getrasa / deep-feed-forward-neural-net-from-scratch

The purpose of this project is to build Deep Feed Forward Neural Network from scratch in python and train it on images of cats and dogs.

Deep-Feed-Forward-Neural-Net-from-scratch

The purpose of this project is to build Deep Feed Forward Neural Network from scratch in python. After the network is built, it will be trained using images of cats and dogs in order to learn the mapping from the inputs to the output and classify them correctly.

Feed Forward Neural network is not the best choice for image classification and is significantly inferior to Convolutional Neural Network. It's merely used for the purpose of showing how it works.

Required Packages

• python 3
• numpy
• cv2
• jupyter
• matplotlib

NbViewer

If you don't have above packages installed. You can still view the notebook on this link: http://nbviewer.jupyter.org/github/getrasa/Deep-Feed-Forward-Neural-Net-from-scratch/blob/master/L-layer%20Feed%20Forward%20Neural%20Network.ipynb