c45决策树分类代码

C4.5决策树是一种经典的机器学习算法，用于分类问题。下面是一个简单的C4.5决策树分类代码的示例：

import numpy as np

class Node:
    def __init__(self, feature=None, threshold=None, label=None):
        self.feature = feature
        self.threshold = threshold
        self.label = label
        self.left = None
        self.right = None

class C45DecisionTree:
    def __init__(self):
        self.root = None

    def entropy(self, labels):
        _, counts = np.unique(labels, return_counts=True)
        probabilities = counts / len(labels)
        entropy = -np.sum(probabilities * np.log2(probabilities))
        return entropy

    def information_gain(self, data, labels, feature, threshold):
        left_mask = data[:, feature] <= threshold
        right_mask = data[:, feature] > threshold
        left_labels = labels[left_mask]
        right_labels = labels[right_mask]
        left_entropy = self.entropy(left_labels)
        right_entropy = self.entropy(right_labels)
        parent_entropy = self.entropy(labels)
        information_gain = parent_entropy - (len(left_labels) / len(labels)) * left_entropy - (len(right_labels) / len(labels)) * right_entropy
        return information_gain

    def choose_best_split(self, data, labels):
        best_gain = 0
        best_feature = None
        best_threshold = None
        num_features = data.shape[1]
        for feature in range(num_features):
            unique_values = np.unique(data[:, feature])
            for threshold in unique_values:
                gain = self.information_gain(data, labels, feature, threshold)
                if gain > best_gain:
                    best_gain = gain
                    best_feature = feature
                    best_threshold = threshold
        return best_feature, best_threshold

    def build_tree(self, data, labels):
        if len(np.unique(labels)) == 1:
            return Node(label=labels[0])
        best_feature, best_threshold = self.choose_best_split(data, labels)
        left_mask = data[:, best_feature] <= best_threshold
        right_mask = data[:, best_feature] > best_threshold
        left_data = data[left_mask]
        left_labels = labels[left_mask]
        right_data = data[right_mask]
        right_labels = labels[right_mask]
        node = Node(feature=best_feature, threshold=best_threshold)
        node.left = self.build_tree(left_data, left_labels)
        node.right = self.build_tree(right_data, right_labels)
        return node

    def fit(self, data, labels):
        self.root = self.build_tree(data, labels)

    def predict(self, data):
        predictions = []
        for sample in data:
            node = self.root
            while node.label is None:
                if sample[node.feature] <= node.threshold:
                    node = node.left
                else:
                    node = node.right
            predictions.append(node.label)
        return predictions

# 示例用法
data = np.array([[1, 2], [2, 3], [3, 4], [4, 5]])
labels = np.array([0, 0, 1, 1])
tree = C45DecisionTree()
tree.fit(data, labels)
test_data = np.array([[1.5, 2.5], [3.5, 4.5]])
predictions = tree.predict(test_data)
print(predictions)

这段代码实现了C4.5决策树的构建和分类功能。你可以根据自己的数据集进行相应的调整和使用。