From 744e0bfa7f5fef2d667bb108dfcdfc039828d6af Mon Sep 17 00:00:00 2001
From: yuhaopan <1159526601@qq.com>
Date: Thu, 26 Sep 2024 17:57:44 +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 | 44 +++++++++++++++++++++-----------------------
 1 file changed, 21 insertions(+), 23 deletions(-)

diff --git a/function.py b/function.py
index df4d153..9b96ddc 100644
--- a/function.py
+++ b/function.py
@@ -1,6 +1,7 @@
 # function.py
 import numpy as np
 from numpy.linalg import inv, det
+from scipy.stats import mode
 """
 挑战说明:
 - 自行编写实现 KNN 分类算法。
@@ -38,7 +39,7 @@ name = "潘宇豪"
 
 
 # 这里写自己的代码实现
-def classify0(inX, dataSet, labels, k=3):
+def classify0(inX, dataSet, labels, ,k_values = [3,9]):
     """
     KNN 分类算法。
 
@@ -52,32 +53,29 @@ def classify0(inX, dataSet, labels, k=3):
     - int: 预测的分类标签。
     """
 
+    def mahalanobis_distance(inX, data, cov_inv):
+        diff = inX - data
+        return np.sqrt(np.dot(np.dot(diff, cov_inv), diff.T))
+
     mean = np.mean(dataSet, axis=0)
     cov_inv = inv(np.cov(dataSet, rowvar=False))
 
-    # 计算马氏距离
-    distances = []
-    for data in dataSet:
-        diff = inX - data
-        distance = np.sqrt(np.dot(np.dot(diff, cov_inv), diff.T))
-        distances.append(distance)
-    
-    distances = np.array(distances)
-    sortedDistIndices = distances.argsort()
+    # 计算所有k值的预测
+    predictions = []
+    for k in k_values:
+        distances = [mahalanobis_distance(inX, data, cov_inv) for data in dataSet]
+        sortedDistIndices = np.argsort(distances)
+        classCount = {}
+        for i in range(k):
+            voteIlabel = labels[sortedDistIndices[i]]
+            classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1
+        
+        sortedClassCount = sorted(classCount.items(), key=lambda item: item[1], reverse=True)
+        predictions.append(sortedClassCount[0][0])
 
-    # 距离加权投票
-    classCount = {}
-    for i in range(k):
-        index = sortedDistIndices[i]
-        voteIlabel = labels[index]
-        weight = 1 / (distances[index] + 1e-5)  # 使用距离的倒数作为权重，避免除以零
-        classCount[voteIlabel] = classCount.get(voteIlabel, 0) + weight
-
-    # 返回权重最高的类别
-    sortedClassCount = sorted(classCount.items(), key=lambda item: item[1], reverse=True)
-    return sortedClassCount[0][0]
-
-         
+    # 多数投票机制
+    final_prediction, _ = mode(predictions)
+    return final_prediction.mode[0]
  
 def load_data(file_path):
     """