feat: check the num of ILayer types and compare with ONNX-TRT in converter tests #3732
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Outdated suggestions cannot be applied.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.
Description
This PR enables checking the num of ILayer types and comparing with ONNX-TRT in converter tests.
To improve perf, we hope Torch-TRT to get the expected ILayer types for every single converter.
Since we observed on many models that ONNX-TRT generated TRT engines typically have better perf than Torch-TRT generated ones, to reduce perf gaps between Torch-TRT and ONNX-TRT, we introduce the mechanism to make sure Torch-TRT has the same num of ILayer types as ONNX-TRT.
Type of change
Please delete options that are not relevant and/or add your own.
Checklist: