[js/webgpu] Fix max pool shape end with 0

[axinging]

Bug: #21386

Description

Motivation and Context

[axinging]

@qjia7 @xhcao @hujiajie @jzm-intel @gyagp PTAL

[qjia7]

If I understand correctly, the error is caused by the 2d shapes of kernelShape/strides/dilations while the 1d shapes are expected since it's pool1d instead of pool2d.

Based on that, I think the right fixing should be passing the right shape to attributes in c++ side instead of modifying it in js. See code here https://github.com/microsoft/onnxruntime/blob/main/onnxruntime/core/providers/js/operators/pool.h#L29-L32, it will always generate a 2d array to attributes.dilations/attributes.kernel_shape/attributes.strides. To get the right attribute shapes, you can refer to conv's implementation here https://github.com/microsoft/onnxruntime/blob/main/onnxruntime/core/providers/js/operators/conv.h#L30-L45 which supports any expected shapes.

[axinging]

If I understand correctly, the error is caused by the 2d shapes of kernelShape/strides/dilations while the 1d shapes are expected since it's pool1d instead of pool2d.

Based on that, I think the right fixing should be passing the right shape to attributes in c++ side instead of modifying it in js. See code here https://github.com/microsoft/onnxruntime/blob/main/onnxruntime/core/providers/js/operators/pool.h#L29-L32, it will always generate a 2d array to attributes.dilations/attributes.kernel_shape/attributes.strides. To get the right attribute shapes, you can refer to conv's implementation here https://github.com/microsoft/onnxruntime/blob/main/onnxruntime/core/providers/js/operators/conv.h#L30-L45 which supports any expected shapes.

Thanks @qjia7, updated.

[qjia7]

LGTM, thanks!

@guschmue @fs-eire @satyajandhyala Please take a look, thanks.

[fs-eire]

/azp run Windows ARM64 QNN CI Pipeline,Windows x64 QNN CI Pipeline,Windows CPU CI Pipeline,Windows GPU CI Pipeline,Windows GPU TensorRT CI Pipeline,ONNX Runtime Web CI Pipeline,Linux CPU CI Pipeline,Linux CPU Minimal Build E2E CI Pipeline,Linux GPU CI Pipeline,Linux GPU TensorRT CI Pipeline

[fs-eire]

/azp run Linux OpenVINO CI Pipeline,Linux QNN CI Pipeline,MacOS CI Pipeline,orttraining-amd-gpu-ci-pipeline,orttraining-linux-ci-pipeline,orttraining-linux-gpu-ci-pipeline,orttraining-ortmodule-distributed,onnxruntime-binary-size-checks-ci-pipeline,Big Models,Linux Android Emulator QNN CI Pipeline

[fs-eire]

/azp run Android CI Pipeline,iOS CI Pipeline,ONNX Runtime React Native CI Pipeline

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[fs-eire]

/azp run Windows GPU CUDA CI Pipeline,Windows GPU DML CI Pipeline,Windows GPU Doc Gen CI Pipeline

[azure-pipelines]

No pipelines are associated with this pull request.

[fs-eire]

/azp run Windows GPU CUDA CI Pipeline, Windows GPU DML CI Pipeline, Windows GPU Doc Gen CI Pipeline

[azure-pipelines]

No pipelines are associated with this pull request.

[fs-eire]

/azp run Windows ARM64 QNN CI Pipeline,Windows x64 QNN CI Pipeline,Windows CPU CI Pipeline,Windows GPU CI Pipeline,Windows GPU TensorRT CI Pipeline,Linux CPU CI Pipeline,Linux CPU Minimal Build E2E CI Pipeline,Linux GPU CI Pipeline,Linux GPU TensorRT CI Pipeline,Linux OpenVINO CI Pipeline

[fs-eire]

/azp run Linux QNN CI Pipeline,MacOS CI Pipeline,orttraining-amd-gpu-ci-pipeline,orttraining-linux-ci-pipeline,orttraining-linux-gpu-ci-pipeline,orttraining-ortmodule-distributed,Big Models,Linux Android Emulator QNN CI Pipeline,Android CI Pipeline,iOS CI Pipeline

[fs-eire]

/azp run ONNX Runtime React Native CI Pipeline

[azure-pipelines]

Azure Pipelines could not run because the pipeline triggers exclude this branch/path.

[azure-pipelines]

Azure Pipelines could not run because the pipeline triggers exclude this branch/path.

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[fs-eire]

/azp run Windows ARM64 QNN CI Pipeline,Windows x64 QNN CI Pipeline,Windows CPU CI Pipeline,Windows GPU CUDA CI Pipeline,Windows GPU DML CI Pipeline,Windows GPU Doc Gen CI Pipeline,Windows GPU TensorRT CI Pipeline,ONNX Runtime Web CI Pipeline,Linux CPU CI Pipeline,Linux CPU Minimal Build E2E CI Pipeline

[fs-eire]

/azp run Linux GPU CI Pipeline,Linux GPU TensorRT CI Pipeline,Linux OpenVINO CI Pipeline,Linux QNN CI Pipeline,MacOS CI Pipeline,orttraining-amd-gpu-ci-pipeline,orttraining-linux-ci-pipeline,orttraining-linux-gpu-ci-pipeline,orttraining-ortmodule-distributed,onnxruntime-binary-size-checks-ci-pipeline

[fs-eire]

/azp run Big Models,Linux Android Emulator QNN CI Pipeline,Android CI Pipeline,iOS CI Pipeline,ONNX Runtime React Native CI Pipeline

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[fs-eire]

This PR is replaced by #21698 because the build pipeline failed to trigger.