From 7eb48c19c81193a6350ab493f6b56a875ad3a757 Mon Sep 17 00:00:00 2001
From: Xionghaoran <719929980@qq.com>
Date: Thu, 31 Oct 2024 15:09:37 +0800
Subject: [PATCH] =?UTF-8?q?=E6=9B=B4=E6=96=B0=20function.py?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 function.py | 42 ++++++++++++++++++++++--------------------
 1 file changed, 22 insertions(+), 20 deletions(-)

diff --git a/function.py b/function.py
index fb8ce50..8b22af3 100644
--- a/function.py
+++ b/function.py
@@ -17,38 +17,50 @@ import numpy as np
 - numpy: 2.1.1
 """
 
-
 """
 请在这里填写学号和姓名，不填写即成绩无效。
 """
-id = "202210423213"
-name = "熊浩燃"
+id = "202210423xxx"
+name = "test_name"
 
-
-# 这里写自己的代码实现
 def sigmoid(z):
     return 1 / (1 + np.exp(-z))
 
-def train_logreg(X, y, learning_rate=0.01, iterations=1000):
+def normalize(X):
+    """ 数据归一化 """
+    return (X - np.mean(X, axis=0)) / np.std(X, axis=0)
+
+def train_logreg(X, y, learning_rate=0.01, iterations=1000, batch_size=32, lambda_=0.01):
     """
-    对率回归训练函数。
+    对率回归训练函数，包含批量梯度下降和L2正则化。
 
     Args:
     - X (numpy array): 特征数据。
     - y (numpy array): 标签数据。
     - learning_rate (float): 学习率。
     - iterations (int): 迭代次数。
+    - batch_size (int): 每次更新的样本数。
+    - lambda_ (float): L2正则化项的系数。
 
     Returns:
     - numpy array: 训练得到的权重。
     """
+    X = normalize(X)  # 归一化数据
     m, n = X.shape
     weights = np.zeros(n)
+    
     for _ in range(iterations):
-        z = np.dot(X, weights)
+        # 随机选择一个小批量
+        indices = np.random.choice(m, batch_size, replace=False)
+        X_batch = X[indices]
+        y_batch = y[indices]
+        
+        # 计算预测和梯度
+        z = np.dot(X_batch, weights)
         h = sigmoid(z)
-        gradient = np.dot(X.T, (h - y)) / m
+        gradient = np.dot(X_batch.T, (h - y_batch)) / batch_size + lambda_ * weights
         weights -= learning_rate * gradient
+    
     return weights
 
 def predict_logreg(inX, weights):
@@ -65,7 +77,6 @@ def predict_logreg(inX, weights):
     probability = sigmoid(np.dot(inX, weights))
     return 1 if probability >= 0.5 else 0
 
-# 这里也写自己的代码实现
 def load_data(filepath):
     """
     从文件中加载数据，若不自行正确实现，则无法通过评测，**不记录成绩**。
@@ -93,16 +104,7 @@ if __name__ == "__main__":
     X_test, y_test = load_data("./data/horse-colic.test")
 
     # 训练对率回归模型
-    weights = train_logreg(X_train, y_train, learning_rate=0.01, iterations=1000)
-
-    # 测试单个数据
-    y_pred = predict_logreg(X_test[11], weights)
-    print(f"预测结果: {y_pred}, 实际结果: {y_test[11]}")
-    # 加载测试集
-    X_test, y_test = load_data("./data/horse-colic.test")
-
-    # 训练对率回归模型
-    weights = train_logreg(X_train, y_train, learning_rate=0.01, iterations=1000)
+    weights = train_logreg(X_train, y_train, learning_rate=0.01, iterations=1000, batch_size=32, lambda_=0.01)
 
     # 测试单个数据
     y_pred = predict_logreg(X_test[11], weights)