Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Tensorboard Logging Issue after multiple consecutive calls to self.learn(..., reset_num_timesteps=False) #1007

Closed
krishpop opened this issue Sep 19, 2020 · 1 comment · May be fixed by #1008
Closed

Tensorboard Logging Issue after multiple consecutive calls to self.learn(..., reset_num_timesteps=False) #1007

krishpop opened this issue Sep 19, 2020 · 1 comment · May be fixed by #1008

Comments

@krishpop
Copy link

Screen Shot 2020-09-19 at 1 53 06 AM

Describe the bug
step used to log self.summary when training SAC resets to 0 when self.learn(reset_num_timesteps=False) is run. This results in some tensorboard logs to become disconnected and have logging begin at step=0 again.

Additional context
This can be resolved by logging to tensorboard with self.num_timesteps instead of step passed into self._train_step.
krishpop added a commit to krishpop/stable-baselines that referenced this issue Sep 19, 2020
@krishpop
Fix step used to log SAC summary
0e79512 
fixes hill-a#1007
This was referenced Sep 19, 2020
Closed
Open
@araffin
Copy link
Collaborator

araffin commented Aug 14, 2021

Should be fixed in SB3: DLR-RM/stable-baselines3#433

@araffin araffin closed this as completed Aug 14, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
2 participants
@araffin @krishpop