Outrageously Large Neural Networks: The Sparsely-Gated Mixture-Of-Experts Layer

Re:Outrageously Large Neural Networks: The Sparsely-Gated Mi... by nl on Monday January 30, 2017 @05:03AM (#13519815)

Cloud GPUs are cost effective if you need to either fine-tune a pretrained network (eg, use pretrained ResNet/VGG/AlexNet for custom classes, ie[1]) or for inference, or if you don't want the upfront costs.

A 4GB GTX1050 is ~$180. A p2 instance on Amazon is $0.9/hour. The cost effectiveness depends on if you have a PC already.

[1] https://blog.keras.io/building-powerful-image-classification...

[ Parent | on HN ]

Re:Outrageously Large Neural Networks: The Sparsely-Gated Mi... by dTal on Monday January 30, 2017 @11:02AM (#13522250)

And the cost of your electricity, don't forget.

[ Parent | on HN ]

Re:Outrageously Large Neural Networks: The Sparsely-Gated Mi... by robotresearcher on Monday January 30, 2017 @04:42PM (#13525313)

Current mean cost to US domestic consumers of $0.12 kW/h, says Google.

TPD Specs: GTX 1050 75W GTX 1080 180W Titan 250W

Triple these numbers for the overall machine's PSU spec.

So even the Titan costs $0.03 per hour to run, or maybe $0.10 if the rest of the machine is flat-out.

edit: $0.19 kWh for San Francisco residents. Domestic rates.

[ Parent | on HN ]