Skip to content

Simple neural network to solve the MNIST digit classification problem

Notifications You must be signed in to change notification settings

br00ks/mnist-digit

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Building a neural network to recognize handwritten digits

Simple neural network to solve the MNIST digit classification problem using Tensorflow. We are using the famous MNIST dataset, which consists of pictures of handwritten digits from 0 to 9. Each picture is a 28x28 pixel grayscale image. The training dataset consists of 60.000 images, the testing dataset of 10.000 images.

Preprocessing

Before we can start we need to normalise the pixel values, so that they lie between 0 and 1 instead of 0 and 255. In order to do so, we divide the training and test datasets by 255.

Model

  • 1st keras layer flattens 28x28 array into 784 pixel vector
  • 2nd keras layer is fully connected 128 node layer with ReLu activation
  • Output layer is fully connected 10 node softmax layer

References:

About

Simple neural network to solve the MNIST digit classification problem

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages