from Hacker News

Using Google Cloud AutoML to Classify Poisonous Australian Spiders

by mattfrasernz on 3/13/18, 10:28 PM with 41 comments

  • by michaelgreen on 3/14/18, 12:36 AM

    This is huge, and it's only an alpha. I begun reading about AutoML/Neural Architecture searches around ~year ago and something I've been thinking about is:

    Why doesn't this just move the optimization problem? Aren't you now just optimizing your DeepRL network rather than the network you're trying to optimize?

  • by Zhenya on 3/14/18, 12:21 AM

    Something like this could actually save lives. Take a photo of what bit you, then the app could provide an answer (with a confidence level) of "do I need to go to the hospital" and provide a few example apps for various matches with level of danger.

    Thanks for sharing!

  • by bastih on 3/14/18, 10:38 AM

    I like the content of this blog post, but the use of stock photos was really off-putting and I had to almost force myself to keep on reading.
  • by juskrey on 3/14/18, 5:16 AM

    Isn't that something available in, say, Mathematica in 3 lines of code? https://wolfram.com/language/11/image-and-signal-processing/...
  • by gajju3588 on 3/14/18, 8:13 AM

    Auto labeling would be way forward for Supervised Algorithms. Get some data to annotate from your team, and tag rest of them using auto-labeling. https://dataturks.com/ could be such player, Not sure how will these survive in front of Google.
  • by thejosh on 3/14/18, 10:02 AM

    How long does it take to get access to AutoML?
  • by rrmoelker on 3/14/18, 7:19 AM

    Does anyone know a bit more about the pricing? All I can find online is Google learning infrastructure cost.
  • by neurostimulant on 3/14/18, 6:28 AM

    Is it possible to export the trained model from AutoML for offline use?
  • by dzhiurgis on 3/14/18, 3:50 AM

    If pictures are already from Google, wouldn't it be simpler to post image to Google's Vision API and then grep results against the list of poisonous spiders?
  • by skeleton on 3/14/18, 7:46 AM

    I'm surprised someone is yet to point out the error of using the word poisonous instead of venomous. I suppose it's not important to the article.