khuntzberry

K-Nearest Neighbors & Model Evaluation

In this video we cover the basics of the most simple machine learning model K-Nearest Neighbors. I also explain what a hyperparameter is and how to use a grid search in scikit-learn to test multiple models. Finally, I cover machine learning model evaluation metrics such as accuracy, recall, precision, and more. We wrap up with ROC AUC curves and how to compare the performance of multiple models in one visualization.

Code Available Here and Data Available Here

Suggested Reading:

  • Machine Learning with R, 3rd Edition, Chapter 3 (p. 65-88)
  • Python Machine Learning, 3rd Edition, Chapter 3 (p. 103-108) and Chapter 6 (p. 191-222)
  • Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd Edition, Chapter 2 (p. 72-80) and Chapter 3 (p. 85-108)
  • Distances in Machine Learning