Skip to content

Latest commit

 

History

History
11 lines (10 loc) · 486 Bytes

README.md

File metadata and controls

11 lines (10 loc) · 486 Bytes

Convolutional Network Using Only Numpy and Scipy

A Neural Network created without using a mahcine learning library such as Keras or PyTorch. Created by Jonathan Levy and Christopher Bloodsworth.

Contributions

Jonathan Levy

I implemented the fully connected layers as well as calculating the back propogation between them.

Christopher Bloodsworth

Chris implemented the convolutional layers and the max pooling layers.

Results

Achieved 78.57% accuracy on MNIST Fashion Test.