Enabling of the Performance model for NNPA / z17#3383
Open
AlexandreEichenberger wants to merge 20 commits intoonnx:mainfrom
Open
Enabling of the Performance model for NNPA / z17#3383AlexandreEichenberger wants to merge 20 commits intoonnx:mainfrom
AlexandreEichenberger wants to merge 20 commits intoonnx:mainfrom
Conversation
Signed-off-by: Alexandre Eichenberger <alexe@us.ibm.com>
Signed-off-by: Alexandre Eichenberger <alexe@us.ibm.com>
Signed-off-by: Alexandre Eichenberger <alexe@us.ibm.com>
…ed as faster than CPU) Signed-off-by: Alexandre Eichenberger <alexe@us.ibm.com>
Signed-off-by: Alexandre Eichenberger <alexe@us.ibm.com>
Signed-off-by: Alexandre Eichenberger <alexe@us.ibm.com>
Signed-off-by: Alexandre Eichenberger <alexe@us.ibm.com>
Signed-off-by: Alexandre Eichenberger <alexe@us.ibm.com>
Signed-off-by: Alexandre Eichenberger <alexe@us.ibm.com>
Signed-off-by: Alexandre Eichenberger <alexe@us.ibm.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
In Granite embedding, I noticed that we are sending a scalar SQRT to NNPA because the current policy is to send any legal operations to the NNPA.
After adding support to missing new operations in z17 (sqrt & leaky relu), I noticed that some patterns transforming ONNX ops to ZHigh ops were firing without considerations for placement (aka the pass that determines where ops should run, NNPA vs CPU). So I added some additional conditions to rules for the ONNX -> ZHigh transition.
I also modified a bit the handling of dynamic dimensions expected values when plugged into the performance models for CPU and NNPA.