Skip to content

sbsrl_offline - sooper compatibility - #248

Merged
yardenas merged 6 commits into
mainfrom
sooper_compatibility
Oct 31, 2025
Merged

sbsrl_offline - sooper compatibility#248
yardenas merged 6 commits into
mainfrom
sooper_compatibility

Conversation

@lucavignola

Copy link
Copy Markdown
Collaborator

No description provided.

Comment thread ss2r/algorithms/sbsrl/train.py Outdated
backup_qc_params = sbsrl_network.backup_qc_network.init(key_qr)
backup_qr_params = sbsrl_network.backup_qc_network.init(
key_qr
) # TODO: can I use the same network for backup_qc/backup_qr?

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Technically yes, but let's separate them to avoid inconsistencies

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ok it should be ok now

Comment thread ss2r/algorithms/sbsrl/train.py Outdated
Comment thread ss2r/algorithms/sbsrl/train.py Outdated
backup_policy_params=params[1],
backup_policy_params=params[
1
], # TODO: can remove backup_policy_params?

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why remove?

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Where are we using them? it seems to me we never touch them in sbsrl

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah, I see. But if you train a prior policy for sooper you'll need them no?

@yardenas
yardenas merged commit fa3fcdd into main Oct 31, 2025
1 check passed
@yardenas
yardenas deleted the sooper_compatibility branch October 31, 2025 17:25
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants