This project extends the idea of the innovative architecture of Kolmogorov-Arnold Networks (KAN) to the Convolutional Layers, changing the classic linear transformation of the convolution to learnable non linear activations in each pixel.
MIT License
757
stars
75
forks
source link
is the b = [kernel_side // 2, kernel_side// 2] used in convolution.py at line 11 or for other usage? #8