[newb] prediction howto

[Neal Becker]

I'm new to statistical modeling. I have a task to use timeseries
measurements related to rain data. Given historical data taken in a
particular location, can we make a model such that looking at new
measurements over say, the last 10 minutes, can we predict the next x
minutes?

First I tried following the example in pyflux here
https://pyflux.readthedocs.io/en/latest/getting_started.html#step-five-analyse-and-predict

Then I realized I don't know how to apply this to predict from _new_ data.

Can anyone suggest an approach, and possibly some example?

Thanks,
Neal