Summary

In this chapter, we built our first machine learning use case based on a real-life scenario. After a brief introduction to the use case, we discovered what linear regression is and how it can be used to predict numerical values.

Before diving into actually developing the machine learning model, we learned that having a clear understanding of the data and checking its quality is fundamental to getting effective machine learning models. To start from a solid foundation, we leveraged the BigQuery public dataset, which collects information about all the rentals for a bike sharing service in New York City.

For training the model, we used different features to understand which variables are relevant to building our BigQuery ML model.

Then, we chose one machine learning model to carry on to the evaluation stage. In this phase, we used the BigQuery evaluation function to verify that the machine learning model could also effectively work on new rows outside the training dataset...