Model Evaluation Metrics

# CONFUSION MATRIX BASICS

import numpy as np
import matplotlib.pyplot as plt
from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay
from sklearn.datasets import make_classification
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression

# Create imbalanced dataset (like fraud detection)
X, y = make_classification(
    n_samples=1000, n_features=10,
    n_informative=5, n_redundant=2,
    weights=[0.95, 0.05],  # 95% negative, 5% positive
    random_state=42
)

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

print("="*60)
print("IMBALANCED DATASET EXAMPLE")
print("="*60)
print(f"Class distribution: {np.bincount(y)}")
print(f"Positive class (fraud): {sum(y)} ({sum(y)/len(y)*100:.1f}%)")

# Train simple model
model = LogisticRegression(random_state=42)
model.fit(X_train, y_train)
y_pred = model.predict(X_test)

# Confusion Matrix
cm = confusion_matrix(y_test, y_pred)
print(f"\nConfusion Matrix:")
print(f"                 Predicted")
print(f"                 Neg    Pos")
print(f"Actual Neg  |  {cm[0,0]:4d}   {cm[0,1]:4d}  | TN, FP")
print(f"Actual Pos  |  {cm[1,0]:4d}   {cm[1,1]:4d}  | FN, TP")

# Accuracy (misleading!)
accuracy = (cm[0,0] + cm[1,1]) / cm.sum()
print(f"\nAccuracy: {accuracy:.1%}")
print(f"⚠️ Looks good but may miss most frauds!")

# Visualize
fig, ax = plt.subplots(figsize=(8, 6))
ConfusionMatrixDisplay(cm, display_labels=['Not Fraud', 'Fraud']).plot(ax=ax, cmap='Blues')
plt.title('Confusion Matrix - Fraud Detection')
plt.show()