CatBoost
Gradient boosting with native categorical support for regression
CatBoost Regressor handles categorical features natively using ordered target statistics, removing the need for manual encoding while delivering competitive accuracy.
When to use:
- Regression datasets with many categorical columns
- When minimal preprocessing is preferred
- Strong baseline for structured data
Input: Tabular data with the feature columns defined during training Output: Continuous predicted value
Model Settings (set during training, used at inference)
Iterations (default: 1000) Number of boosting rounds.
Learning Rate (default: auto) Step size for gradient updates.
Depth (default: 6) Symmetric tree depth.
L2 Leaf Reg (default: 3.0) L2 regularization on leaf values.
Loss Function (default: RMSE)
RMSE for mean squared error; MAE for mean absolute error; Huber for robust regression.
Inference Settings
No dedicated inference-time settings.