[WebNN] Fallback int64 indices to int32 - #26308
Conversation
|
Dwayne Robinson (@fdwr), Guenther Schmuelling (@guschmue), PTAL, thanks! This will make Whisper Base demo on WebNN Developer Preview can fully run behind TFLite backend without any fallback. |
Dwayne Robinson (fdwr)
left a comment
Dwayne Robinson (fdwr)
left a comment
There was a problem hiding this comment.
👍 Minor thought, else LGTM.
|
/azp run ONNX Runtime Web CI Pipeline,Windows GPU CI Pipeline,Linux Android Emulator QNN CI Pipeline,Windows GPU WebGPU CI Pipeline,Windows OpenVINO CI Pipeline |
|
/azp run Linux CPU CI Pipeline,Linux CPU Minimal Build E2E CI Pipeline,Linux GPU CI Pipeline,Linux GPU TensorRT CI Pipeline,Linux OpenVINO CI Pipeline,MacOS CI Pipeline,Windows ARM64 QNN CI Pipeline,Windows CPU CI Pipeline |
|
/azp run Windows GPU CUDA CI Pipeline,Windows GPU DML CI Pipeline,Windows GPU Doc Gen CI Pipeline,Win_TRT_Minimal_CUDA_Test_CI |
|
/azp run Windows GPU TensorRT CI Pipeline,onnxruntime-binary-size-checks-ci-pipeline,orttraining-linux-ci-pipeline,orttraining-linux-gpu-ci-pipeline,orttraining-ortmodule-distributed,Windows x64 QNN CI Pipeline,Big Models |
|
Azure Pipelines successfully started running 1 pipeline(s). |
|
/azp run Test Linux CUDA x64 Release,Test Linux TensorRT x64 Release,web_Debug / build_onnxruntime_web,web_Release / build_onnxruntime_web |
|
/azp run Linux QNN CI Pipeline |
|
Azure Pipelines successfully started running 1 pipeline(s). |
|
No pipelines are associated with this pull request. |
1 similar comment
|
No pipelines are associated with this pull request. |
|
Azure Pipelines successfully started running 2 pipeline(s). |
|
Azure Pipelines successfully started running 1 pipeline(s). |
|
/azp run ONNX Runtime Web CI Pipeline,Windows GPU CI Pipeline,Linux Android Emulator QNN CI Pipeline,Windows GPU WebGPU CI Pipeline,Windows OpenVINO CI Pipeline |
|
/azp run Linux CPU CI Pipeline,Linux CPU Minimal Build E2E CI Pipeline,Linux GPU CI Pipeline,Linux GPU TensorRT CI Pipeline,Linux OpenVINO CI Pipeline,MacOS CI Pipeline,Windows ARM64 QNN CI Pipeline,Windows CPU CI Pipeline |
|
/azp run Windows GPU CUDA CI Pipeline,Windows GPU DML CI Pipeline,Windows GPU Doc Gen CI Pipeline,Win_TRT_Minimal_CUDA_Test_CI |
|
/azp run Windows GPU TensorRT CI Pipeline,onnxruntime-binary-size-checks-ci-pipeline,orttraining-linux-ci-pipeline,orttraining-linux-gpu-ci-pipeline,orttraining-ortmodule-distributed,Windows x64 QNN CI Pipeline,Big Models |
|
Azure Pipelines successfully started running 1 pipeline(s). |
|
/azp run Test Linux CUDA x64 Release,Test Linux TensorRT x64 Release,web_Debug / build_onnxruntime_web,web_Release / build_onnxruntime_web |
|
/azp run Linux QNN CI Pipeline |
|
Azure Pipelines successfully started running 1 pipeline(s). |
|
No pipelines are associated with this pull request. |
1 similar comment
|
No pipelines are associated with this pull request. |
|
Azure Pipelines successfully started running 2 pipeline(s). |
|
Azure Pipelines successfully started running 1 pipeline(s). |
|
Hmm, I've retried the Linux CI build and CUDA build twice now. Wonder if this happens in main now. 
|
ONNX's ScatterND and ScatterElements limit their indices input to int64, but some WebNN backends only support int32. As a workaround for such backends, we can insert a Cast operation to convert the data type.
d5beb4f to
d74c228
Compare
Let me rebase it to main and pls. help retrigger the CI. |
|
/azp run Linux QNN CI Pipeline,Win_TRT_Minimal_CUDA_Test_CI,Windows ARM64 QNN CI Pipeline,Windows GPU Doc Gen CI Pipeline |
|
Azure Pipelines successfully started running 4 pipeline(s). |
|
Guenther Schmuelling (@guschmue), all CLs passed, could you please merge this one? Thanks! |
ONNX's ScatterND and ScatterElements limit their indices input to int64, but some WebNN backends only support int32 indices. As a workaround for such backends, we can insert a Cast operation to convert the data type.
Upgrade the onnxruntime-web to 1.24.0-dev.20251104-75d35474d5 which includes the PR [1] to support ScatterND with int64 indices for TFLite backend. With this PR, Whisper Base demo can fully run behind TFLite backend without any fallback. Upgrade the test dist as well BTW. [1] microsoft/onnxruntime#26308
3 participants
ONNX's ScatterND and ScatterElements limit their indices input to int64, but some WebNN backends only support int32 indices. As a workaround for such backends, we can insert a Cast operation to convert the data type.