Mandatory to create Task for RL training

[GRINGOLOCO7]

I was wondering, seeing the examples in the GitHub, all Reinforcment Learning algorithms has the structure of:
./isaaclab.sh -p source/standalone/workflows/sb3/train.py --task Isaac-Cartpole-v0 --headless --cpu

Without having the template avaible at this GitHub that is inexisting,

I WAS WONDERING, IF I NEED TO CREATE A TASK IN ORDER TO MAKE A REINFORCMENT LEARNING ALGORITM.
If yes, do I do this form here?

[Mayankm96]

No, if you only care about the algorithm, then you need to implement that through your preferred choice of the library. You can then run the algorithm with the bash command similar to the one above.

If you want to integrate your own library, please check here: https://isaac-sim.github.io/IsaacLab/source/how-to/add_own_library.html

[GRINGOLOCO7]

what if I want to train my own robot... This mean I won't use one of the avaible tasks (can be found with ./isaaclab.sh -p source/standalone/environments/list_envs.py).