immich-app/immich
1
0
Fork 0
mirror of https://github.com/immich-app/immich.git synced 2026-02-05 00:30:57 +03:00
Code Issues 11 Packages Projects Releases 10 Wiki Activity

[Feature]: Add the ability to correct/train machine learning matches #408

New Issue
Closed
opened 2026-02-04 20:19:19 +03:00 by OVERLORD · 2 comments
OVERLORD commented 2026-02-04 20:19:19 +03:00
Owner
Copy Link

Originally created by @esmondmissen on GitHub (Nov 8, 2022).

Feature detail

Just a simple option to correct false positives in the search results.

If I teach "carrot" and it returns a beach umbrella, an option to say "Not a carrot" would be nice.

Could this also be used to train the model going forward?

Platform

Server

Originally created by @esmondmissen on GitHub (Nov 8, 2022). ### Feature detail Just a simple option to correct false positives in the search results. If I teach "carrot" and it returns a beach umbrella, an option to say "Not a carrot" would be nice. Could this also be used to train the model going forward? ### Platform Server
OVERLORD commented 2026-02-04 20:19:29 +03:00
Author
Owner
Copy Link

@alextran1502 commented on GitHub (Nov 8, 2022):

Currently, we don't have a re-train system yet. It would be quite a project of its own to provide the training platform for the application. I am keeping an eyes out for a more accurate or community train model though

@alextran1502 commented on GitHub (Nov 8, 2022): Currently, we don't have a re-train system yet. It would be quite a project of its own to provide the training platform for the application. I am keeping an eyes out for a more accurate or community train model though
OVERLORD commented 2026-02-04 20:19:32 +03:00
Author
Owner
Copy Link

@bo0tzz commented on GitHub (Nov 8, 2022):

Editing tags will be supported in the future. Dynamic model training is unfortunately not as simple as adding some ui, it'd need some solid ML knowledge as well. And it wouldn't really be effective at the scale of just one immich instance, so we'd need to share training data around etc - it gets hairy pretty quick.

@bo0tzz commented on GitHub (Nov 8, 2022): Editing tags will be supported in the future. Dynamic model training is unfortunately not as simple as adding some ui, it'd need some solid ML knowledge as well. And it wouldn't really be effective at the scale of just one immich instance, so we'd need to share training data around etc - it gets hairy pretty quick.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
accessibility
changelog:enhancement
changelog:security
changelog:translation
cli
date-time
documentation
external-library
format
good first issue
needs-answer
nice to have
sharing
tech-debt
📱mobile
🖥️web
🗄️server
🧠machine-learning
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
OVERLORD
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: immich-app/immich#408
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?