Overview

Computer vision is an exciting domain within machine learning that sees a range of interesting applications. This educational challenge aims to introduce you to convolutional neural networks (CNNs) through a multi-class image classification challenge based on the CIFAR-10 dataset.

This is a great opportunity to learn about some core ideas in computer vision using PyTorch.

Getting started 💡

To get started, take a look at our tutorial notebook:

https://github.com/DoxaAI/educational-challenges/blob/main/cifar-10/getting-started.ipynb

A simple convolutional neural network is provided to help you get started:

If you have any questions, feel free to ask them in the DOXA AI community Discord server.

The data 💾

The CIFAR-10 dataset was originally collected by Alex Krizhevsky, Vinod Nair and Geoffrey Hinton at the University of Toronto. It consists of 60,000 32×32 colour RGB images in total (split across a 50,000-image training dataset and a 10,000-image test dataset), split across ten classes outlined below.

Evaluation 📐

Your submissions are ranked according to their accuracy in this competition.