Repo for offline reinforcement learning methods
Current support dataset:
- D4RL(website)
Planning to support:
- DQN Replay Dataset(website)
Behavior Clone methods:
- Behavior cloning
Basic Off-policy methods
- DDPG
- SAC
- Basic SAC
- Twin SAC without V
Offline RL methods:
- BCQ
- Continuous BCQ