Tabular regression

In Chapter 5, Time Series Forecasting as Regression, we saw how we can convert a time series problem into a standard regression problem with temporal embedding and time delay embedding. In Chapter 6, Feature Engineering for Time Series Forecasting, we have already created the necessary features for the household energy consumption dataset we have been working on, and in Chapter 8, Forecasting Time Series with Machine Learning Models, Chapter 9, Ensembling and Stacking, and Chapter 10, Global Forecasting Models, we used traditional machine learning (ML) models to create a forecast.

Just as we used standard ML models for forecasting, we can also use DL models built for tabular data using the feature-engineered dataset we have created. We already talked about data-driven methods and how they are better when given larger amounts of data. DL models take that paradigm even further and enable us to learn highly data-driven models. One of the advantages of using a DL...