Arm backend: Add support for quant. decomposition #15993
+304
−44
Conversation
This file contains hidden or 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
For mixed type models we need to be able switch between FP and INT, meaning quantize and dequantize online. As there's no quantize or dequantize operator in TOSA, we need to decompose these operators to TOSA operators. This commit introduces a pass that decomposes q-dq nodes into more primitive nodes. Signed-off-by: Oscar Andersson <[email protected]> Change-Id: I05d52572a8d614f9f77119f031915ef8bf2a00e3
Enable q-dq decomposition in ArmPassManager. Signed-off-by: Oscar Andersson <[email protected]> Change-Id: Idc0d502c2ef1e6081abe55637ba21caad6c847d2
Tag quantized nodes with ArmAnnotationInfo to make q-dq-folding pass work as expected. Signed-off-by: Oscar Andersson <[email protected]> Change-Id: I2e473c1a268a01a9bf31a85802b56d7d2fd9dc38
oscarandersson8218
requested review from
digantdesai and
kimishpatel
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/15993
Note: Links to docs will display an error until the docs builds have been completed. ❗ 2 Active SEVsThere are 2 currently active SEVs. If your PR is affected, please view them below: This comment was automatically generated by Dr. CI and updates every 15 minutes. |
meta-cla
bot
added
the
CLA Signed
oscarandersson8218
added
ciflow/trunk
release notes: none
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
For mixed type models we need to be able switch between FP and INT,
meaning quantize and dequantize at runtime. As there are no quantize or
dequantize operators in TOSA, we need to decompose these operators to
TOSA operators. This commit introduces a pass that decomposes q-dq nodes
into more primitive nodes.
This also affects Cortex-M backend as it uses the qdq-folding pass in arm/_passes.
cc @freddan80 @per @zingo @digantdesai