search

pranz24 / pytorch-soft-actor-critic

PyTorch implementation of soft actor critic
MIT License
810 stars 180 forks source link

multiplying action_scale in the log_prob computation #33

Closed sungsulim closed 4 years ago

sungsulim commented 4 years ago

Could anyone help me understand why the action_scale is multiplied here? Thanks in advance!

https://github.com/pranz24/pytorch-soft-actor-critic/blob/1bd1158116edb02b09a8f8f361640eaeee84447e/model.py#L103

sungsulim commented 4 years ago

Sorry there was already an issue explaining this. https://github.com/pranz24/pytorch-soft-actor-critic/issues/14#issuecomment-534974752