ML之xgboost：利用xgboost算法(sklearn+GridSearchCV)训练mushroom蘑菇数据集(22+1,6513+1611)来预测蘑菇是否毒性(二分类预测)
目录
输出结果
正在更新……
设计思路
正在更新……
核心代码
from sklearn.grid_search import GridSearchCV
param_test = { 'n_estimators': range(1, 51, 1)}
clf = GridSearchCV(estimator = bst, param_grid = param_test, cv=5)
clf.fit(X_train, y_train)
clf.grid_scores_, clf.best_params_, clf.best_score_
grid_scores_mean= [0.90542, 0.94749, 0.90542, 0.94749, 0.90573, 0.94718,
0.90542, 0.94242, 0.94473, 0.97482, 0.94887, 0.97850,
0.97298, 0.97850, 0.97298, 0.97850, 0.97850, 0.97850,
0.97850, 0.97850, 0.97850, 0.97850, 0.97850, 0.97850,
0.97850, 0.97804, 0.97774, 0.97835, 0.98296, 0.98419,
0.98342, 0.98372, 0.98419, 0.98419, 0.98419, 0.98419,
0.98419, 0.98419, 0.98419, 0.98419, 0.98419, 0.98419,
0.98419, 0.98419, 0.98419, 0.98419, 0.98419, 0.98419,
0.98419 ]
grid_scores_std = [0.08996, 0.07458, 0.08996, 0.07458, 0.09028, 0.07436,
0.08996, 0.07331, 0.07739, 0.02235, 0.07621, 0.02387,
0.03186, 0.02387, 0.03186, 0.02387, 0.02387, 0.02387,
0.02387, 0.02387, 0.02387, 0.02387, 0.02387, 0.02387,
0.02387, 0.02365, 0.02337, 0.02383, 0.01963, 0.02040,
0.01988, 0.02008, 0.02040, 0.02040, 0.02040, 0.02040,
0.02040, 0.02040, 0.02040, 0.02040, 0.02040, 0.02040,
0.02040, 0.02040, 0.02040, 0.02040, 0.02040, 0.02040,
0.02040 ]
#7-CrVa交叉验证曲线可视化
import matplotlib.pyplot as plt
x = range(0,len(grid_scores_mean))
y1 = grid_scores_mean
y2 = grid_scores_std
Xlabel = 'n_estimators'
Ylabel = 'value'
title = 'mushroom datase: xgboost(sklearn+GridSearchCV) model'
plt.plot(x,y1,'r',label='Mean') #绘制mean曲线
plt.plot(x,y2,'g',label='Std') #绘制std曲线
plt.rcParams['font.sans-serif']=['Times New Roman'] #手动添加中文字体，或者['font.sans-serif'] = ['FangSong'] SimHei
#myfont = matplotlib.font_manager.FontProperties(fname='C:/Windows/Fonts/msyh.ttf') #也可以指定win系统字体路径
plt.rcParams['axes.unicode_minus'] = False #对坐标轴的负号进行正常显示
plt.xlabel(Xlabel)
plt.ylabel(Ylabel)
plt.title(title)
plt.legend(loc=1)
plt.show()
更多输出
GridSearchCV time： 79.7655139499154
clf.grid_scores_： [mean: 0.90542, std: 0.08996, params: {'n_estimators': 1}, mean: 0.94749, std: 0.07458, params: {'n_estimators': 2},
mean: 0.90542, std: 0.08996, params: {'n_estimators': 3}, mean: 0.94749, std: 0.07458, params: {'n_estimators': 4},
mean: 0.90573, std: 0.09028, params: {'n_estimators': 5}, mean: 0.94718, std: 0.07436, params: {'n_estimators': 6},
mean: 0.90542, std: 0.08996, params: {'n_estimators': 7}, mean: 0.94242, std: 0.07331, params: {'n_estimators': 8},
mean: 0.94473, std: 0.07739, params: {'n_estimators': 9}, mean: 0.97482, std: 0.02235, params: {'n_estimators': 10},
mean: 0.94887, std: 0.07621, params: {'n_estimators': 11}, mean: 0.97850, std: 0.02387, params: {'n_estimators': 12},
mean: 0.97298, std: 0.03186, params: {'n_estimators': 13}, mean: 0.97850, std: 0.02387, params: {'n_estimators': 14},
mean: 0.97298, std: 0.03186, params: {'n_estimators': 15}, mean: 0.97850, std: 0.02387, params: {'n_estimators': 16},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 17}, mean: 0.97850, std: 0.02387, params: {'n_estimators': 18},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 19}, mean: 0.97850, std: 0.02387, params: {'n_estimators': 20},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 21}, mean: 0.97850, std: 0.02387, params: {'n_estimators': 22},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 23}, mean: 0.97850, std: 0.02387, params: {'n_estimators': 24},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 25}, mean: 0.97804, std: 0.02365, params: {'n_estimators': 26},
mean: 0.97774, std: 0.02337, params: {'n_estimators': 27}, mean: 0.97835, std: 0.02383, params: {'n_estimators': 28},
mean: 0.98296, std: 0.01963, params: {'n_estimators': 29}, mean: 0.98419, std: 0.02040, params: {'n_estimators': 30},
mean: 0.98342, std: 0.01988, params: {'n_estimators': 31}, mean: 0.98372, std: 0.02008, params: {'n_estimators': 32},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 33}, mean: 0.98419, std: 0.02040, params: {'n_estimators': 34},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 35}, mean: 0.98419, std: 0.02040, params: {'n_estimators': 36},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 37}, mean: 0.98419, std: 0.02040, params: {'n_estimators': 38},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 39}, mean: 0.98419, std: 0.02040, params: {'n_estimators': 40},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 41}, mean: 0.98419, std: 0.02040, params: {'n_estimators': 42},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 43}, mean: 0.98419, std: 0.02040, params: {'n_estimators': 44},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 45}, mean: 0.98419, std: 0.02040, params: {'n_estimators': 46},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 47}, mean: 0.98419, std: 0.02040, params: {'n_estimators': 48},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 49}, mean: 0.98419, std: 0.02040, params: {'n_estimators': 50}]
