袋装决策树
袋装决策树详细操作教程
我们知道，装袋合奏方法与方差较大的算法配合使用效果很好，因此，最好的方法是决策树算法。在以下Python配方中，我们将通过在Simaarn糖尿病数据集上使用sklearn的BaggingClassifier函数和DecisionTreeClasifier（分类和回归树算法）来构建袋装决策树集成模型。
首先，按如下所示导入所需的程序包-
# Filename : example.py
# Copyright : 2020 By Lidihuo
# Author by : www.lidihuo.com
# Date : 2020-08-27
from pandas import read_csv
from sklearn.model_selection import KFold
from sklearn.model_selection import cross_val_score
from sklearn.ensemble import BaggingClassifier
from sklearn.tree import DecisionTreeClassifier
现在，我们需要像前面的示例中一样加载Pima糖尿病数据集-
# Filename : example.py
# Copyright : 2020 By Lidihuo
# Author by : www.lidihuo.com
# Date : 2020-08-27
path = r"C:\pima-indians-diabetes.csv"
headernames = ['preg', 'plas', 'pres', 'skin', 'test', 'mass', 'pedi', 'age', 'class']
data = read_csv(path, names=headernames)
array = data.values
X = array[:,0:8]
Y = array[:,8]
接下来，如下所示进行10倍交叉验证的输入-
# Filename : example.py
# Copyright : 2020 By Lidihuo
# Author by : www.lidihuo.com
# Date : 2020-08-27
seed = 7
kfold = KFold(n_splits = 10, random_state = seed)
cart = DecisionTreeClassifier()
我们需要提供要建造的树木数量。在这里，我们正在建造150棵树-
# Filename : example.py
# Copyright : 2020 By Lidihuo
# Author by : www.lidihuo.com
# Date : 2020-08-27
num_trees = 150
接下来，在以下脚本的帮助下构建模型-
# Filename : example.py
# Copyright : 2020 By Lidihuo
# Author by : www.lidihuo.com
# Date : 2020-08-27
model = BaggingClassifier(base_estimator = cart, n_estimators = num_trees, random_state = seed)
计算并打印结果如下-
# Filename : example.py
# Copyright : 2020 By Lidihuo
# Author by : www.lidihuo.com
# Date : 2020-08-27
results = cross_val_score(model, X, Y, cv=kfold)
print(results.mean())
输出
# Filename : example.py
# Copyright : 2020 By Lidihuo
# Author by : www.lidihuo.com
# Date : 2020-08-27
0.7733766233766234
上面的输出表明，我们的袋装决策树分类器模型的准确性达到了约77％。
