This repository contains implementations and training logs for both Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), along with visualizations used in our paper.
CNNs.ipynb– Contains all CNN architectures used in our experiments.Vision_Transformers.ipynb– Contains all Vision Transformer architectures used in our experiments.images– Includes visualizations and training logs featured in our paper (e.g. loss curves, accuracy plots, and references to architectures)
All datasets are automatically downloaded and loaded within the dataset class definitions. You do not need to manually download any files. The dataset will be retrieved the first time you run a training script.