Support multi-output models with bioimageio spec 0.5 #117
Conversation
This file contains 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
…nto multi-output-model
alanocallaghan
requested review from
petebankhead
and removed request for
petebankhead
petebankhead
reviewed
Feb 7, 2025
There was a problem hiding this comment.
I haven't tried running it, but added a few comments, mostly related to clarity.
I don't really have an overview of the latest spec and instanseg developments, so I'll go with what you think is best. Will discover if I have any strong opinions after it's merged & I've played around with it some more.
src/main/java/qupath/ext/instanseg/core/InstanSegOutputToObjectConverter.java
Outdated
Show resolved
Hide resolved
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Add support for multi-output models using the bioimageio 0.5.x model spec.
We currently handle logits and embeddings by adding them to the measurements of output objects. We don't have a good way to name measurements yet; hopefully we can get axis names at some point in the future of the bioimageio spec. Detection classes are set as PathClass on the output objects.
Detection classes are supported with a NominalOrOrdinalDataDescr. If logits are present too, we try to match the names of outputs and logits. Otherwise, we default to "Class 1", "Logit 1" etc.
It's not totally clear how to handle multiple classes. "Class 1: Class A"?
Semantic segmentation is recognised as an output type, but support is not currently implemented (suggest this could be a future PR for now?).
Superset of #100
See qupath/qupath-bioimageio-spec#12