Datasets
by @stats385
· last added 2026-09-24
· 5 items
[datasets]
Canonical vision benchmarks used throughout the course: MNIST, CIFAR-10, ImageNet.
*
ImageNet (overview)
Background on ILSVRC and why ImageNet mattered historically for deep learning.
*
ImageNet
Large-scale image database that powered AlexNet and the modern deep learning boom.
*
CIFAR-10
60k 32×32 color images, 10 classes. Standard small-scale vision benchmark for theory papers.
*
MNIST
60k train / 10k test handwritten digits, 28×28. The ‘hello world’ of classification.