Hi, I was just wondering if it was possible to access the policy gradients in between training epochs and how this could be done? Thanks
Hi there,
You could look into the rl_games code (GitHub - Denys88/rl_games: RL implementations), which is the RL library we used for our IsaacGymEnvs examples with. It could possibly be under rl_games/algos_torch/a2c_continuous.py
.