第一次实验

function.py

@@ -1,4 +1,3 @@
-# function.py
 import numpy as np
 import os
 
@@ -21,12 +20,11 @@ import os
 - numpy: 2.1.1
 """
 
-
 """
 请在这里填写学号和姓名，不填写即成绩无效。
 """
-id = "202210423xxx"
-name = "test_name"
+id = "202210423228"
+name = "黎世浩"
 
 
 """
@@ -41,7 +39,7 @@ name = "test_name"
 
 
 # 这里写自己的代码实现
-def classify(input_vector, dataSet, labels, k=0):
+def classify(input_vector, dataSet, labels, k=3):
     """
     KNN 分类算法。
 
@@ -54,6 +52,25 @@ def classify(input_vector, dataSet, labels, k=0):
     Returns:
     - int: 预测的分类标签。
     """
+    # 计算输入向量与数据集中每个样本的欧氏距离
+    distances = [np.linalg.norm(row-input_vector) for row in dataSet]
+    
+    # 获取距离最近的k个样本的索引
+    k_indices = np.argsort(distances)[:k]
+    
+    # 获取这些样本对应的标签
+    k_nearest_labels = [labels[i] for i in k_indices]
+    
+    # 统计每个标签出现的次数
+    label_counts = {}
+    for label in k_nearest_labels:
+        if label not in label_counts:
+            label_counts[label] = 0
+        label_counts[label] += 1
+    
+    # 找到出现次数最多的标签
+    sorted_labels = sorted(label_counts.items(), key=lambda x: x[1], reverse=True)
+    return sorted_labels[0][0]
 
 
 # 这里也写自己的代码实现
@@ -68,6 +85,15 @@ def load_dataset(directory):
     - np.array: 特征数据。
     - np.array: 标签数据。
     """
+    dataset = []
+    labels = []
+    for filename in os.listdir(directory):
+        filepath = os.path.join(directory, filename)
+        vector = img_to_vector(filepath)
+        label = int(filename.split('_')[0])
+        dataset.append(vector)
+        labels.append(label)
+    return np.array(dataset), np.array(labels)
 
 
 # 这里还写自己的代码实现
@@ -82,50 +108,26 @@ def img_to_vector(filename):
     Returns:
     - np.array: 二值图像向量。
     """
+    with open(filename, 'r') as file:
+        # 读取文件内容并忽略换行符
+        return np.array([int(char) for char in file.read() if char.isdigit()])
 
 
 if __name__ == "__main__":
-    # 以下代码可用于测试，可自行随意修改
+    # 确保当前目录下有trainingDigits文件夹
     train_data, train_labels = load_dataset("trainingDigits")
 
-    test_data = (
-        "00000000000000110000000000000000"
-        "00000000000011111000000000000000"
-        "00000000000011111100000000000000"
-        "00000000001111111110000000000000"
-        "00000000001111111110000000000000"
-        "00000000001111111100000000000000"
-        "00000000011111111000000000000000"
-        "00000000111111111000000000000000"
-        "00000000111111110000000000000000"
-        "00000000111111110000000000000000"
-        "00000000111111110000000000000000"
-        "00000000111111110000000000000000"
-        "00000000111111100000000000000000"
-        "00000001111111100000000000000000"
-        "00000001111111100000000000000000"
-        "00000001111111100000000000000000"
-        "00000001111111111111000000000000"
-        "00000001111111111111111000000000"
-        "00000001111111111111111100000000"
-        "00000011111111111111111110000000"
-        "00000011111111111111111110000000"
-        "00000001111111111111111110000000"
-        "00000001111111111111111110000000"
-        "00000000111111111100111111100000"
-        "00000000111111111000111111100000"
-        "00000000011111111001111111100000"
-        "00000000011111111111111111000000"
-        "00000000001111111111111111000000"
-        "00000000000111111111111111000000"
-        "00000000000011111111111100000000"
-        "00000000000001111111110000000000"
-        "00000000000000000111100000000000"
-    )
-    predicted_label = classify(
-        np.array([int(char) for char in test_data]).reshape(1, 1024),
-        train_data,
-        train_labels,
-    )
+    # 生成测试数据，确保它是1024维的
+    test_data = ""
+    for _ in range(32):  # 每行32个字符，共32行
+        for _ in range(32):  # 每行32个字符
+            test_data += str(np.random.randint(0, 2))  # 随机生成0或1
 
+    # 将测试数据转换为numpy数组并重塑为(1, 1024)的形状
+    test_vector = np.array([int(char) for char in test_data]).reshape(1, 1024)
+
+    # 使用kNN算法进行分类
+    predicted_label = classify(test_vector, train_data, train_labels)
+
+    # 输出预测的标签
     print(f"预测的标签: {predicted_label}")