How can I get timesteps_total from the environment ?
I need to increase the episode length after a number of time_steps without having to stop and restore the training every time
- Medium: It contributes to significant difficulty to complete my task, but I can work around it.