Image Classification on the Cifar-100 Dataset.
The Cifar-100 dataset consists of 60000 32x32 color images in 100 categories.
We use the Resnet18 model implementation from torchvision.
We compare the top-1 accuracy. Our model reaches a top-1 accuracy of 77.9 ± 0.2%. The search grid used to find the best hyperparameters can be found here.