I created a custom environment with Gym and trained it with stable baseline 3 algorithms. The observation and space action are both continues. The observation space includes 10 values and action space has 2. The action space is [0,1] and I know it's not symmetric.
I want to normalize my inputs to the NN and I Don't Know how to do it. The environment observes Temperature and Power (input values) and takes action (continuous between [0,1]) and returns state (10 values not normalized) I believe I have to normalized all inputs to get better results but I don't know how can I normalize it bf feeding to neuronal network and unwrapped it afterwards.
I appreciate any help specially if you can link some dummy code or a practical case.
