Can I modify MDP functions in Isaac Lab without restarting Isaac Sim? Any tips for faster iteration?

[leo01110111]

I’m using Isaac Lab to train a robot with reinforcement learning in Isaac Sim, and my RL environment is based on the manager-based MDP structure (reward managers, observation managers, etc.).

Right now, every time I make a change to my Python code for the MDP functions (e.g., reward calculation, observation extraction), I stop my training script and re-run it — but I’m not sure if I also need to restart Isaac Sim, or if there’s a way to reload those Python changes while keeping the simulator running.

My questions are:

Is it possible to update MDP-related code (reward/observation/action managers) without restarting Isaac Sim?

Are there any recommended tips for faster iteration when testing code changes in Isaac Lab?

Thanks!