Request for Reinforcement Learning (RL) Training Script

[Cavalry0531]

Thank you for releasing SRL-AssistedAFM. I have been reading the repository and found it very helpful.

I noticed that the repository mainly contains the supervised / SRL-related components. May I ask whether the reinforcement learning (RL) part of the code (e.g., training script, policy optimization, environment interface, or any RL-specific implementation) is available?

If it cannot be made fully public, could you kindly share:

the RL training entry point (main script),

key configs / hyperparameters,

or a minimal example of how the RL module is integrated with the supervised model?

Any guidance would be greatly appreciated.
Thank you very much for your time and support.