Training the K-Means clustering model

In this section, we'll create two different K-Means machine learning models. The first model will be created using taxi_miles_per_minute as a training dataset, while the second will include also tot_income as a feature and will leverage taxi_speed_and_income. Let's proceed as follows:

1. As a first step, let's start training a machine learning model named clustering_by_speed by running the following code:

   CREATE OR REPLACE MODEL `07_chicago_taxi_drivers.clustering_by_speed`
   OPTIONS(model_type='kmeans', num_clusters=3, kmeans_init_method = 'KMEANS++') AS
     SELECT * EXCEPT (taxi_id)
     FROM `07_chicago_taxi_drivers.taxi_miles_per_minute`;

   The first lines of the SQL statement are composed of the CREATE OR REPLACE MODEL keywords, followed by the identifier of the `07_chicago_taxi_drivers.clustering_by_speed` machine learning model and the OPTIONS clause.

   Now, let's take a look at the options...