Back to GalleryGradient Boosting
Gradient Boosting builds an ensemble of weak decision trees sequentially. Each new tree is fitted to the pseudo-residuals (gradients of the loss function) of the existing ensemble, performing gradient descent in function space.
Key Parameters & Visual Influence:
• Number of Trees: More trees capture complex residual patterns, but risk overfitting.
• Learning Rate: Scales the contribution of each tree. Smaller values slow down training but improve generalizability.
• Subsample Ratio: Trains trees on random fractions of the dataset, reducing variance.
Dataset Studio
CSV format: Headers in first row, last column = label/target. All values must be numeric.
Example: x1, x2, label