更新 function.py

function.py

@@ -1,7 +1,9 @@
-# function.py
+#function.py
 import numpy as np
 import os
-
+from collections import Counter
+id="202210423115"
+name="李煜锋"
 def img_to_vector(filename):
     """
     将二值图像转换为向量。
@@ -31,6 +33,7 @@ def load_dataset(directory):
     feature_list = []
     label_list = []
     
+    print(f"正在加载目录: {directory}")
     # 遍历目录，加载数据
     for filename in os.listdir(directory):
         if filename.endswith('.txt'):
@@ -49,7 +52,6 @@ def load_dataset(directory):
     labels = np.array(label_list)
     
     return features, labels
-from collections import Counter
 
 def knn_classify(input_vector, dataSet, labels, k=3):
     """
@@ -77,14 +79,29 @@ def knn_classify(input_vector, dataSet, labels, k=3):
     label_count = Counter(top_k_labels)
     
     return label_count.most_common(1)[0][0]
-# 加载训练数据
-train_data, train_labels = load_dataset('C:\\Users\\Administrator\\Desktop\\trainingDigits')
 
-# 加载测试数据
-test_data, test_labels = load_dataset('C:\\Users\\Administrator\\Desktop\\testDigits')
-
-# 选择一个测试样本，进行分类
-for i in range(len(test_data)):
-    predicted_label = knn_classify(test_data[i], train_data, train_labels, k=3)
-    true_label = test_labels[i]
-    print(f"预测标签: {predicted_label}, 真实标签: {true_label}")
+# 主程序
+if __name__ == '__main__':
+    # 使用绝对路径替换为你数据集的路径
+    train_directory = r'C:\Users\Administrator\Desktop\trainingDigits'
+    test_directory = r'C:\Users\Administrator\Desktop\testDigits'
+    
+    # 加载训练数据
+    train_data, train_labels = load_dataset(train_directory)
+    
+    # 加载测试数据
+    test_data, test_labels = load_dataset(test_directory)
+    
+    # 测试 kNN 分类器
+    correct_predictions = 0
+    
+    for i in range(len(test_data)):
+        predicted_label = knn_classify(test_data[i], train_data, train_labels, k=3)
+        true_label = test_labels[i]
+        print(f"预测标签: {predicted_label}, 真实标签: {true_label}")
+        
+        if predicted_label == true_label:
+            correct_predictions += 1
+            
+    accuracy = correct_predictions / len(test_data) * 100
+    print(f"分类准确率: {accuracy:.2f}%")