Handwritten Digit Recognition
with Neural Networks
Python
TensorFlow / Keras
Jupyter Notebook
Deep Learning
This project compares two neural network architectures — a Multi-Layer Perceptron (MLP) and a Convolutional Neural Network (CNN) — on the MNIST handwritten digit dataset. The notebook walks through data preparation, model building, training, and evaluation to show how adding spatial awareness through convolutions improves classification accuracy, even on a simple image task.
GitHub RepoFull Notebook
