Updated CREF to handle bias=null for FC,Conv 8x16 CREF code#3492
Open
unmeshna017 wants to merge 5 commits intotensorflow:mainfrom
Open
Updated CREF to handle bias=null for FC,Conv 8x16 CREF code#3492unmeshna017 wants to merge 5 commits intotensorflow:mainfrom
unmeshna017 wants to merge 5 commits intotensorflow:mainfrom
Conversation
- Streamlined the usage of GetTensorData() and GetOptionaltensorData() - Added bias != nullptr condiition check wherever necessary.
Contributor
|
Hi @unmeshna017, Thanks for creating this PR! Can you please create a issue and add it in this PR description as BUG number? And also please to add the detailed description in the issue or the PR. |
rameshkunasi
added
the
ci:full
rameshkunasi
temporarily deployed
to
integration-test
GitHub Actions
Inactive
Contributor
|
@unmeshna017 these changes are failing some test cases. Can you please resolve and update the PR! |
unmeshna017
had a problem deploying
to
integration-test
GitHub Actions
Error
…ed operator" This reverts commit 452cca9.
unmeshna017
temporarily deployed
to
integration-test
GitHub Actions
Inactive
Comment out type equality check for input and output tensors.
rameshkunasi
temporarily deployed
to
integration-test
GitHub Actions
Inactive
Contributor
|
Looks good to me! @veblush can you please review this PR! |
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This change enhances the CREF kernels for Fully Connected and Convolutional layers to use higher precision accumulation when no bias term is present. Previously, the implementation would default to int32 accumulation when bias=NULL or INT32. This change is applicable only for the Int16x8 kernel type.
Adding @vp-cad and @joshih-cad as watchers.
BUG=3498