Toby Roseman
Toby Roseman
> Not sure if this is a framework issue, or one with `coremltools`. My hunch is the latter, so I'm asking here. I think this is much more likely to...
@tritolol thanks for the pull request. I apologize that no one seems to have looked at this. Please rebase this change on top of tip of `main`. Then I will...
Thanks @tritolol. CI: https://gitlab.com/coremltools1/coremltools/-/pipelines/1978748030
@tritolol - I don't think those failures are related to this change. Please rebase this change on top the latest `main`.
I'm not understanding the issue. Can you give us some toy code that demonstrates the issue?
I think it's going to be a lot easier to add NMS to your Keras model, then convert that updated model to Core ML. I would recommend taking that approach.
The converted model should not be any slower. For TensorFlow we support both `NonMaxSuppression` and `NonMaxSuppressionV5`.
I welcome the discussion here, but this isn't something which can be fixed in the coremltools repository. This a Core ML framework issue. Please submit a bug report here: https://developer.apple.com/bug-reporting/
No build failure with Python 3.12. We do have plenty of test failures, mostly related to the dependency versions we're using.
Thanks for the pull request. The fix looks good to me. CI: https://gitlab.com/coremltools1/coremltools/-/pipelines/2189637393