更新 function.py

function.py

@@ -39,38 +39,30 @@ name = "陈垣豪"
 - int: 预测的分类标签。
 """
 
-def classify0(inX, dataSet, labels, k=3):
+def classify0(inX, 数据集， 标签， k=3):
-    # Normalize the dataset
-    dataSet_mean = np.mean(dataSet, axis=0)
-    dataSet_std = np.std(dataSet, axis=0)
-    normalized_dataSet = (dataSet - dataSet_mean) / dataSet_std
-    normalized_inX = (inX - dataSet_mean) / dataSet_std
-    
     # 计算协方差矩阵并求其逆矩阵
-    cov_matrix = np.cov(normalized_dataSet, rowvar=False)
+    cov_matrix = np.cov(数据集,rowvar=False)
     inv_cov_matrix = np.linalg.inv(cov_matrix)
-    
+
     # 计算马氏距离
-    diff = normalized_dataSet - normalized_inX
+    diff = dataSet - inX
-    distances = np.einsum('ij,jk->i', diff, inv_cov_matrix @ diff.T)  # 使用 einsum 简化计算
+    distances = np.sum(diff @ inv_cov_matrix * diff, axis=1)
-    
+
     # 获取距离最小的 k 个样本的索引
     sorted_indices = np.argsort(distances)[:k] 
-    
+
     # 获取 k 个最近邻的标签
     nearest_labels = [labels[i] for i in sorted_indices]
-    
+
     # 统计 k 个最近邻中每个类别的出现次数
     label_count = {}
-    weights = 1 / (distances[sorted_indices] + 1e-10)  # Avoid division by zero
+    for label in nearest_labels:
-    
-    for i, label in enumerate(nearest_labels):
         if label in label_count:
-            label_count[label] += weights[i]
+            label_count[label] += 1
         else:
-            label_count[label] = weights[i]
+            label_count[label] = 1
-    
+
     # 找到出现次数最多的标签
     predicted_label = max(label_count, key=label_count.get)
-    
+
-    return predicted_label
+    return predicted_label提高准确率