/Minimal-Neural-Classifier

An artificial neural network that identifies the correct class [0..9] of handwritten digits from the MNIST dataset. It's derived from first principles implemented in only 300 lines of code. (No dependencies)

Primary LanguageC#

Watchers