Ildefons Magrans de Abril
Ildefons Magrans de Abril
Error message: --------------------------------------------------------------------------- RuntimeError Traceback (most recent call last) in 1 for x in xs: 2 print(1) ----> 3 hs, cs = rim_model(x, hs, cs) ~\anaconda3\envs\eg2\lib\site-packages\torch\nn\modules\module.py in __call__(self, *input, **kwargs)...
Changing botocore==1.9 for botocore==1.11.9 in environment.yml seems to solved the issue. Is this a correct fix?
Is there support for other reward functions besides return?
I am curious, what is the motivation of using MARL?
The last stable version of stable3_baseline (2.0.0a9) is buggy. Just install the previous working version: `pip install stable-baselines3==2.0.0a5`
> > The last stable version of stable3_baseline (2.0.0a9) is buggy. Just install the previous working version: `pip install stable-baselines3==2.0.0a5` > > Would you like to submit a PR to...
> > The last stable version of stable3_baseline (2.0.0a9) is buggy. Just install the previous working version: `pip install stable-baselines3==2.0.0a5` > > This doesn't work either. Slightly different error message...
Start a new setup from scratch: 1) create a new conda environment 2) activate conda environment 3) pip install FinRL project 4) pip install stable-baselines3==2.0.0a5 > > pip list >...