GitHub

[convolution] Intermediate matrix rather than list for

Convolution seems like a good target example for #270, since we’ll need a way to specify/navigate to the appropriate intermediate value for the weightedSum computation. Currently weightedSum is ...
IEEE

Automatic Generation of High-Performance Convolution Kernels on ARM CPUs for Deep Learning

Abstract: We present FastConv, a template-based code auto-generation open-source library that can automatically generate high-performance deep learning convolution kernels of arbitrary ...
EDN

Automotive vision system recognizes road signs: Part 3 – Implementation of a Sobel filter in C and assembler

Traffic sign recognition is based on pattern recognition algorithms that are well understood by machine vision specialists. The first two parts of this series detailed a camera-based traffic-sign ...
IEEE

Graph Polynomial Convolution Models for Node Classification of Non-Homophilous Graphs

Abstract: We investigate graph convolution networks with efficient learning from higher-order graph convolutions and direct learning from adjacency matrices for node classification. We revisit the ...
GitHub

Feature-Base-Jump-Encoding-for-Convolutional-Matrices

This repository demonstrates a novel technique for feature encoding in neural network (NN) matrices. Instead of expanding matrix size (by adding channels) or increasing compute via multiplication, ...