feat(workflows): enable "best" confidence for semantic segmentation#2425
Open
leeclemnet wants to merge 3 commits into
Open
feat(workflows): enable "best" confidence for semantic segmentation#2425leeclemnet wants to merge 3 commits into
leeclemnet wants to merge 3 commits into
Conversation
leeclemnet
changed the title
leeclemnet
changed the title
Mirror the object-detection v3 / instance-segmentation v4 pattern so the semantic segmentation model block (@v2) exposes the "best" confidence mode, which uses F1-optimal thresholds from model evaluation. "best" becomes the default, matching the sibling blocks. The mode was previously gated by a deliberate validator on SemanticSegmentationInferenceRequest because model eval did not yet produce per-class thresholds for this task; that gate is now removed. Note: this flips the default confidence_mode from "default" to "best" for existing v2 workflows that did not set the field explicitly. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
…ation The semantic-segmentation request entity no longer rejects "best", so move it from the rejects list into the accepted-confidence-modes set. keypoint-detection remains gated (model eval still produces no thresholds for it). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
leeclemnet
force-pushed
the
feat/semantic-seg-best-confidence
branch
from
2972289 to
18006b5
Compare
leeclemnet
requested review from
PawelPeczek-Roboflow,
dkosowski87,
grzegorz-roboflow,
hansent,
probicheaux,
rafel-roboflow and
yeldarby
as code owners
No sibling model block asserts the default confidence_mode value, so remove that assertion for consistency. The accepts-"best" test and the existing minimalistic-config test still cover the relevant behavior. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What does this PR do?
Allow "best" confidence for semantic segmentation inference now that model-eval supports it (https://github.com/roboflow/model_eval/pull/161), and make it the new default for the semantic seg model workflow block, matching other supported model blocks.
Related Issue(s): n/a
Type of Change
Testing
Test details:
Checklist
Additional Context