diff --git a/function.py b/function.py
index 43bdb94..78409d3 100644
--- a/function.py
+++ b/function.py
@@ -41,7 +41,7 @@ name = "唐一鸣"
 
 
 # 这里写自己的代码实现
-def classify(input_vector, dataSet, labels, k=3):
+def classify(input_vector, dataSet, labels, k=1):
 
     """
     KNN 分类算法。
@@ -84,8 +84,8 @@ def load_dataset(directory):
     file_list = os.listdir(directory)
     num_files = len(file_list)
     
-    # 初始化数据矩阵，假设每个图像转化为 28x28=784 的向量
-    data_matrix = np.zeros((num_files, 784))
+    # 初始化数据矩阵，假设每个图像转化为 32x32=1024 的向量
+    data_matrix = np.zeros((num_files, 1024))
     
     for i, filename in enumerate(file_list):
         # 提取文件名中的数字作为标签
@@ -113,13 +113,13 @@ def img_to_vector(filename):
     """
 
    # 初始化一个 1D 向量
-    vector = np.zeros((1, 784))
+    vector = np.zeros((1, 1024))
     
     with open(filename, 'r') as file:
-        for i in range(28):
+        for i in range(32):
             line = file.readline()
-            for j in range(28):
-                vector[0, 28 * i + j] = int(line[j])
+            for j in range(32):
+                vector[0, 32 * i + j] = int(line[j])
     
     return vector
 
@@ -162,7 +162,7 @@ if __name__ == "__main__":
         "00000000000000000111100000000000"
     )
     predicted_label = classify(
-        np.array([int(char) for char in test_data]).reshape(1, 784),
+        np.array([int(char) for char in test_data]).reshape(1, 1024),
         train_data,
         train_labels,
     )