Check out my rundown of all-things-Data-Science in November. Especially Amazon had some nice machine learning related announcements.

Google announcements

This month Google announced the release of AI Hub - a one-stop destination for plug-and-play ML content. Also they set the pricing of the V2 TPU pods at 4.95 USD per hour.

Amazon's Machine Learning University now available for all

This is the news that personally got me excited the most. With 30 self-paced online courses and 45 hours of video materials they aim to educate not only data scientists, but also developers, data platform engineers and even business professionals.

Amazon has also:

  • Made it easier to collaborate on building models in AWS by adding Git integration to SageMaker.
  • Made QuickSight more intelligent by adding some ML functionality.
  • Added search to SageMaker, making it easier to find that one model you made a long time ago. Of course this is only valuable if you are a heavy user.

KDNugget Interpretability Poll

KDNuggets conducted a poll amongst its readers asking about interpretability of machine learning models. Do you find the result surprising?

Dating pool based on creepiness rule

Although the analysis doesn't seem to have much practical use, there's a nice blog post visualizing the 'half-your-age-plus-seven' rule on the US dating pool With my favourite quote: It’s on Wikipedia, so it’s basically law.