Tag：Fourier
-
Fourier transform in convolutional neural network: 1024×1024 Fourier convolution
Convolutional neural network (CNN) has been widely used and proved to be very successful. However, the calculation of convolution is very inefficient. The sliding window requires a lot of calculation and limits the size of the filter. Usually, small nuclei between [3,3] and [7,7] limit the receptive field (the recent large kernel convolution can refer […]
-
Fast Fourier transform theory
There are many formulas in this paper. You are welcome to make corrections 1. Fourier transform of periodic discrete signal The formula of discrete signal Fourier transform is as follows: The principle of discrete Fourier transform is to copy and expand the original aperiodic signal into periodic signal. In the actual digital circuit processing, the […]
-
Graph neural network Chapter 5 graph signal processing and graph convolution neural network reading notes
A few words before: Finally! Here comes the most relevant content of GNN! The first four chapters are all preliminary knowledge, or introductory knowledge. In fact, they are not particularly related to GNN. But from the beginning of this chapter, it is the core of GNN: graph signal processing. This part is actually very critical, […]
-
Fourier series expansion of common waveforms
introduction Recently, it is necessary to carry out the trapezoidal waveFourier series After searching some information (I’m sorry, I didn’t think about doing integral by myself at the beginning), and then I didn’t find the result I wanted (in fact, it was similar, but it wasn’t arbitrary period, and it didn’t change at that time). […]