Scikit中使用Grid_Search来获取模型的最佳参数

1. grid search是用来寻找模型的最佳参数

先导入一些依赖包

from sklearn.ensemble import GradientBoostingClassifier
from sklearn.grid_search import GridSearchCV
from sklearn import metrics
import numnpy as np
import pandas as pd

2. 设置要查找的参数

params={'learning_rate':np.linspace(0.05,0.25,5), 'max_depth':[x for x in range(1,8,1)], 'min_samples_leaf':[x for x in range(1,5,1)], 'n_estimators':[x for x in range(50,100,10)]}

3. 设置模型和评价指标，开始用不同的参数训练模型

clf = GradientBoostingClassifier()
grid = GridSearchCV(clf, params, cv=10, scoring="f1")
grid.fit(X, y)

scoring所有可能情况如下:

• Classification

scoring	function	comment
accuracy	metrics.accuracy_score
average_precision	metrics.average_precision_score
f1	metrics.f1_score	for binary targets
f1_micro	metrics.f1_score	micro-averaged
f1_macro	metrics.f1_score	macro-averaged
f1_weighted	metrics.f1_score	weighted average
f1_samples	metrics.f1_score	by multilabel sample
neg_log_loss	metrics.log_loss	requires predict_proba support
precision etc.	metrics.precision_score	suffixes apply as with “f1”
recall etc.	metrics.recall_score	suffixes apply as with “f1”
roc_auc	metrics.roc_auc_score

• Clustering

scoring	function	comment
adjusted_rand_score	metrics.adjusted_rand_score

• Regression

scoring	function	comment
neg_mean_absolute_error	metrics.mean_absolute_error
neg_mean_squared_error	metrics.mean_squared_error
neg_median_absolute_error	metrics.median_absolute_error
r2	metrics.r2_score

4. 查看最佳分数和最佳参数

grid.best_score_    #查看最佳分数(此处为f1_score)
grid.best_params_   #查看最佳参数

5. 获取最佳模型

grid.best_estimator_

6. 利用最佳模型来进行预测

best_model=grid.best_estimator_
predict_y=best_model.predict(Test_X)
metrics.f1_score(y, predict_y)