adaptgym

Environments for assessing adaptation and exploration of reinforcement learning agents.

Many are inspired by animal assays, e.g. a virtual labyrinth (based on Rosenberg et al.), and a virtual object interaction assay (based on Ahmadlou et al.).

Adaptgym is based on dm_control, and is a flexible framework for making environments.

Installation

Install from pypi: pip install adaptgym

Install from test pypi: pip install -i https://test.pypi.org/simple/ adaptgym

or to specify a specific version:

pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple --no-cache-dir adaptgym==0.1.48

Playing around with the environments

See adaptgym/envs/scripts/fiddle_env.py to visualize and interact with the environments.

To run this demo from the command line:

python -c 'from adaptgym.fiddle_env import main; main()'

Notes:

Local install (from the adaptgym/ directory) poetry lock --no-update poetry install

To update and publish. Change the version number in pyproject.toml, then: poetry build poetry publish -r testpypi