NM512 / dreamerv3-torch

Implementation of Dreamer v3 in pytorch.
MIT License
425 stars 96 forks source link

Set agent to eval mode in simulate function #33

Closed zoharri closed 1 year ago

zoharri commented 1 year ago

Hi! thank you for the great work! In line 165 in tools.py (the simulate function), you call the agent as follows: action, agent_state = agent(obs, done, agent_state)

I think this should be: action, agent_state = agent(obs, done, agent_state, not is_eval) right?

Thanks!

NM512 commented 1 year ago

Hi,

Thank you for your comment. I believe the policy is set to evaluation mode by the following line of code before calling the simulate function.https://github.com/NM512/dreamerv3-torch/blob/d94a719421401baab3a38696ab23f15b9c98b4ae/dreamer.py#L330