Onnx is the name of the file format and it's possible we'll end up supporting different formats to do the same thing. Also we now have 3 different features that use Onnx compute graphs so there is a risk of confusion.
We have the first Onnx plugin that is run on sources during measurement. I don't really have a good name for it (suggestions?)
We have the machine learning detection (segmentation algorithm "ML") that uses the --segmentation-onnx-model configuration option
(maybe we can rename that to -segmentation-machine-learning-model ?)
And then we have the OnnxModel in the Python configuration,. We can rename to ComputeGraphModel?
If anybody (@astrorama/developers ) has ideas, let me know, we need to do this for the release.
Onnx is the name of the file format and it's possible we'll end up supporting different formats to do the same thing. Also we now have 3 different features that use Onnx compute graphs so there is a risk of confusion.
We have the first Onnx plugin that is run on sources during measurement. I don't really have a good name for it (suggestions?)
We have the machine learning detection (segmentation algorithm "ML") that uses the --segmentation-onnx-model configuration option (maybe we can rename that to -segmentation-machine-learning-model ?)
And then we have the OnnxModel in the Python configuration,. We can rename to ComputeGraphModel?
If anybody (@astrorama/developers ) has ideas, let me know, we need to do this for the release.