Skip to content

Conversation

@gprateek93
Copy link
Collaborator

@gprateek93 gprateek93 commented Mar 16, 2022

This commit fixes the 2nd and 3rd return types of the aten.native_layer_norm.
Previously the mean and rSTD were returned with reduction dims removed.
This commit fixes this and keeps the reduction dims of the results.

Signed-Off-By: Prateek Gupta prateek@nord-labs.com

fixes: #665

Copy link
Contributor

@cathyzhyi cathyzhyi left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for the fix!

This commit fixes the 2nd and 3rd return types of the `aten.native_layer_norm`.
Previously the mean and rSTD were returned with reduction dims removed.
This commit fixes this and keeps the reduction dims of the results.

Signed-Off-By: Prateek Gupta <prateek@nord-labs.com>
@gprateek93 gprateek93 force-pushed the prateek/fix-aten-native-layer-norm branch from 3ed910c to 9a46f7c Compare March 17, 2022 06:24
@gprateek93 gprateek93 merged commit 7256c9e into llvm:main Mar 17, 2022
@gprateek93 gprateek93 deleted the prateek/fix-aten-native-layer-norm branch March 17, 2022 06:38
@silvasean
Copy link
Contributor

Thanks for the quick follow-up!

qedawkins pushed a commit to nod-ai/torch-mlir that referenced this pull request Oct 3, 2022
* Various small fixes

Some typos; some white space; removing duplicate lib in OMLibs; removing a couple of unused include directories; removing unused definition INCLUDE_ONNX_ML; moving a dependency to the correct cmake; removing add_c_unit_test function which is only used in one place; making debug.py work (better) on windows

Signed-off-by: Stella Stamenova <stilis@microsoft.com>

* Run clang-format

Signed-off-by: Stella Stamenova <stilis@microsoft.com>

Co-authored-by: Kevin O'Brien <caomhin@us.ibm.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

aten::native_layer_norm 2nd and 3rd results not implemented correctly

4 participants