Deep Learning on AWS

(Before It's News)

This will be a very short post. But yesterday I spent some time trying to manually set up gpu on AWS.My reflections:

1. Use someone else’s AMI see http://erikbern.com/2015/11/12/installing-tensorflow-on-aws/ for example.
2. I had a lot of trouble updating cudnn from nvidia to work with the newest keras. I expect this to improve in the next version.
3. Use spot instances they are much cheaper!
4. There seems to be some cumbersome setup cost with modern deep learning frameworks. So using a GPU is a long way from being super easy.

• Enzyme Catalysis Enhanced Dark-field Imaging as a Novel Immunohistochemical Method
• Nanoscale Bio-Platforms for Living Cell Interrogation: Current Status and Future Perspectives
• More acupuncture Tooth Fairy science as 2015 approaches its end [Respectful Insolence]
• Hurrem Sultan, the Cheerful Rose of Suleiman I and a Powerful Woman of the Ottoman Empire
• Like Father, like Son: Altar shows heroic son of Hercules slaying a many-headed Hydra
• Unearthing the Ancient Port of Naukratis, a Bridge between Egyptian and Greek Civilization
• Cold Fusion Again
• Hands-off with Tesla’s Autopilot system
• Open Source “Baby Blue’ Project Could Finally Kill the Bluebook
• What animals are the best range buddies for your chickens?