Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: Machine learning – the easy way
4 points by kenjackson on July 6, 2015 | hide | past | favorite | 1 comment
I just watched this TED talk (https://www.ted.com/talks/rajiv_maheswaran_the_math_behind_basketball_s_wildest_moves) and at one point in the talk he talks about using machine learning to determine what is and isn't a pick and rolls That makes sense, that's exactly what I would have said too. Except I wouldn't have actually done it.

I feel like I often run into situations like these where I think, "makes sense to use machine learning here". I'd love to just drop my data into some service that then spits out code that will classify future data. I trust the AI experts will build a much better system (and update it) better than I could.

Are there services that do this?




I think Azure ML Studio has some of the aspects of what you are looking for...https://studio.azureml.net/

There is a small amount of work involved, but you can mostly browse the gallery, find a situation that fits your needs, and even publish a REST Webservice with a single click that your applications can call for a classification task. It makes ML very easy (they do all of the hard work of building decision tree algorithms, neural networks, etc. -- all you do is drag and drop modules).




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: