Hello, thank you for providing the code. I am now doing the wavelet transform with 1D (shape is [1,64,L]) features. But you get a different scale, it's the same length when you add it up. This is different from image processing, where they get the same length of feature scale. Don't know if you can provide code formulas that increase the understanding of the algorithm.
Hello, thank you for providing the code. I am now doing the wavelet transform with 1D (shape is [1,64,L]) features. But you get a different scale, it's the same length when you add it up. This is different from image processing, where they get the same length of feature scale. Don't know if you can provide code formulas that increase the understanding of the algorithm.