impossible → routine

Training ImageNet

Training an image classifier to state-of-the-art accuracy on ImageNet.

Impossible

AlexNet wins ILSVRC-2012 with a 15.3% top-5 test error, against 26.2% for the runner-up, after five to six days of training on two GTX 580 GPUs.

five to six days on two consumer GPUs

source

5 years, 8 months

Routine

fast.ai trains ImageNet to 93% accuracy in 18 minutes on 16 rented public cloud instances, for about $40.

18 minutes, about $40

source

The impossible end is dated to the opening day of NIPS 2012 (December 3–8, Lake Tahoe), where the AlexNet paper was presented.

Two notes on reading the receipts. The cited PDF gives two results, and this end quotes the second: 37.5% top-1 and 17.0% top-5 error on LSVRC-2010, and then "a winning top-5 test error rate of 15.3%, compared to 26.2% achieved by the second-best entry" for the ILSVRC-2012 competition the routine end's benchmark descends from. The NeurIPS HTML landing page for the same paper still serves an earlier draft of the abstract, giving 39.7% and 18.9% against 1.3 million images where the camera-ready PDF gives 37.5%, 17.0% and 1.2 million; the figures here are the PDF's, which is the version linked.

The two ends measure time and cost, not accuracy, and that is deliberate. The fast.ai post reports "93% accuracy" without saying whether it means top-1 or top-5, so the two accuracy figures are not directly comparable and no comparison between them is claimed.

All dated pairs