knn回归怎么调参，svr回归怎么调参，随机森林回归怎么调参，决策树回归怎么调参，给出python代码

在机器学习中，KNN、SVR（支持向量回归）、随机森林和决策树回归都有各自的超参数需要调优。以下是这四个模型的一些常见调参方法和Python代码示例：

### KNN回归 (KNeighborsRegressor)
- 超参数: n_neighbors（邻居数）、weights（距离加权）

from sklearn.neighbors import KNeighborsRegressor
from sklearn.model_selection import GridSearchCV

knn = KNeighborsRegressor()
params = {'n_neighbors': list(range(1, 21)), 'weights': ['uniform', 'distance']}
grid_knn = GridSearchCV(knn, params, cv=5)
grid_knn.fit(X_train, y_train)
best_knn = grid_knn.best_estimator_

### SVR回归 (SVC)
- 超参数: C（惩罚项系数）、gamma（内核函数参数）

from sklearn.svm import SVR
from sklearn.model_selection import GridSearchCV

svr = SVR()
params = {'C': [0.1, 1, 10], 'kernel': ['linear', 'poly', 'rbf'], 'gamma': ['scale', 'auto']}
grid_svr = GridSearchCV(svr, params, cv=5)
grid_svr.fit(X_train, y_train)
best_svr = grid_svr.best_estimator_

### 随机森林回归 (RandomForestRegressor)
- 超参数: n_estimators（树的数量）、max_depth（最大深度）

from sklearn.ensemble import RandomForestRegressor
from sklearn.model_selection import GridSearchCV

rf = RandomForestRegressor()
params = {'n_estimators': [10, 50, 100], 'max_depth': range(1, 21)}
grid_rf = GridSearchCV(rf, params, cv=5)
grid_rf.fit(X_train, y_train)
best_rf = grid_rf.best_estimator_

### 决策树回归 (DecisionTreeRegressor)
- 超参数: max_depth（最大深度）、min_samples_split（分割最小样本数）

from sklearn.tree import DecisionTreeRegressor
from sklearn.model_selection import GridSearchCV

dt = DecisionTreeRegressor()
params = {'max_depth': range(1, 21), 'min_samples_split': [2, 5, 10]}
grid_dt = GridSearchCV(dt, params, cv=5)
grid_dt.fit(X_train, y_train)
best_dt = grid_dt.best_estimator_