梯度下降法代码 python

梯度下降法是一种常用的优化算法，用于求解函数的最小值。下面是一个简单的梯度下降法的 Python 代码示例：

import numpy as np

def gradient_descent(x, y, learning_rate, num_iterations):
    num_samples = len(y)
    num_features = len(x[0])
    theta = np.zeros(num_features)  # 初始化参数

    for _ in range(num_iterations):
        gradient = np.zeros(num_features)  # 初始化梯度
        for i in range(num_samples):
            h = np.dot(theta, x[i])  # 计算预测值
            loss = h - y[i]
            gradient += loss * x[i]  # 计算梯度
        theta -= (learning_rate / num_samples) * gradient  # 更新参数

    return theta

# 示例数据
x = np.array([[1, 2, 3], [1, 3, 4], [1, 4, 5], [1, 5, 6], [1, 6, 7]])
y = np.array([3, 4, 5, 6, 7])

# 超参数设置
learning_rate = 0.01
num_iterations = 1000

# 调用梯度下降算法
theta = gradient_descent(x, y, learning_rate, num_iterations)

print("参数 theta:", theta)