feat(compression): implement tensor decompression in op concatenation #3014
+282
−57
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
Implement tensor decompression in op concatenation. Extend tests to validate operation on compressed tensors. BUG=part of tensorflow#2636
yushhuang
reviewed
Dec 16, 2024
| output_tensor->params.zero_point); | ||
| } else if (input_type == kTfLiteInt16) { | ||
| // Make sure that all Int16 inputs have a null zero-point. | ||
| TF_LITE_ENSURE_EQ(context, input->params.zero_point, 0); |
There was a problem hiding this comment.
Is there a reason why the scale is not checked in the int16 case? AFAICS the (at least the reference) concatenation kernel does not take into account any scaling/quantization at all.
There was a problem hiding this comment.
I agree with your conclusion, however this is a copy-paste of the TfLite code.
There was a problem hiding this comment.
Yes, also is there are reason why it has to be zero? Seems that concatenation in general shouldn't care about quantization parameters, as long as they match in and out.
There was a problem hiding this comment.
@tinskip This same check is performed by the LiteRT (TfLite) reference implementation.
yushhuang
approved these changes
Dec 16, 2024
suleshahid
reviewed
Dec 16, 2024
suleshahid
approved these changes
Dec 16, 2024
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Outdated suggestions cannot be applied.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.
Implement tensor decompression in op concatenation. Extend
tests to validate operation on compressed tensors.
BUG=part of #2636