openai / vime

Code for the paper "Curiosity-driven Exploration in Deep Reinforcement Learning via Bayesian Neural Networks"
https://arxiv.org/abs/1605.09674
342 stars 90 forks source link

rllab version problem #8

Closed hcwang95 closed 5 years ago

hcwang95 commented 5 years ago

It seems that the new rllab is in Python3 but the code for vime is in Python2.

I am wondering if any of you successfully reproduce the results and could you tell me what version of rllab are you using?

Thanks a lot.

hcwang95 commented 5 years ago

The new version of rllab can be used by only change all xrange() in implementation in Vime to range().

BarelySentient commented 5 years ago

Sorry for missing this, I was able to reproduce the results for the Swimmer Gather domain. I switched to python2. The key was to download the py2 branch of rllab rather than the whole rllab github repo.

hcwang95 commented 5 years ago

Oh, thanks a lot!