nrontsis / PILCO

Bayesian Reinforcement Learning in Tensorflow
MIT License
311 stars 84 forks source link

Could you please share exact version of some dependency packages #58

Open gitAugust opened 3 years ago

gitAugust commented 3 years ago

Hello,

I have installed dependency packages required in requirement.txt, but code can't run well under my environment.

when I run the example safe_swimmer_run.py, I meet the errorOperatorNotAllowedInGraphError: iterating overtf.Tensoris not allowed: AutoGraph did convert this function. This might indicate you are trying to use an unsupported feature. I guess that may be caused by wrong version of some packages.

my version of some main package is tensorflow 2.4.1 gpflow 2.1.4 gym 0.18.0 mujoco-py 1.50.1.0 numpy 1.19.2

and my python is 3.7

thanks in advance!