From 7ceb6f3decf879ebc6a0c0fa4bb50ab4244a872c Mon Sep 17 00:00:00 2001
From: Xionghaoran <719929980@qq.com>
Date: Thu, 7 Nov 2024 16:26:09 +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 | 58 +++++++++++++++++++++++++++++++++++++----------------
 1 file changed, 41 insertions(+), 17 deletions(-)

diff --git a/function.py b/function.py
index d16310e..dba13aa 100644
--- a/function.py
+++ b/function.py
@@ -1,6 +1,7 @@
 # function.py
 import numpy as np
 
+
 """
 挑战说明:
 - 自行编写实现 KNN 分类算法。
@@ -21,7 +22,6 @@ 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。
 """
@@ -39,36 +39,57 @@ name = "熊浩燃"
 - int: 预测的分类标签。
 """
 
+def mahalanobis_distance(x1, x2, cov_inv, mean):
+    """
+    计算两点之间的马氏距离。
+    
+    Args:
+    x1 (numpy array): 第一个点。
+    x2 (numpy array): 第二个点。
+    cov_inv (numpy array): 协方差矩阵的逆矩阵。
+    mean (numpy array): 数据集的均值。
+    
+    Returns:
+    float: 两点之间的马氏距离。
+    """
+    delta = x1 - x2
+    dist = delta @ cov_inv @ delta.T
+    return dist
 
-# 这里写自己的代码实现
 def classify0(inX, dataSet, labels, k=3):
     """
-    KNN 分类算法。
+    使用马氏距离的 KNN 分类算法。
 
     Args:
     - inX (numpy array): 输入待分类的样本。
     - dataSet (numpy array): 训练数据集。
     - labels (list or numpy array): 训练集对应的标签。
-    - k (int): kNN 算法参数，请**自行设置默认值**，评测时仅会使用其默认值，不修改默认值将**无法通过**评测，**成绩为零**。
+    - k (int): kNN 算法参数，默认值为 3。
 
     Returns:
     - int: 预测的分类标签。
     """
-    # 计算协方差矩阵的逆
+    # 计算训练数据集的协方差矩阵和均值
     cov_matrix = np.cov(dataSet, rowvar=False)
-    inv_cov_matrix = np.linalg.inv(cov_matrix)
+    mean = np.mean(dataSet, axis=0)
+    cov_inv = np.linalg.inv(cov_matrix + np.eye(cov_matrix.shape[0]) * 1e-5)  # 加上一个小的单位矩阵以避免奇异矩阵
     
-    # 计算马氏距离
-    diff = dataSet - inX
-    distances = np.sqrt(np.sum(diff @ inv_cov_matrix * diff, axis=1))
+    # 计算待分类样本与训练集中每个样本的马氏距离
+    distances = np.array([mahalanobis_distance(inX, x, cov_inv, mean) for x in dataSet])
     
-    # 按距离排序并返回 k 个最近的标签
-    sorted_indices = distances.argsort()
-    nearest_labels = [labels[i] for i in sorted_indices[:k]]
+    # 获取距离最近的 k 个样本的索引
+    sortedDistIndices = np.argpartition(distances, k)[:k]
     
-    # 找出出现次数最多的标签
-    label_counts = np.bincount(nearest_labels)
-    return label_counts.argmax()
+    # 获取这些样本对应的标签
+    classCount = {}
+    for i in sortedDistIndices:
+        voteLabel = labels[i]
+        classCount[voteLabel] = classCount.get(voteLabel, 0) + 1
+    
+    # 选择出现次数最多的标签作为预测标签
+    sortedClassCount = sorted(classCount.items(), key=lambda x: x[1], reverse=True)
+    return sortedClassCount[0][0]
+
 
 def load_data(file_path):
     """
@@ -82,7 +103,10 @@ def load_data(file_path):
 
 if __name__ == "__main__":
     # 以下代码可用于测试，可自行随意修改
-
     train_file = "datingTrainSet2.tsv"
     train_data, train_labels = load_data(train_file)
-    print(classify0(np.array([65915, 3.868359, 0.128078]), train_data, train_labels))
\ No newline at end of file
+
+    # 测试一个样本
+    test_sample = np.array([65915, 3.868359, 0.128078])
+    predicted_label = classify0(test_sample, train_data, train_labels)
+    print(f"The predicted label for the test sample is: {predicted_label}")
\ No newline at end of file