Understand the Python Code

First, Look at the Full Code

import numpy as np
from sklearn.cluster import KMeans

# Customer data
X = np.array([
    [1000, 2],
    [1200, 3],
    [1100, 2],
    [10000, 15],
    [11000, 17],
    [10500, 16]
])

# Create the model
model = KMeans(
    n_clusters=2,
    random_state=42
)

# Train the model
model.fit(X)

# Get cluster labels
labels = model.labels_

# Get cluster centers
centers = model.cluster_centers_

print("Labels:")
print(labels)

print("Centers:")
print(centers)

Now let's break this code into small pieces.

Import NumPy

import numpy as np

NumPy is a Python library used for working with numerical data and arrays.

We use it here to create our dataset.

Import KMeans

from sklearn.cluster import KMeans

This imports the KMeans algorithm from scikit-learn.

After this line, we can create a K-Means model in Python.

Create the Dataset

X = np.array([
    [1000, 2],
    [1200, 3],
    [1100, 2],
    [10000, 15],
    [11000, 17],
    [10500, 16]
])

The variable X contains our input data.

Each row represents one customer.

So:

Create the K-Means Model

model = KMeans(
    n_clusters=2,
    random_state=42
)

This creates a K-Means model.

We have not trained it yet.

What does n_clusters=2 mean?

We are telling K-Means:

What Is random_state=42?

random_state=42

K-Means uses random initialization when starting its cluster centers.

Setting random_state gives us reproducible results.

The number 42 is not special. You could use another fixed number.

Train the Model

model.fit(X)

This is one of the most important lines.

fit() tells K-Means to learn the cluster structure from our data.

Internally, K-Means repeatedly assigns points to nearby centroids and moves the centroids until the solution becomes stable.

Get the Cluster Labels

labels = model.labels_

After training, labels_ tells us which cluster each data point was assigned to.

For example, we might get:

[1 1 1 0 0 0]

This corresponds to:

Get the Cluster Centers

centers = model.cluster_centers_

cluster_centers_ contains the centroid of every cluster.

A centroid represents the average position of the data points belonging to that cluster.

Print the Results

print("Labels:")
print(labels)

print("Centers:")
print(centers)

print() simply displays the values in the terminal or notebook.

Labels:
[1 1 1 0 0 0]

Centers:
[[10500.    16. ]
 [ 1100.     2.33]]

The exact ordering of cluster labels can vary, so don't depend on Cluster 0 always being the lower-spending group or Cluster 1 always being the higher-spending group.

The Whole Code in Plain English

Don't Forget Feature Scaling

Our example uses spending and purchases, but these features have very different numerical scales.

Because K-Means uses distances, the spending feature can dominate the distance calculation.

In a real project, you would usually scale the numerical features before clustering when their scales differ substantially.

Check Your Understanding