search

reinforcement-learning-kr / lets-do-irl

Inverse RL algorithms (APP, MaxEnt, GAIL, VAIL)
MIT License
672 stars 109 forks source link

running test.py for VAIL #7

Open ranahanocka opened 4 years ago

ranahanocka commented 4 years ago

Hello,

I am running the VAIL example as explained in the README.

main.py works as expected.

When running test (e.g., as written:python test.py --load_model ckpt_4000_vail.pth.tar), there is an import error.

Specifically: https://github.com/reinforcement-learning-kr/lets-do-irl/blob/ad04bf1856deb7ac2acda46275eb0c927c635ae9/mujoco/vail/test.py#L8

there is no file running_state. In the main.py the running state was defined from Zfilter.

csufangyu commented 4 years ago

Hello,

I am running the VAIL example as explained in the README.

main.py works as expected.

When running test (e.g., as written:python test.py --load_model ckpt_4000_vail.pth.tar), there is an import error.

Specifically: https://github.com/reinforcement-learning-kr/lets-do-irl/blob/ad04bf1856deb7ac2acda46275eb0c927c635ae9/mujoco/vail/test.py#L8

there is no file running_state. In the main.py the running state was defined from Zfilter.

maybe you should change it.the Zfliter is in utils.zfilter.you can change it as : from utils.zfilter import ZFilter

ranahanocka commented 4 years ago

Thanks @csufangyu. I previously modified the code to make it work, but thought the authors might want to fix the repo