带有正则化的逻辑回归代码

以下是带有L2正则化的逻辑回归代码：

import numpy as np

class LogisticRegression:
    def __init__(self, alpha=0.01, lambda_val=0.01, num_iterations=1000):
        self.alpha = alpha
        self.lambda_val = lambda_val
        self.num_iterations = num_iterations
        self.theta = None

    def sigmoid(self, z):
        return 1 / (1 + np.exp(-z))

    def cost_function(self, X, y, theta):
        m = len(y)
        h = self.sigmoid(X.dot(theta))
        J = (-1 / m) * (y.T.dot(np.log(h)) + (1 - y).T.dot(np.log(1 - h)))
        reg = (self.lambda_val / (2 * m)) * np.sum(np.square(theta[1:]))
        J += reg
        return J

    def gradient_descent(self, X, y, theta):
        m = len(y)
        for i in range(self.num_iterations):
            h = self.sigmoid(X.dot(theta))
            reg = (self.lambda_val / m) * theta[1:]
            reg = np.insert(reg, 0, 0, axis=0)
            gradient = (1 / m) * X.T.dot(h - y) + reg
            theta -= self.alpha * gradient
        return theta

    def fit(self, X, y):
        m, n = X.shape
        X = np.insert(X, 0, 1, axis=1)
        self.theta = np.zeros((n + 1, 1))
        self.theta = self.gradient_descent(X, y, self.theta)

    def predict(self, X):
        X = np.insert(X, 0, 1, axis=1)
        return np.round(self.sigmoid(X.dot(self.theta)))

其中，alpha是学习率，lambda_val是正则化参数，num_iterations是梯度下降的迭代次数。在cost_function中，加入了正则化项，用于惩罚模型的复杂度。在gradient_descent中，对theta进行更新时，也加入了正则化项。