Feat (export): deprecate FINN ONNX export #753
Merged
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
Giuseppe5
force-pushed
the
deprecate_finn_export
branch
from
b3b9941 to
d27799c
Compare
Giuseppe5
force-pushed
the
deprecate_finn_export
branch
from
a99d4cc to
328d3a6
Compare
volcacius
approved these changes
Nov 15, 2023
Giuseppe5
force-pushed
the
deprecate_finn_export
branch
from
328d3a6 to
14823fa
Compare
Giuseppe5
force-pushed
the
deprecate_finn_export
branch
from
14823fa to
358517f
Compare
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Outdated suggestions cannot be applied.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.
The WBIOL tests have been merged all under Brevitas-ORT. This introduces two ONNX validation path (one actually calling ORT, the other one using QONNX), but allows us to stress test QONNX export with all the settings we already use for QCDQ, etc..
The other option to avoid this would be to have code duplication, which makes maintainability more complicated.
All the other tests have been kept in place and adapted to export to QONNX, with the main difference being that now we export a float tensor, or we add a QuantNode (in the case of TruncAvgPool).