This is a project for the Stanford Car Classification Challenge on Kaggle. The goal is to classify cars into 196 classes.
The dataset contains 16,185 car images distributed over 196 classes/brands. There are 8,144 images for training and 8,041 images for testing in this dataset. Each class roughly has a 50-50 split in the training and validation set. The dataset is available https://www.kaggle.com/datasets/jutrera/stanford-car-dataset-by-classes-folder
This folder contains results observed from transfer-learning the EfficientNet-b0 model on the Stanford Car dataset.