CNN-using-Numpy

This repository contains an implementation of a basic CNN from scratch using just the Numpy library.
The code used here is what i learned in the coursera deeplearning.ai specialization course 4 Convolutional Neural Networks in addition with implemented of an ANN.
the implementation is hardcoded to have a softmax output layer and performing mini-batch gradient descent with 60 batches.
Furthermore the data used in this repo is MNIST dataset which is available at this link: http://yann.lecun.com/exdb/mnist/ with instructions to download and use.