Skip to content

SImple feed forward neural network in Python to detect handwritten diggets using MNIST Data set.

Notifications You must be signed in to change notification settings

tobiasbrinker/nn-in-python

Repository files navigation

Simple Neural Network in Python

This is a simple feedforward neural network programm i created by using the book "Make your own Neural Network" by Tariq Rashids. It can read in pictures of handwritten digits from 0 to 9 of an 28x28 PNG-Image. It's using a 3 layer network and the sigmoid activiation function. For Training i used the MNIST-Dataset in a csv-format of 66000 examples. I tested it on several learning rates and epoches aswell as 100 and 200 hiddennodes. My best rate of success was 94,83% which is good enough for this simple network, even though there way better ones out there.

About

SImple feed forward neural network in Python to detect handwritten diggets using MNIST Data set.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages