python KNN回归预测

KNN回归是一种基于邻近样本的回归算法，它通过计算待预测样本与训练集中各个样本之间的距离，选取距离最近的K个样本，并根据这K个样本的输出值进行预测。下面是两种实现KNN回归的方式：

方式一：使用sklearn库中的KNeighborsRegressor类实现KNN回归算法。

from sklearn.neighbors import KNeighborsRegressor

# 创建KNN回归模型，设置K值为3
knn = KNeighborsRegressor(n_neighbors=3)

# 训练模型
X_train = [[0], [1], [2], [3]]
y_train = [0, 0.2, 0.4, 0.6]
knn.fit(X_train, y_train)

# 预测新样本
X_test = [[1.5]]
y_pred = knn.predict(X_test)
print("预测结果:", y_pred)  # 输出：[0.2]

方式二：自己实现KNN回归算法。

import numpy as np

# 计算欧氏距离
def euclidean_distance(x1, x2):
    return np.sqrt(np.sum((x1 - x2) ** 2))

# KNN回归算法
def knn_regression(X_train, y_train, X_test, k):
    y_pred = []
    for x_test in X_test:
        # 计算测试样本与训练样本的距离
        distances = [euclidean_distance(x_test, x_train) for x_train in X_train]
        # 根据距离排序，选取距离最近的K个样本
        k_indices = np.argsort(distances)[:k]
        # 取K个样本的输出值的平均作为预测值
        y_pred.append(np.mean(y_train[k_indices]))
    return y_pred

# 训练样本
X_train = np.array([[0], [1], [2], [3]])
y_train = np.array([0, 0.2, 0.4, 0.6])

# 预测样本
X_test = np.array([[1.5]])

# 调用KNN回归算法进行预测
y_pred = knn_regression(X_train, y_train, X_test, k=3)
print("预测结果:", y_pred)  # 输出：[0.2]