Experiments with Convolutional Neural Networks (CNNs) to classify images from the Intel Image Classification dataset.

This repository implements a 3-layer neural network with L2 and Dropout regularization using Python and NumPy. It focuses on reducing overfitting and improving generalization. The project includes forward/backward propagation, cost functions, and decision boundary visualization. Inspired by the Deep Learning Specialization from deeplearning.ai.