CIFAR-10 Image Classification using Neural Networks and Statistical methods like KNN and Bayes Classifiers
Analysis of CIFAR-10 image dataset to compare the difference between CNN, KNN and Bayes Classifier. The CIFAR-10 dataset consists of 60,000 32x32 colour images in 10 classes. The purpose is to find which methods are more effective in comparison which can have practical applications in computer vision.
Our objective in this project to implement statistical methods like KNN & Bayes Classifier and neural network like CNN to analyse the varying classification of each method. The CIFAR-10 dataset is a collection of 60,000 images, each of size 32x32, categorized into 10 classes, with 6,000 images per class. The 10 different classes are namely airplane, automobile, bird, cat, deer, dog, frog, horse, ship and truck. The dataset is commonly used for object recognition tasks in computer vision. In order to compare these methods on CIFAR-10, we can evaluate their performance on a held-out test set using metrics such as accuracy, precision, recall, F1-score and support. After evaluation we observed that CNN performed way better compared to statistical methods. //maybe we can add a brief description of each method
Libraries : tensorflow, numpy, sklearn , matplotlib
- Laasya Nellore : laasyanellore@arizona.edu
- Tanisha Lohchab : tanishal@arizona.edu
This project is licensed under the MIT License - see the LICENSE.md file for details
- GeeksforGeeks