上传文件至 /

function.py

@@ -1,129 +1,63 @@
 # function.py
+
 import numpy as np
 import os
-from collections import Counter
+from collections import defaultdict
 import operator
 
 id = "202210423104"
 name = "张钦诚"
 
-
-"""
-参数说明:
-- input_vector (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，请**自行设置默认值**，评测时仅会使用其默认值，不修改默认值将**无法通过**评测，**成绩为零**。
-返回值:
-- int: 预测的分类标签。
-"""
-
 def classify(input_vector, dataSet, labels, k=5):
-    """
-    KNN 分类算法。
 
-    Args:
-    - input_vector (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，请**自行设置默认值**，评测时仅会使用其默认值，不修改默认值将**无法通过**评测，**成绩为零**。
-
-    Returns:
-    - int: 预测的分类标签。
-    """
     # 计算欧式距离
+    # 使用广播和向量化计算距离，避免循环
     distances = np.linalg.norm(dataSet - input_vector, axis=1)
-    sortedDistIndices = distances.argsort()
-    k_labels = [labels[i] for i in sortedDistIndices[:k]]
-    count = Counter(k_labels)
-    return count.most_common(1)[0][0]
+
+    # 获取k个最近邻的索引
+    sortedDistIndices = distances.argsort()[:k]
+
+    # 获取k个最近邻的标签
+    k_labels = [labels[i] for i in sortedDistIndices]
+
+    # 获取k个最近邻的距离
+    k_distances = distances[sortedDistIndices]
+
+    # 加权多数表决：权重为1/(距离 + 1e-5) 避免除以零
+    weight_dict = defaultdict(float)
+    for label, dist in zip(k_labels, k_distances):
+        weight = 1 / (dist + 1e-5)
+        weight_dict[label] += weight
+
+    # 排序权重并选择权重最大的类别
+    sorted_weight = sorted(weight_dict.items(), key=operator.itemgetter(1), reverse=True)
+
+    return sorted_weight[0][0]
 
 
 def load_dataset(directory):
-    """
-    从文件夹中加载数据，若不自行正确实现，则无法通过评测，**不记录成绩**。
 
-    Args:
-    - directory (str): 数据文件夹路径。
-
-    Returns:
-    - np.array: 特征数据。
-    - np.array: 标签数据。
-    """
     labels = []
     files = os.listdir(directory)
     m = len(files)
-    dataset = np.zeros((m, 1024))
+    dataset = np.zeros((m, 1024), dtype=np.int8)
+
     for i, file in enumerate(files):
         filename = os.path.join(directory, file)
         dataset[i, :] = img_to_vector(filename)
         label = int(file.split('_')[0])
         labels.append(label)
+
     return dataset, labels
 
 
 def img_to_vector(filename):
-    """
-    将二值图像转换为向量。
-    该部分应为 load_dataset 函数的辅助子函数，可单独在这里实现，也可直接集成至 load_dataset 函数中。
 
-    Args:
-    - filename (str): 图像文件路径。
-
-    Returns:
-    - np.array: 二值图像向量。
-    """
-    returnVect = np.zeros((1, 1024))
-    with open(filename) as fr:
+    returnVect = np.zeros((1024,), dtype=np.int8)
+    with open(filename, 'r') as fr:
         for i in range(32):
-            lineStr = fr.readline()
+            lineStr = fr.readline().strip()
             for j in range(32):
-                returnVect[0, 32*i + j] = int(lineStr[j])
+                returnVect[32 * i + j] = int(lineStr[j])
     return returnVect
 
-
-if __name__ == "__main__":
-    # 以下代码可用于测试，可自行随意修改
-    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,
-    )
-
-    print(f"预测的标签: {predicted_label}")