Model Evaluation Metrics

Duration: 5 min

This module delves into the various metrics used to evaluate supervised learning models, such as Linear Regression, Logistic Regression, Decision Trees, Random Forests, SVM, and Gradient Boosting. Understanding these metrics is crucial for assessing model performance and making informed decisions in machine learning projects.

Mean Squared Error (MSE) for Regression Models

Mean Squared Error (MSE) is a common metric for evaluating regression models. It measures the average squared difference between the actual and predicted values. Lower MSE values indicate better model performance. MSE is sensitive to outliers, which can significantly affect the model evaluation.

import numpy as np
from sklearn.metrics import mean_squared_error

# Actual values
y_true = np.array([3, -0.5, 2, 7])
# Predicted values
y_pred = np.array([2.5, 0.0, 2, 8])

# Calculate MSE
mse = mean_squared_error(y_true, y_pred)
mse

Try it in Google Colab:

0.375

Accuracy for Classification Models

Accuracy is a straightforward metric for evaluating classification models. It represents the proportion of correct predictions out of the total predictions. While easy to interpret, accuracy can be misleading for imbalanced datasets, where the majority class dominates the metric.

from sklearn.metrics import accuracy_score

# Actual labels
y_true = [0, 1, 1, 0]
# Predicted labels
y_pred = [1, 1, 1, 0]

# Calculate accuracy
accuracy = accuracy_score(y_true, y_pred)
accuracy

💡 Tip: When dealing with imbalanced datasets, consider using additional metrics like precision, recall, and F1-score alongside accuracy to get a comprehensive evaluation of your classification model.

Key Concepts

Check Your Understanding