决策树简介

from sklearn.tree import DecisionTreeClassifier, DecisionTreeRegressor
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
import numpy as np

# 1. CART分类示例
def cart_classification():
    """CART分类"""
    # 生成示例数据：女孩相亲决策
    # 特征：[年龄, 长相(0丑1中等2帅), 收入(0低1中2高), 是否公务员(0否1是)]
    X = np.array([
        [28, 2, 2, 0],  # 年轻、帅、高收入、非公务员 → 见
        [35, 1, 1, 0],  # 年龄大、中等、中收入、非公务员 → 不见
        [25, 0, 1, 1],  # 年轻、丑、中收入、公务员 → 不见
        [27, 1, 1, 1],  # 年轻、中等、中收入、公务员 → 见
        [32, 2, 2, 0],  # 年龄大、帅、高收入、非公务员 → 不见
        [26, 2, 2, 0],  # 年轻、帅、高收入、非公务员 → 见
    ])
    y = np.array([1, 0, 0, 1, 0, 1])  # 1:见, 0:不见
    
    # 划分训练集和测试集
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
    
    # 创建CART分类树
    # criterion: gini(基尼) 或 entropy(信息增益)
    clf = DecisionTreeClassifier(criterion='gini', random_state=42)
    clf.fit(X_train, y_train)
    
    # 预测
    y_pred = clf.predict(X_test)
    
    # 评估
    print("=== CART分类 ===")
    print(f"准确率: {accuracy_score(y_test, y_pred):.4f}")
    
    # 预测新样本
    new_sample = np.array([[28, 2, 2, 0]])  # 年轻、帅、高收入、非公务员
    prediction = clf.predict(new_sample)
    print(f"新样本预测: {'见' if prediction[0] == 1 else '不见'}")

# 2. CART回归示例
def cart_regression():
    """CART回归"""
    # 生成示例数据
    np.random.seed(42)
    X = np.sort(5 * np.random.rand(100, 1), axis=0)
    y = np.sin(X).ravel() + np.random.normal(0, 0.1, X.shape[0])
    
    # 划分训练集和测试集
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
    
    # 创建CART回归树
    reg = DecisionTreeRegressor(criterion='squared_error', max_depth=3, random_state=42)
    reg.fit(X_train, y_train)
    
    # 预测
    y_pred = reg.predict(X_test)
    
    # 评估
    from sklearn.metrics import mean_squared_error
    mse = mean_squared_error(y_test, y_pred)
    print("\n=== CART回归 ===")
    print(f"MSE: {mse:.4f}")
    print(f"R²: {reg.score(X_test, y_test):.4f}")

# 运行
if __name__ == '__main__':
    cart_classification()
    cart_regression()