Lab_ML(03/02/25)

import matplotlib.pyplot as plt
import numpy as np
from scipy.spatial.distance import euclidean, cityblock, chebyshev, minkowski, hamming, cosine, jaccard

# Define the two points for Euclidean, Manhattan, Chebyshev, and Minkowski
point1 = (1, 2)
point2 = (4, 6)

# Define binary vectors for Hamming
point1_binary = np.array([1, 0, 1, 0])
point2_binary = np.array([1, 1, 0, 0])

# Define vectors for Cosine
point1_vector = np.array([1, 2, 3])
point2_vector = np.array([4, 6, 8])

# Define sets for Jaccard and convert them to binary vectors
point1_set = set([1, 2, 3])
point2_set = set([3, 4, 5])
# Convert sets to binary vectors
all_elements = point1_set.union(point2_set)
point1_binary_vector = np.array([1 if x in point1_set else 0 for x in all_elements])
point2_binary_vector = np.array([1 if x in point2_set else 0 for x in all_elements])

# List of distance functions and their names
distance_functions = [
    ("Euclidean", euclidean, point1, point2),
    ("Manhattan", cityblock, point1, point2),
    ("Chebyshev", chebyshev, point1, point2),
    ("Minkowski (p=3)", lambda p1, p2: minkowski(p1, p2, p=3), point1, point2),
    ("Hamming", hamming, point1_binary, point2_binary),
    ("Cosine", cosine, point1_vector, point2_vector),
    ("Jaccard", jaccard, point1_binary_vector, point2_binary_vector)
]

# Calculate distances and store them
distances = []
for name, distance_func, p1, p2 in distance_functions:
    distance = distance_func(p1, p2)
    distances.append((name, distance))
    print(f"{name} distance:", distance)

# Plot the distances using a bar chart
fig, ax = plt.subplots(figsize=(10, 6))
distance_names = [d[0] for d in distances]
distance_values = [d[1] for d in distances]

bars = ax.bar(distance_names, distance_values, color='skyblue')
ax.set_xlabel('Distance Metrics')
ax.set_ylabel('Distance Values')
ax.set_title('Comparison of Different Distance Metrics')

# Set the tick positions and labels explicitly
ax.set_xticks(range(len(distance_names)))
ax.set_xticklabels(distance_names, rotation=45, ha='right')

# Add value labels on top of the bars
for bar in bars:
    yval = bar.get_height()
    ax.text(bar.get_x() + bar.get_width()/2, yval + 0.05, round(yval, 2), ha='center', va='bottom')

plt.tight_layout()
plt.show()