Scribd
Upload
1 views

Vertopal.com_ML LAB 9

The document presents a Python script that implements a clustering algorithm using two normal distributions. It generates data points, randomly selects initial cluster centers, and iteratively assigns labels based on distance until convergence. Finally, it visualizes the clustering results using a scatter plot.

Uploaded by

vanshikasehrawat1085
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
1 views

Vertopal.com_ML LAB 9

The document presents a Python script that implements a clustering algorithm using two normal distributions. It generates data points, randomly selects initial cluster centers, and iteratively assigns labels based on distance until convergence. Finally, it visualizes the clustering results using a scatter plot.

Uploaded by

vanshikasehrawat1085
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

#06701192023 Sugandh ML LAB 9

import numpy as np
import matplotlib.pyplot as plt

# Generate two normal distributions


d1 = np.random.normal(loc=[2, 2], scale=1.0, size=(500, 2))
d2 = np.random.normal(loc=[8, 8], scale=1.0, size=(500, 2))

data = np.vstack((d1, d2)) # Combine both distributions

# Randomly select two initial points


indices = np.random.choice(data.shape[0], 2, replace=False)
x1, x2 = data[indices]

while True:
# Assign labels based on the closest center
distances_to_x1 = np.linalg.norm(data - x1, axis=1)
distances_to_x2 = np.linalg.norm(data - x2, axis=1)
labels = np.where(distances_to_x1 < distances_to_x2, 1, 2)

# Compute new means


new_x1 = data[labels == 1].mean(axis=0)
new_x2 = data[labels == 2].mean(axis=0)

# Check for convergence


if np.allclose(new_x1, x1) and np.allclose(new_x2, x2):
break

x1, x2 = new_x1, new_x2

# Plot results
plt.scatter(data[:, 0], data[:, 1], c=labels, cmap='bwr', alpha=0.6)
plt.scatter([x1[0], x2[0]], [x1[1], x2[1]], c='black', marker='x',
s=200, label='Centers')
plt.legend()
plt.xlabel('X')
plt.ylabel('Y')
plt.title('Clustering using Distance-based Labeling')
plt.show()

You might also like