How to use GPU for trainings with ML Agents #6055
Description
Just finished installing everything to get the ML Agents running, using Anaconda with Python 3.10.12 and PyTorch 2.2.0 and everything seems to be working, except the PyTorch seems to be using cpu for the trainings.
As you can see i ran mlagents-learn --torch-device cuda and it didn't seem to work for me.