A Downsampled Variant of ImageNet as an Alternative to the CIFAR datasets

TL;DR

This paper introduces downsampled variants of ImageNet at 16x16, 32x32, and 64x64 pixels, enabling faster experiments while maintaining similar dataset characteristics, as an alternative to CIFAR datasets.

Contribution

It presents a new set of downsampled ImageNet datasets with the same classes and images, facilitating faster research without losing key properties.

Findings

Downsampled datasets enable faster training and experimentation.

Characteristics of downsampled datasets remain similar to original ImageNet.

Scripts and datasets are publicly available for reproducibility.

Abstract

The original ImageNet dataset is a popular large-scale benchmark for training Deep Neural Networks. Since the cost of performing experiments (e.g, algorithm design, architecture search, and hyperparameter tuning) on the original dataset might be prohibitive, we propose to consider a downsampled version of ImageNet. In contrast to the CIFAR datasets and earlier downsampled versions of ImageNet, our proposed ImageNet32$\times$32 (and its variants ImageNet64$\times$64 and ImageNet16$\times$16) contains exactly the same number of classes and images as ImageNet, with the only difference that the images are downsampled to 32$\times$32 pixels per image (64$\times$64 and 16$\times$16 pixels for the variants, respectively). Experiments on these downsampled variants are dramatically faster than on the original ImageNet and the characteristics of the downsampled datasets with respect to optimal hyperparameters appear to remain similar. The proposed datasets and scripts to reproduce our results are available at http://image-net.org/download-images and https://github.com/PatrykChrabaszcz/Imagenet32_Scripts