search

denisyarats / drq

DrQ: Data regularized Q
https://sites.google.com/view/data-regularized-q
MIT License
407 stars 52 forks source link

Action_repeat settings to reproduce your paper results #5

Closed teslacool closed 4 years ago

teslacool commented 4 years ago

In Readme, you mean that if I want to reproduce the results, I just python train.py env=cartpole_swingup batch_size=512

But I notice the action_repeat number in config.yaml is not 8 for cartpole_swingup.

Maybe you should check this point.

denisyarats commented 4 years ago

Yes, action_repeat needs to be set according to the paper. See table 2 in the appendix.

I also update the README. Thanks!

teslacool commented 4 years ago

Thanks, this solved my doubt.