ML_5.py

import numpy as np
import matplotlib.pyplot as plt
from collections import Counter

# Generate data
np.random.seed(42)  # Added for reproducibility
data = np.random.rand(40)
train_data, test_data = data[:20], data[20:]
train_labels = ["Class1" if x <= 0.5 else "Class2" for x in train_data]

# Define KNN classifier
def knn_predict(train_data, train_labels, test_data, k):
    return [Counter(sorted([(abs(x-point), label) for x, label in zip(train_data, train_labels)], 
                    key=lambda x: x[0])[:k]).most_common(1)[0][0] for point in test_data]

# Test with multiple k values
k_values = [1, 2, 3, 4, 5, 20, 30]
results = {k: knn_predict(train_data, train_labels, test_data, k) for k in k_values}

# Print results
print("--- k-Nearest Neighbors Classification ---")
for k, labels in results.items():
    print(f"\nResults for k = {k}:")
    for i, (point, label) in enumerate(zip(test_data, labels)):
        print(f"Point x{i+21} (value: {point:.4f}) is classified as {label}")

# Visualize results
for k, classified_labels in results.items():
    plt.figure(figsize=(10, 8))
    # Plot training data
    plt.scatter(train_data, [0]*len(train_data), 
                c=["blue" if l == "Class1" else "red" for l in train_labels], 
                marker="o", label="Training Data")
    # Plot test data by class
    class_points = {"Class1": [], "Class2": []}
    for point, label in zip(test_data, classified_labels):
        class_points[label].append(point)
    
    for label, color in [("Class1", "blue"), ("Class2", "red")]:
        if class_points[label]:
            plt.scatter(class_points[label], [1]*len(class_points[label]), 
                        c=color, marker="x", label=f"{label} (Test)")
            
    plt.title(f"k-NN Classification Results for k = {k}")
    plt.xlabel("Data Points"), plt.ylabel("Classification Level")
    plt.legend(), plt.grid(True)
    plt.show()