CNN is widely used in image classification. In this project, we are going to classify car images.

stanford car dataset from https://ai.stanford.edu/~jkrause/cars/car_dataset.html

download car dataset, prepare the data for the model.

1. download training image http://imagenet.stanford.edu/internal/car196/cars_train.tgz uncompress to ./data/cars_train
2. download testing image http://imagenet.stanford.edu/internal/car196/cars_test.tgz uncompress to ./data/cars_test
3. download devkit https://ai.stanford.edu/~jkrause/cars/car_devkit.tgz uncompress to ./data/devkit
4. download test annotation with class label http://imagenet.stanford.edu/internal/car196/cars_test_annos_withlabels.mat move it to ./data/devkit
5. (optional) download bounding box annotations for all images http://imagenet.stanford.edu/internal/car196/cars_annos.mat move it to ./data/devkit
6. run data_prepare.py to prepare the training and testing data for the model (you may need modify some paths in the file)

run python train.py -t /path/to/car_dataset/train/ -v /path/to/car_dataset/test/ -m vgg16 -s car196 -e 20 -n 196