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

Remove user customized search model #946

Merged
merged 14 commits into from
Oct 24, 2024
Merged

Remove user customized search model #946

merged 14 commits into from
Oct 24, 2024

Conversation

sabaimran
Copy link
Member

  • Use a single standard search model across the server. There's diminishing benefits for having multiple user-customizable search models.
  • We may want to add server-level customization for specific tasks
  • Store the search model used to generate a given entry on the Entry object
  • Remove user-facing APIs and view
  • Add a management command for migrating the default search model on the server

In a future PR (after running the migration), we'll also remove the UserSearchModelConfig

@sabaimran
Add outline, code for a management command to update search model con…
6a98215 
…figs

- Not yet tested
- Updates all Entry and User configs to point to a given default config. This helps us keep models up to date and management them across the server.
- Otherwise, can run into issues with searching against the knowledge base when the model that generated embeddings for different documents differ.
@sabaimran
Process Entry objects for users who did not specify any thing explicitly
f69c057
@sabaimran
Remove the search model option card
961a6a4
@sabaimran
Fix entry / entries misnaming in regenerate_entry method
e993517
@sabaimran
Merge branch 'master' of github.com:khoj-ai/khoj into fix/user-config…
f6dc1c4 
…ured-search-models
@sabaimran
Associate Entry objects with a search model, hide search model from u…
3b8eea1 
…ser settings

- Currently, we run a lot of operations by pulling the user's search model from the DB and adding personalization. This has some serious limitations, especially when there are multiple embeddings models used in the user's knowledge base.
- Shift to a model where the server-wide setting is used to maintain the default search model, which will be used by everyone.
- For each Entry, associate it with the search model that was used to generate it
- Remove APIs for configuring search model
@sabaimran
Add some exception handling, and update the default search model at t…
cf67daa 
…he end of the change_default_model management command
@sabaimran
Fix issue and do some local runs
50c4a87 
- Add transaction around the regenerate_entries and do a bulk update
@sabaimran sabaimran requested a review from debanjum October 21, 2024 23:26
@sabaimran sabaimran added the maintain Maintain code, documentation or project label Oct 21, 2024
debanjum
debanjum approved these changes Oct 23, 2024
View reviewed changes
@sabaimran sabaimran merged commit 5120597 into master Oct 24, 2024
9 checks passed
@sabaimran sabaimran deleted the fix/user-configured-search-models branch October 24, 2024 00:38
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@debanjum debanjum debanjum approved these changes

Assignees
No one assigned
Labels
maintain Maintain code, documentation or project
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@sabaimran @debanjum