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=5):
+def classify0(inX, dataSet, labels, k=3):
     """
     KNN 分类算法。
 
@@ -51,18 +51,20 @@ def classify0(inX, dataSet, labels, k=5):
     返回值:
     - int: 预测的分类标签。
     """
-    # 计算输入样本与训练数据集中每个样本的欧氏距离
-    diffMat = np.tile(inX, (dataSet.shape[0], 1)) - dataSet
-    sqDiffMat = diffMat**2
-    sqDistances = sqDiffMat.sum(axis=1)
-    distances = np.sqrt(sqDistances)
+ # 计算协方差矩阵及其逆
+    covariance_matrix = np.cov(dataSet, rowvar=False)
+    inv_covmat = inv(covariance_matrix)
+
+    # 计算马氏距离
+    distances = np.array([mahalanobis(inX, dataSet[i], inv_covmat) for i in range(dataSet.shape[0])])
 
     # 对距离进行排序并找到最近的k个邻居
-    sortedDistIndices = distances.argsort()
-    classCount = {}
+    sortedDistIndices = distances.argsort()[:k]
     
-    for i in range(k):
-        voteIlabel = labels[sortedDistIndices[i]]
+    # 进行简单投票
+    classCount = {}
+    for i in sortedDistIndices:
+        voteIlabel = labels[i]
         classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1
     
     # 返回出现次数最多的类别