Latte is a convolutional neural network (CNN) inference engine written in C++ and uses AVX to vectorize operations. The engine runs on Windows 10, Linux and macOS Sierra.
Primary LanguageC++GNU General Public License v3.0GPL-3.0