UML with scikit-learn on Vertex AI
In this section, we will start using our Vertex AI Workbench notebook to train models. We will begin with a relatively simple use case in which we will create an unsupervised model to discover clustering patterns in our data. Before we dive into the code, we will first take a minute to discuss the clustering algorithm we will use in this section, which is called K-means.
K-means
You may remember that we discussed unsupervised learning (UL) mechanisms such as clustering in Chapter 1. Remember that in clustering, data points are grouped together based on similarities between features or characteristics that are observed by the model. Figure 5.5 provides a visual representation of this concept, showing the input data on the left and the resulting data clusters on the right:
Figure 5.5: Clustering
K-means is an example of a clustering algorithm, and it is categorized as a centroid-based clustering algorithm. What this...
