Summary

In this chapter, we discussed how the quality of our features, and the ratio of features to observations in our dataset, influences how our algorithms learn from our data. We discussed challenges that can occur when our dataset contains many features and how to address those challenges by using mechanisms such as dimensionality reduction. We dived into details on dimensionality reduction techniques such as feature selection and feature projection, including algorithms such as PCA, LDA, and t-SNE, and we looked at examples of how to use some of these algorithms using hands-on activities.

Next, we dived into feature engineering techniques in which we augmented a source dataset to create new features that contained information that was not readily available in the original dataset. Finally, we dived into Vertex AI Feature Store to learn about how we can use that service to store and serve our engineered feature sets.

In the next chapter, we will shift our focus away from...