The index to the articles in this series is found here.

In the last message, we talked about looking at the total weight of rain pixels in a training entry, since ones with a low total weight are less interesting to us. I decided that this total weight was something that would be useful to have in the header of the intermediate binary file. So, I’ve bumped the version number of that file, and made appropriate adjustments. At this point, I figured it might get confusing to follow along as I make edits to source code, so I’ve set up a github repository for the python files I’ve been writing here. You can find it at: https://github.com/ChristopherNeufeld/rain-predictor

I’ll update the first post to mention the github source.

UPDATE #1 (2019-08-23): Included a link to an index of articles in this series.