Time series forecasting with the functime library
Time series forecasting is a form of predictive analytics for time series data. It is to predict the future values based on historical data using statistical models. functime is a machine learning library for time series forecasting and feature extraction in Polars. It enables you to build time series forecasting models utilizing the Polars speed. functime is the Polars’ version of tsfresh, the popular time series feature extraction library.
In this recipe, we’ll cover how to build a simple time series forecasting model with the functime library, including feature extraction and plotting.
Important
As of the time of writing, Polars just upgraded to version 1.0.0 and functime has some compatibility issues with it. You may encounter an error after step 2; however, you can run those later steps with Polars version 0.20.31 (you’ll also need to change lf.collect_schema().names() to lf.columns for the code in...
