OIGE task with stable baselines3

We customized a reinforcement learning task UR5Doortask by referring to some of the tasks that come with OIGE such as Cartpoletask. it is a subclass of RLTask. By defining the yaml files we can implement the UR5Doortask as normal as the Cartpoletask, but of course the reinforcement learning algorithm is based on PPO. So we would like to try other algorithms inside stable baselines3 such as A2C, DRPG, etc… But in the official tutorials (9.8. Custom RL Example using Stable Baselines — Omniverse IsaacSim latest documentation), the custom RL task is based on the BaseTask class instead of RLTask, so we only need to pass a name parameter when creating an instance of the task in the training script. However, if the task is based on RLTask, you need to pass a name, a sim_config and an env. How can I solve this problem.

If I need to revert the originally written UR5Doortask task based on RLTask to BaseTask based, how can I clone a lot of instances to train in set_up_scene. Because in the official tutorial about cloning it just clones the objects, i don’t know how to make them to train together