Low-Code AI

Chapter 6. Using BigQuery ML to Train a Linear Regression Model

In this chapter you learn how to build a linear regression model and a neural network model from scratch to forecast power plant production. You perform this task using SQL for data analysis, Jupyter Notebook for data exploration, and BigQuery Machine Learning (BigQuery ML) for training the ML model. Along the way, you learn new techniques for understanding your data in preparation for ML and how to apply this knowledge in improving your model performance.

The Business Use Case: Power Plant Production

Your goal in this project will be to predict the net hourly electrical energy output for a combined cycle power plant (CCPP) given the weather conditions near the plant at the time.

A CCPP is composed of gas turbines, steam turbines, and heat recovery steam generators. The electricity is generated by gas and steam turbines, which are combined in one cycle, and is transferred from one turbine to another. While the vacuum is collected from the steam turbine, the other three ambient variables (temperature, ambient pressure, and relative humidity) affect the gas turbine performance.

The dataset in this section contains data points collected from a CCPP over a six-year period (2006–2011) when the power plant was set to work with a full load. The data is aggregated per hour, though the exact hour for the recorded weather conditions and energy production is not supplied in the dataset. From a practical viewpoint, this means ...