Yifei Cheng 43792bf5ab

Upgrade gym (#613) ...

fixes some deprecation warnings due to new changes in gym version 0.23:
- use `env.np_random.integers` instead of `env.np_random.randint`
- support `seed` and `return_info` arguments for reset (addresses https://github.com/thu-ml/tianshou/issues/605)

2022-06-28 06:52:21 +08:00

..

atari

Upgrade gym (#613)

2022-06-28 06:52:21 +08:00

box2d

Add BranchingDQN for large discrete action spaces (#618)

2022-05-15 21:40:32 +08:00

inverse

rename save_fn to save_best_fn to avoid ambiguity (#575)

2022-03-22 04:29:27 +08:00

modelbased

Fix SAC loss explode (#333)

2021-04-04 17:33:35 +08:00

mujoco

Fix num_envs=test_num (#653)

2022-05-30 12:38:47 +08:00

offline

Implement TD3+BC for offline RL (#660)

2022-06-07 00:39:37 +08:00

vizdoom

use envpool in vizdoom example, update doc (#634)

2022-05-09 00:42:16 +08:00

__init__.py

Fix critic network for Discrete CRR (#485)

2021-11-28 23:10:28 +08:00