MachineLearning/gradient_boosting library

⚡ Gradient Boosting Regressor (stumps)

A concise gradient boosting regressor using decision stumps (one-feature threshold learners) as weak learners. Uses squared-error loss and a learning rate. Designed for clarity and small datasets.

Contract:

  • Input: X (n x m), y numeric targets.
  • Output: GradientBoostingRegressor with fit and predict.

Time Complexity: O(n * m * n_estimators) Space Complexity: O(n_estimators)