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.

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

feat: Add support for preloading textual clip model without visual model #13368

Closed
wants to merge 5 commits into from

Conversation

1-tempest
Copy link

Preloading the textual clip model is required for quick smart searches.

I'd like to suggest having an ability to load the textual model without the dependency of the visual model to preserve system ram.

@1-tempest 1-tempest requested a review from mertalev as a code owner October 11, 2024 04:15
@github-actions github-actions bot added documentation Improvements or additions to documentation 🧠machine-learning labels Oct 11, 2024
@1-tempest 1-tempest changed the title Add support for preloading textual clip model without visual model feat:Add support for preloading textual clip model without visual model Oct 11, 2024
@1-tempest 1-tempest changed the title feat:Add support for preloading textual clip model without visual model feat: Add support for preloading textual clip model without visual model Oct 11, 2024
@bo0tzz
Copy link
Member

bo0tzz commented Oct 11, 2024

How does this PR relate to #12729?

@1-tempest
Copy link
Author

It's similar for sure.

I only have about 60 IOPS available on my machine due to my drives (smr unfortunately), so I'd be a little concerned about increasing IO on user connect.

Ideally I leave all IO for populating thumbs.

I'd need to try that PR to know if it makes any noticeable impact on user connect in my case (I assume I'd be worst case scenario), if the impact is marginal then my approach is indeed unnecessary.

@1-tempest 1-tempest closed this Oct 13, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
changelog:enhancement documentation Improvements or additions to documentation 🧠machine-learning
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants