importonnxfromonnx2kerasimportonnx_to_keras# Load the ONNX model from the specified pathonnx_model=onnx.load('C:/Users/wood/Desktop/Anamoly _Detection/anomalib/results/Patchcore/MVTec/toothbrush/v0/weights/onnx/weights/onnx/model.onnx')# Convert the ONNX model to Keras modelk_model=onnx...
To me this is not a strange phenomenon at all. One of the two inputs ofMulin this ONNX model is a constant. TensorFlow Lite has by far the most powerful optimizer than ONNX, with the ability to optimize and precompute all meaningless operations. ThisTransposeseems to be a specification th...