From a86348a74121bf9ec727e533629c07f57dfc9364 Mon Sep 17 00:00:00 2001
From: 202210423203 <2579478480@qq.com>
Date: Tue, 15 Oct 2024 20:00:05 +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 | 50 ++++++++++----------------------------------------
 1 file changed, 10 insertions(+), 40 deletions(-)

diff --git a/function.py b/function.py
index b734d88..86fe43e 100644
--- a/function.py
+++ b/function.py
@@ -41,31 +41,20 @@ name = "赵昊龙"
 
 
 # 这里写自己的代码实现
-def euclidean_distance(inX, dataSet):
-    # 将inX转换为numpy数组
+import numpy as np
+
+def mahalanobis_distance(inX, dataSet, cov_mat):
+  
     inX = np.array(inX).reshape(1, -1)
-    # 计算inX和dataSet中每一行的差的平方和
-    sq_diff = (inX - dataSet) ** 2
-    # 计算每行的和，然后开方得到欧氏距离
-    sq_distances = sq_diff.sum(axis=1)
-    distances = np.sqrt(sq_distances)
+    diff = inX - dataSet
+    distances = np.sqrt(diff @ cov_mat @ diff.T)
     return distances
 
 def classify0(inX, dataSet, labels, k=5):
-    """
-    KNN 分类算法。
-    
-    参数:
-    inX (numpy array): 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，默认值为5。
-    
-    返回:
-    int: 预测的分类标签。
-    """
-    # 计算待分类样本与数据集中每个样本的距离
-    distances = euclidean_distance(inX, dataSet)
+    # 计算训练数据集的协方差矩阵
+    cov_mat = np.cov(dataSet.T)
+    # 计算待分类样本与数据集中每个样本的马氏距离
+    distances = mahalanobis_distance(inX, dataSet, cov_mat)
     # 获取距离最小的k个样本的索引
     sortedDistIndices = distances.argsort()
     # 为分类计数器初始化
@@ -80,16 +69,6 @@ def classify0(inX, dataSet, labels, k=5):
     return sortedClassCount[0][0]
 
 def load_data(file_path):
-    """
-    加载 TSV 格式数据文件。
-    
-    参数:
-    file_path (str): 数据文件的路径。
-    
-    返回:
-    features (numpy.ndarray): 数据特征。
-    labels (numpy.ndarray): 数据标签。
-    """
     data = np.loadtxt(file_path, delimiter="\t")
     features = data[:, :-1]
     labels = data[:, -1].astype(int)
@@ -100,12 +79,3 @@ if __name__ == "__main__":
     train_file = "datingTrainSet2.tsv"
     train_data, train_labels = load_data(train_file)
     print(classify0([65915, 3.868359, 0.128078], train_data, train_labels))
-
-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([65915, 3.868359, 0.128078], train_data, train_labels))