GBDT之GradientBoostingRegressor参数详解以及调参
- 一、参数、属性及方法
-
- 1、参数
-
- (1)loss
- (2)learning_rate
- (3)n_estimators
- (4)subsample
- (5)criterion
- (6)min_samples_split
- (7)min_samples_leaf
- (8)min_weight_fraction_leaf
- (9)max_depth
- (10)min_impurity_decrease
- (11)init
- (12)random_state
- (13)max_features
- (14)alpha
- (15)verbose
- (16)max_leaf_nodes
- (17)warm_start
- (18)validation_fraction
- (19)n_iter_no_change
- (20)tol
- (21)ccp_alphanon-negative
- 2、属性
- 3、方法
- 代码:波士顿房价为例
- 二、参数优化
sklearn官网:
https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.GradientBoostingRegressor.html#sklearn.ensemble.GradientBoostingRegressor
一、参数、属性及方法
1、参数
class sklearn.ensemble.GradientBoostingRegressor(*, loss=‘squared_error’, learning_rate=0.1, n_estimators=100,
subsample=1.0, criterion=‘friedman_mse’, min_samples_split=2, min_samples_leaf&#