fix(torch): handle 'update_after' set to zero

rickstaa / stable-learning-control

A framework for training theoretically stable (and robust) Reinforcement Learning control algorithms.

https://rickstaa.dev/stable-learning-control

MIT License

6 stars 1 forks source link

fix(torch): handle 'update_after' set to zero #408

Closed rickstaa closed 6 months ago

rickstaa commented 6 months ago

This pull request addresses a bug in the learning rate decay logic when 'update_after' is set to zero. Previously, the algorithm would malfunction under these conditions. With this fix, the algorithm can now correctly handle 'update_after' being set to zero, ensuring proper learning rate decay.