An example of running an FR-PPO-based algorithm against the Farama Gymnasium Car Racing environemnt.
Significant learning effort is about 12h on Tesla V100 GPU with 32GB memory for the NN model and 40 CPUs running the environments.