Applying DQN or any DRL algorithm to kuka_bin.py or franka_cube_ik.py

I would be happy if anybody worked on similar or same topic and may answer the following questions.

1-) How can I apply a DQN algorithm to the Kuka Robot (kuka_bin.py) or franka_cube_ik.py?
2-) How can I manipulate my agent in the Isaac Gym environment?

Thanks in advance.

Best,
Faruk