“So the big aim here is obviously to predict the rain in the future. We’ll build a system that does just that from scratch!
First a few words about this project the motivation, the technologies involved and the end product that we’re going to build.
So the big aim here is obviously to predict the rain in the future (we’ll try 6 hours). The prediction will be a yes or no (boolean in programming terms). I’ve searched from tutorials on this matter and I haven’t found one that is complete in every way. So mine will take this to a whole new approach and get into every aspect of it. For that we’re going to:
- build the weather station ourselves. The station should be completely off grid with a solar panel, and a extremely low power mode (a few dozens microamps hour)
- program the station so it gather data and transmits it every ten minutes to a base station
- collect the data on the base station and store it (in a database)
- using neural networks (Keras library) and other Python libraries like pandas filter, clean and preprocess the data, then feed it to a neural network to train a “model” for predicting that it shall or not rain
- finally predict whether or not it will rain on the next 6 hours and notify users via email”