ikostrikov / pytorch-a2c-ppo-acktr-gail

PyTorch implementation of Advantage Actor Critic (A2C), Proximal Policy Optimization (PPO), Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation (ACKTR) and Generative Adversarial Imitation Learning (GAIL).
MIT License
3.57k stars 829 forks source link

enjoy.py failes. Unexpected argument 'ret' #263

Closed jakefoster954 closed 3 years ago

jakefoster954 commented 3 years ago

This was run on the master branch. Help appreciated.

Traceback (most recent call last): File "enjoy.py", line 46, in allow_early_resets=False) File "..../pytorch-a2c-ppo-acktr-gail/a2c_ppo_acktr/envs.py", line 99, in make_vec_envs envs = VecNormalize(envs, ret=False) File "..../pytorch-a2c-ppo-acktr-gail/a2c_ppo_acktr/envs.py", line 191, in init super(VecNormalize, self).init(*args, **kwargs) TypeError: init() got an unexpected keyword argument 'ret'

ikostrikov commented 3 years ago

Fixed in https://github.com/ikostrikov/pytorch-a2c-ppo-acktr-gail/commit/b03eba0158b59d0b2412e2844a68b9115508066a

Let me know if it fixes the issue.

jakefoster954 commented 3 years ago

Yes, that fix worked.