CIFAR-10 Image Classification 🧠

About The Project

This project is a part of the Computational Intelligence course. The purpose of this project is to build an image classifier using the CIFAR-10 dataset from scratch and without relying on pre-built deep learning frameworks such as TensorFlow or PyTorch.

CIFAR-10 is a well-known image classification dataset that consists of 60,000 32x32 color images in 10 classes, with 6,000 images per class.

This project will give us an opportunity to implement a neural network from scratch (implementig vectorized and non-vectorized gradient descent) and apply the concepts we have learned in class to solve a real-world image classification problem. The project will involve the following steps:

• Loading data
• Feed forward
• Back propagation using gradient descent (vectorized / non-vectorized)
• Clculate accuracy of model on test set

Library we used in this project:

• numpy
• matplotlib

In this repository, you will find the following files:

• CI_project1.ipynb: main project and implementation
• CI_project1_report.pdf: report and extra information
• CI_project1.pdf: instruction to project