Project on exploring how different vision models “see” and analyze the images. We compare ConvNeXt, DeiT, and MLP-Mixer using Grad-CAM, attention maps, and saliency, with tools to run inference, analyze results, and visualize model focus regions.

This repository contains image classification experiments on CIFAR-10, comparing MLPs, ResNets, and MLP-Mixer architectures. Conducted as part of IFT6135 – Representation Learning, it studies how architectural choices affect accuracy, training dynamics, and practical trade-offs.

Deep Learning: Mlp, Resnet18, MlpMixer