clf.best_params_： {'n_estimators': 30}
clf.best_score_： 0.9841854752034392[mean: 0.90542, std: 0.08996, params: {'n_estimators': 1},
mean: 0.94749, std: 0.07458, params: {'n_estimators': 2},
mean: 0.90542, std: 0.08996, params: {'n_estimators': 3},
mean: 0.94749, std: 0.07458, params: {'n_estimators': 4},
mean: 0.90573, std: 0.09028, params: {'n_estimators': 5},
mean: 0.94718, std: 0.07436, params: {'n_estimators': 6},
mean: 0.90542, std: 0.08996, params: {'n_estimators': 7},
mean: 0.94242, std: 0.07331, params: {'n_estimators': 8},
mean: 0.94473, std: 0.07739, params: {'n_estimators': 9},
mean: 0.97482, std: 0.02235, params: {'n_estimators': 10},
mean: 0.94887, std: 0.07621, params: {'n_estimators': 11},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 12},
mean: 0.97298, std: 0.03186, params: {'n_estimators': 13},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 14},
mean: 0.97298, std: 0.03186, params: {'n_estimators': 15},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 16},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 17},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 18},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 19},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 20},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 21},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 22},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 23},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 24},
mean: 0.97850, std: 0.02387, params: {'n_estimators': 25},
mean: 0.97804, std: 0.02365, params: {'n_estimators': 26},
mean: 0.97774, std: 0.02337, params: {'n_estimators': 27},
mean: 0.97835, std: 0.02383, params: {'n_estimators': 28},
mean: 0.98296, std: 0.01963, params: {'n_estimators': 29},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 30},
mean: 0.98342, std: 0.01988, params: {'n_estimators': 31},
mean: 0.98372, std: 0.02008, params: {'n_estimators': 32},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 33},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 34},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 35},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 36},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 37},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 38},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 39},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 40},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 41},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 42},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 43},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 44},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 45},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 46},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 47},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 48},
mean: 0.98419, std: 0.02040, params: {'n_estimators': 49}]
grid_scores_ = [mean: 0.90542, std: 0.08996,
mean: 0.94749, std: 0.07458,
mean: 0.90542, std: 0.08996,
mean: 0.94749, std: 0.07458,
mean: 0.90573, std: 0.09028,
mean: 0.94718, std: 0.07436,
mean: 0.90542, std: 0.08996,
mean: 0.94242, std: 0.07331,
mean: 0.94473, std: 0.07739,
mean: 0.97482, std: 0.02235,
mean: 0.94887, std: 0.07621,
mean: 0.97850, std: 0.02387,
mean: 0.97298, std: 0.03186,
mean: 0.97850, std: 0.02387,
mean: 0.97298, std: 0.03186,
mean: 0.97850, std: 0.02387,
mean: 0.97850, std: 0.02387,
mean: 0.97850, std: 0.02387,
mean: 0.97850, std: 0.02387,
mean: 0.97850, std: 0.02387,
mean: 0.97850, std: 0.02387,
mean: 0.97850, std: 0.02387,
mean: 0.97850, std: 0.02387,
mean: 0.97850, std: 0.02387,
mean: 0.97850, std: 0.02387,
mean: 0.97804, std: 0.02365,
mean: 0.97774, std: 0.02337,
mean: 0.97835, std: 0.02383,
mean: 0.98296, std: 0.01963,
mean: 0.98419, std: 0.02040,
mean: 0.98342, std: 0.01988,
mean: 0.98372, std: 0.02008,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040,
mean: 0.98419, std: 0.02040 ]