Image-Classification

This project's main idea is to classify the digits (0-9) using the digit dataset and determine whether a given image is a face or not using Face dataset. The images are already pre-processed through an edge detector. The classification is done using 3 Algorithms 1. Naive Bayes Classifier 2. Perceptron Classifier 3. MIRA(Margin Infused Relaxed Algorithm) Classifier using the Edge Features. The results are then analyzed based on classification accuracy vs the size of the training sets.