🧠 Neural Network with Convolution Using as_strided (NumPy)

This project showcases a custom implementation of a convolutional neural network (CNN) using NumPy's as_strided method. The goal is to perform efficient 2D convolutions without relying on high-level deep learning frameworks like PyTorch or TensorFlow.

📌 Features

• Custom CNN Implementation: Built from scratch using NumPy, providing a deeper understanding of underlying operations.
• Efficient Convolution: Utilizes numpy.lib.stride_tricks.as_strided for memory-efficient convolution operations.
• Educational Purpose: Designed to reinforce concepts in deep learning by manually implementing core components.