更新 function.py

function.py

@@ -44,61 +44,65 @@ name = "陈瑞琪祥"
 
 
 
-def classify0(inX, dataSet, labels, k=5):
-    """
-    KNN 分类算法。
 
-    Args:
-    - inX (numpy array): 输入待分类的样本。
-    - dataSet (numpy array): 训练数据集。
-    - labels (list or numpy array): 训练集对应的标签。
-    - k (int): kNN 算法参数，默认值为 5。
 
-    Returns:
-    - int: 预测的分类标签。
-    """
-    # 计算待分类样本与训练集中每个样本的欧氏距离
+def classify0(inX, dataSet, labels, k):
     diffMat = np.tile(inX, (dataSet.shape[0], 1)) - dataSet
     sqDiffMat = diffMat**2
     sqDistances = sqDiffMat.sum(axis=1)
     distances = sqDistances**0.5
-    
-    # 对距离进行排序，获取距离最近的 k 个样本的索引
     sortedDistIndices = distances.argsort()
-    
-    # 提取距离最近的 k 个样本的标签
     classCount = {}
     for i in range(k):
         voteIlabel = labels[sortedDistIndices[i]]
         classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1
-    
-    # 根据出现次数最多的标签来预测待分类样本的标签
     sortedClassCount = sorted(classCount.items(), key=lambda x: x[1], reverse=True)
     return sortedClassCount[0][0]
 
 def load_data(file_path):
-    """
-    加载 TSV 格式数据文件，并进行标准化处理。
-    """
     data = np.loadtxt(file_path, delimiter="\t")
     features = data[:, :-1]
     labels = data[:, -1].astype(int)
-    
-    # 标准化特征
     features = (features - np.mean(features, axis=0)) / np.std(features, axis=0)
-    
     return features, labels
 
+def find_best_k(train_data, train_labels):
+    best_accuracy = 0
+    best_k = 1
+    for k in range(1, 26, 2):  # 尝试奇数的 k 值，从 1 到 25
+        correct_count = 0
+        for i in range(train_data.shape[0]):
+            if classify0(train_data[i], train_data, train_labels, k) == train_labels[i]:
+                correct_count += 1
+        accuracy = correct_count / train_data.shape[0]
+        if accuracy > best_accuracy:
+            best_accuracy = accuracy
+            best_k = k
+    return best_k
+
+def preprocess_data(features):
+    # 简单的异常值处理
+    Q1 = np.percentile(features, 25, axis=0)
+    Q3 = np.percentile(features, 75, axis=0)
+    IQR = Q3 - Q1
+    lower_bound = Q1 - 1.5 * IQR
+    upper_bound = Q3 + 1.5 * IQR
+    return np.clip(features, lower_bound, upper_bound)
+
 if __name__ == "__main__":
     train_file = "datingTrainSet2.tsv"
-    train_data, train_labels = load_data(train_file)
     test_file = "datingTestSet2.tsv"
+    train_data, train_labels = load_data(train_file)
     test_data, test_labels = load_data(test_file)
     
-    # 测试 KNN 分类器
+    train_data = preprocess_data(train_data)
+    test_data = preprocess_data(test_data)
+    
+    best_k = find_best_k(train_data, train_labels)
+    
     correct_count = 0
     for i in range(test_data.shape[0]):
-        if classify0(test_data[i], train_data, train_labels) == test_labels[i]:
+        if classify0(test_data[i], train_data, train_labels, best_k) == test_labels[i]:
             correct_count += 1
     
-    print(f"Accuracy: {correct_count / test_data.shape[0] * 100:.2f}%")
+    print(f"Accuracy with best k = {best_k}: {correct_count / test_data.shape[0] * 100:.2f}%")