MachineLearning/xgboost_like library

🧰 Simplified XGBoost-like Regressor (stump + second-order)

A simplified, educational XGBoost-style regressor using decision stumps and second-order approximation (gradients & hessians) to compute optimal leaf weights. This implements regularization lambda and learning rate.

Contract:

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

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