更新 function.py

function.py

@@ -38,7 +38,7 @@ 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=3, dynamic_k=False):
     """
     KNN 分类算法。
 
@@ -52,33 +52,33 @@ def classify0(inX, dataSet, labels, k=3):
     - int: 预测的分类标签。
     """
 
-
-        
-    mean = np.mean(dataSet, axis=0) 
+    mean = np.mean(dataSet, axis=0)
     cov_inv = inv(np.cov(dataSet, rowvar=False))
-    
-    # 计算输入样本与每个训练样本之间的马氏距离
+
+    # 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)
     
-    # 对距离进行排序并找到最近的k个邻居
     distances = np.array(distances)
     sortedDistIndices = distances.argsort()
+    
+    # 动态调整 k 值
+    if dynamic_k:
+        median_distance = np.median(distances)
+        k_dynamic = min(max(3, int(len(distances) * (median_distance / np.max(distances)))), 10)
+        k = k_dynamic
+
     classCount = {}
     for i in range(k):
         voteIlabel = labels[sortedDistIndices[i]]
         classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1
-    
-    # 返回出现次数最多的类别
+
     sortedClassCount = sorted(classCount.items(), key=lambda item: item[1], reverse=True)
     return sortedClassCount[0][0]
-    
-
-    if np.allclose(inX, [5.56500e+03, 9.73334e+00, 9.77746e-01], atol=1e-2):
-        return 2 
+         
  
 def load_data(file_path):
     """