NBEATS

A Deep Learning model for zero-shot time-series forecastingCreated with DALLE There’s one thing that makes Time-Series Forecasting special.It was the only area of Data Science where Deep Learning and Transformers didn’t decisively outperform the other models.Let’s use the prestigious Makridakis M-competitions as a benchmark — a series of large-scale challenges that showcase the latest advances in the time-series forecasting area.In the fourth iteration of the competition, known as M4, the winning solution was ES-RNN , a…

From theory to practice, learn how N-BEATS works and apply it in a real-life forecasting project using PythonPhoto by Pablo de la Fuente on UnsplashIf, like me, you are interested in time series forecasting, chances are that you stumbled on the model N-BEATS. The model promises state-of-the-art results using a pure deep learning architecture. In other words, it does not need time-series specific components, like trend or seasonality.Chances are that you read the 2020 paper on N-BEATS written by Oreshkin et al. Although…