0.3.5

• Minor improvements to docs, PyPI metadata.
• Fix a versioning conflict for TensorFlow/TensorFlow probability.
• Make bandit environments more configurable.

0.3.4

Bump version again (incorrect tag was used before) and use PyPI's rlax.

PiperOrigin-RevId: 334553236
Change-Id: I88eca16269fabf975dd0c8f7ccd8a8c2374b78c9

0.3.3 -- Minor bugfixes

Internal change.

PiperOrigin-RevId: 334334765
Change-Id: I1080d6b302fb77b3d75633c4d2ed487e6b26730d

0.3.2

Fixes a version discrepancy between PyPI and GitHub.

0.3.1 -- Minor bugfixes and improved test coverage

• Fixed a bug where some experiments were assigned to the wrong tag (736b9fe).
• Add GitHub CI (pytype + pytest) and improve agent tests.
• Some minor quality-of-life changes for getting the bsuite_num_episodes constant.

0.3.0 -- Cartpole bugfix and more

• Fixed an issue with cartpole_swingup's best_episode not being computed correctly due to negative rewards (beb1630). This change will affect scores for this experiment, so we bump the minor version in keeping with semantic versioning.
• Fix versioning issues in colab and setup.py.
• Improve test.sh coverage.
• Make TAGS accessible as sweep (e.g. sweep.TAGS['exploration'] to retrieve all exploration experiments.
• Add a bootstrapped DQN baseline in JAX.
• Extract environments to their own package for easier direct access for debugging/development.

0.2.0 -- Miscellaneous bugfixes and improvements

Contents include:

• Drop Python 2 support.
• Improve installation instructions.
• Upgrade baseline agents to TensorFlow 2 / JAX.
• Improvements to logging behaviour.