使用SGD实现具有对数损失和L2正则化的SGD分类器而不使用sklearn

import numpy as np

class SGDClassifier:
    def __init__(self, learning_rate=0.01, penalty=0.01, max_iter=100):
        self.learning_rate = learning_rate
        self.penalty = penalty
        self.max_iter = max_iter
        self.weights = None
    
    def sigmoid(self, z):
        return 1 / (1 + np.exp(-z))
    
    def fit(self, X, y):
        n_samples, n_features = X.shape
        self.weights = np.zeros(n_features)
        
        for _ in range(self.max_iter):
            for i in range(n_samples):
                z = np.dot(X[i], self.weights)
                h = self.sigmoid(z)
                gradient = (h - y[i]) * X[i] + self.penalty * self.weights
                self.weights -= self.learning_rate * gradient
    
    def predict(self, X):
        z = np.dot(X, self.weights)
        h = self.sigmoid(z)
        return np.round(h)