JiangHao
JiangHao
@huningxin Do you have any suggestions in this problem?
> Generally WebNN uses channel-last layout, OpenVINO uses channel-first layout. Could you please share what are the shapes (convolution, reshape, permute) when you run it with webml-polyfill? Yes, the tensor...
> Is this figure just for openvino model? I suppose WebNN uses NHWC layout, e.g. so the shape of Convolution should be `[1, 360, 640, 72]`, is it? Yes, this...
> Did you try to permute `[1, 360, 720, 8, 3, 3]` to `[1, 360, 3, 640, 3, 8]`, then reshape to `[1, 1080, 1920, 8]`? Yes,I tried to permute...
I replace rehspape->permute->reshape with depthToSpace op: ```JavaScript case 'Reshape': { let nextNode = graph.nodes[i + 1]; let next2Node = graph.nodes[i + 3]; if(nextNode && next2Node && nextNode.operator === 'Permute' &&...