Microsoft just announced yesterday the release of ML.NET, an open source machine learning framework for .Net.
I'm excited by the possibilities this brings, and will be spending some time over the next several weeks trying this out combined with the AF SDK as the means of extracting source data from PI. Whilst there are plenty of options out there for machine learning based advanced analytics, a framework like this calls to my inner geek.
As I build some familiarity with this I will post further updates and examples based on analysing data in my development PI system.
In the meantime, I'm curious as to whether other community members would look at this as an analytics option in future projects, especially as a (potentially) lower cost option.