Workflow
Policy and Value Networks
Trainers
Concepts and Structure
Environment Customization
Best Practices and Tutorials
Logging
Scaling the Training Process
maze.core.wrappers.reward_scaling_wrapper.
RewardScalingWrapper
Scales original step reward by a multiplicative scaling factor.
env – The underlying environment.
scale – Multiplicative reward scaling factor.
reward
Scales the original reward.
reward – The original reward.
The scaled reward.