From ee58302173983c612cf695f26ceeba482ed70b22 Mon Sep 17 00:00:00 2001
From: Xionghaoran <719929980@qq.com>
Date: Thu, 7 Nov 2024 16:29:42 +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 | 31 +++++++++++++------------------
 1 file changed, 13 insertions(+), 18 deletions(-)

diff --git a/function.py b/function.py
index dba13aa..2cf3dac 100644
--- a/function.py
+++ b/function.py
@@ -39,7 +39,7 @@ name = "熊浩燃"
 - int: 预测的分类标签。
 """
 
-def mahalanobis_distance(x1, x2, cov_inv, mean):
+def mahalanobis_distance(x1, x2, cov_inv):
     """
     计算两点之间的马氏距离。
     
@@ -47,13 +47,12 @@ def mahalanobis_distance(x1, x2, cov_inv, mean):
     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
+    dist = np.dot(np.dot(delta, cov_inv), delta.T)
     return dist
 
 def classify0(inX, dataSet, labels, k=3):
@@ -69,26 +68,22 @@ def classify0(inX, dataSet, labels, k=3):
     Returns:
     - int: 预测的分类标签。
     """
-    # 计算训练数据集的协方差矩阵和均值
+    # 计算训练数据集的协方差矩阵和均值，避免重复计算
     cov_matrix = np.cov(dataSet, rowvar=False)
-    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])
+    # 计算所有训练样本与待分类样本的马氏距离
+    distances = np.array([mahalanobis_distance(inX, x, cov_inv) for x in dataSet])
+
+    # 获取与待分类样本距离最近的 k 个样本的索引
+    nearest_indices = distances.argsort()[:k]
     
-    # 获取距离最近的 k 个样本的索引
-    sortedDistIndices = np.argpartition(distances, k)[:k]
+    # 获取 k 个最近样本的标签
+    nearest_labels = [labels[i] for i in nearest_indices]
     
-    # 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]
+    # 返回出现频率最高的标签作为预测结果
+    unique_labels, counts = np.unique(nearest_labels, return_counts=True)
+    return unique_labels[np.argmax(counts)]
 
 
 def load_data(file_path):