Skip to content

[QNN EP] Fix error handling for Softmax/ReduceOps#16700

Merged
adrianlizarraga merged 3 commits into
mainfrom
adrianl/qnn-better-error-reducesum-softmax
Jul 14, 2023
Merged

[QNN EP] Fix error handling for Softmax/ReduceOps#16700
adrianlizarraga merged 3 commits into
mainfrom
adrianl/qnn-better-error-reducesum-softmax

Conversation

@adrianlizarraga
Copy link
Copy Markdown
Contributor

Description

  • Fix check for Softmax with axis attributes not equal to -1. QNN EP only supports axis values equal to -1 (or rank - 1).
  • Explicit error when Reduce* ops have an input with rank > 4 on HTP backend (unsupported).
  • Correctly filter out partitions that only contain a single QuantizeLinear or DequantizeLinear node.
  • Add tests for the above and clean up unnecessary usage of test description labels.

Motivation and Context

Make it easier to debug why a model may not be supported.

Comment thread onnxruntime/core/providers/qnn/qnn_execution_provider.cc
@adrianlizarraga adrianlizarraga marked this pull request as ready for review July 14, 2023 00:29
@adrianlizarraga adrianlizarraga changed the title [QNN EP] Better error handling for Softmax/ReduceOps [QNN EP] Fix error handling for Softmax/ReduceOps Jul 14, 2023
Comment thread onnxruntime/test/optimizer/qdq_test_utils.h
Copy link
Copy Markdown
Contributor

@HectorSVC HectorSVC left a comment

Choose a reason for hiding this comment

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

:shipit:

@adrianlizarraga adrianlizarraga merged commit a189e76 into main Jul 14, 2023
@adrianlizarraga adrianlizarraga deleted the adrianl/qnn-better-error-reducesum-softmax branch July 14, 2023 20:47
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.

2 participants