决策树和随机森林对经典鸢尾花数据集分类

1.决策树

import pandas as pd
import numpy as np
from sklearn.datasets import load_iris
from sklearn.tree import DecisionTreeClassifier
from sklearn.tree import export_graphviz
from sklearn.tree import DecisionTreeRegressor
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
import matplotlib.pyplot as plt
import matplotlib as mpl

#导入数据
iris = load_iris()

#用pandas读取数据
data = pd.DataFrame(iris.data)

#添加特征名称
data.columns = iris.feature_names

#添加目标值
data['Species'] = load_iris().target

# 特征数据只取花萼长度和宽度
x = data.iloc[:, :2]
#取出目标数据
y = data.iloc[:, -1]

#切分数据集，指定随机种子
x_train, x_test, y_train, y_test = train_test_split(x, y, train_size=0.75, random_state=42)

#实例化一个决策树对象  指定最大树深为2，信息增益计算标准为基尼系数
tree_clf = DecisionTreeClassifier(max_depth=4, criterion='gini')

#传入训练集数据进行训练
tree_clf.fit(x_train, y_train)
#传入测试集进行预测
y_test_hat = tree_clf.predict(x_test)

#打印准确率
print("acc score:", accuracy_score(y_test, y_test_hat))

#验证树深与准确率关系
depth = np.arange(1, 15)
err_list = []
for d in depth:
    clf = DecisionTreeClassifier(criterion='entropy', max_depth=d)
    clf.fit(x_train, y_train)
    y_test_hat = clf.predict(x_test)
    result = (y_test_hat == y_test)
    err = 1 - np.mean(result)
    err_list.append(err)
    print(d, '错误率：%.2f%%' % (100 * err))

mpl.rcParams['font.sans-serif'] = ['SimHei']
plt.figure(facecolor='w')
plt.plot(depth, err_list, 'ro-', lw=2)
plt.xlabel('决策树深度', fontsize=15)
plt.ylabel('错误率', fontsize=15)
plt.title('决策树深度和过拟合', fontsize=18)
plt.grid(True)
plt.show()

运行结果：
acc score: 0.7631578947368421
1 错误率：36.84%
2 错误率：34.21%
3 错误率：31.58%
4 错误率：26.32%
5 错误率：34.21%
6 错误率：28.95%
7 错误率：31.58%
8 错误率：26.32%
9 错误率：26.32%
10 错误率：39.47%
11 错误率：42.11%
12 错误率：34.21%
13 错误率：36.84%
14 错误率：36.84%

2.随机森林

from sklearn.ensemble import RandomForestClassifier
from sklearn.datasets import make_moons
from sklearn.model_selection import train_test_split
from sklearn.ensemble import BaggingClassifier
from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import accuracy_score
from sklearn.datasets import load_iris

iris = load_iris()
X = iris.data[:, :2]  # 花萼长度和宽度
y = iris.target
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33, random_state=42)

#实例化一个随机森林分类器对象   指定包含5棵决策树，最大叶子节点数为16，用5个线程进行训练
rnd_clf = RandomForestClassifier(n_estimators=15, max_leaf_nodes=20, n_jobs=5)
rnd_clf.fit(X_train, y_train)

y_pred_rf = rnd_clf.predict(X_test)
print(accuracy_score(y_test, y_pred_rf))

# 计算特征重要性
iris = load_iris()
rnd_clf = RandomForestClassifier(n_estimators=500, n_jobs=-1)
rnd_clf.fit(iris["data"], iris['target'])
for name, score in zip(iris['feature_names'], rnd_clf.feature_importances_):
    print(name, score)

运行结构:
0.78
sepal length (cm) 0.10222109801813306
sepal width (cm) 0.02548487947686973
petal length (cm) 0.4183213187649062
petal width (cm) 0.4539727037400911

Process finished with exit code 0