v0.1.0

• add selfplay with self
• add selfplay strategy

v0.0.15

• add sac
• fix offpolicy bugs

v0.0.14

• can train tictactoe
• add isaac example

v0.0.13

add GAIL
add more callbacks and callbacks tests
add behavior cloning algorithm

v0.0.12

• optimize vdn algorithm
• add Atari
• Successfully train dqn in gridworld
• add StarCraft II: SMAC
• can save checkpoints with callbacks
• Add connect3
• can evaluate and save best model with callbacks

v0.0.11

• add MAT algorithm
• add mujoco

v0.0.10

• Add retro
• Update runners for off-policy algorithm
• Update module, buffer, and driver for off-policy algorithm
• Add SuperMario Environment
• Fix bug in test_case: https://openrl-docs.readthedocs.io/en/latest/quick_start/multi_agent_RL.html#loading-trained-agent

v0.0.9

• add .vscode to gitignore
• add colab
• fix: PPONet parse_args for colab
• fix: python version check bugs
• abstract rl agent

v0.0.8

• improve code testing coverage
• add test for train examples
• add docker image
• fix minor bugs