale-pyThe Arcade Learning Environment (ALE) is a simple framework that allows researchers and hobbyists to develop AI agents for Atari 2600 games.
A difference between the PyPI distribution of ale-py and the conda-forge package
is that the conda-forge package was unable to verify the copyright status of the ROM
required to use the library, and if conda-forge is allowed to redistribute them.
For these reason, to actually work with the library, if you are legally allowed to
access the ROMs, download them (for example with the https://github.com/Farama-Foundation/Arcade-Learning-Environment/blob/master/scripts/downloadunpackroms.sh
script) and then set the ALE_ROMS_DIR
environment variable to point to the directory that contains the ROMs.