Building neural networks using the MNIST dataset has become a classic problem in the field of machine learning and deep learning.
MNIST, a collection of handwritten digits, serves as a fundamental benchmark for testing and evaluating various machine learning algorithms, particularly neural networks. It's a simpler problem and an ideal starting point to grasp the concepts of neural networks and their training process.
This is a simple implementation of a neural network based on this classical problem.
I tried implementing a somewhat Object Oriented approach to the neural network, also type coding it.
Before trying to run the code, unzip the mnist dataset handwritten_numbers.csv found dataset folder.
To run the code, simply download (or upload it to an online service) the jupyter file and run it, it should download the required packages and run without problem.
Hope you enjoy =-)