更新 function.py

2024-10-16 20:53:42 +08:00
parent 4a340d7eef
commit f06746ec17
1 changed files with 76 additions and 101 deletions
@@ -2,124 +2,99 @@
 import numpy as np
 import os

-"""
-实验说明:
- 自行编写实现数据加载与 kNN 的手写数字分类算法。
- 函数输入: 待分类的样本 input_vector, 训练数据集 dataSet, 训练集对应的标签 labels。
- 函数返回: 预测的分类标签。
-
-注意事项:
- 在 function.py 中实现 kNN 分类算法。
- 可以自行导入其他库，需要在 requirements.txt 添加依赖，若未在 requirements.txt 中添加使用到的依赖，则无法自动评测，成绩为零。
- 自动评测脚本仅会加载 classify 和 load_data 函数，请勿修改函数名。
- 训练集和测试集需手动下载解压，并放入当前目录下。
-
-测评环境:
- CPU: Intel Xeon(R) CPU E3-1231 v3 @ 3.40GHz
- OS: Ubuntu 22.04.4 LTS
- Python: 3.10.12
- numpy: 2.1.1
-"""


 """
 请在这里填写学号和姓名，不填写即成绩无效。
 """
-id = "202210423109"
+id = "202210423219"
 name = "李燕如"


-"""
-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。
-"""
+


 # 这里写自己的代码实现
 def classify(input_vector, dataSet, labels, k=3):
-    # 计算输入向量与数据集中每个样本的欧氏距离
-    distances = np.sqrt(((dataSet - input_vector) ** 2).sum(axis=1))
-    # 获取距离最近的k个样本的索引
-    sortedDistIndices = distances.argsort()
-    # 获取这些样本对应的标签
-    classCount = {}
-    for i in range(k):
-        voteIlabel = labels[sortedDistIndices[i]]
-        classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1
-    # 选择票数最多的类别作为预测标签
-    sortedClassCount = sorted(classCount.items(), key=lambda x: x[1], reverse=True)
-    return sortedClassCount[0][0]
+    # 计算输入向量与数据集中每个样本的欧氏距离
+    distances = np.sqrt(((dataSet - input_vector) ** 2).sum(axis=1))
+    # 获取距离最近的k个样本的索引
+    sortedDistIndices = distances.argsort()
+    # 获取这些样本对应的标签
+    classCount = {}
+    for i in range(k):
+        voteIlabel = labels[sortedDistIndices[i]]
+        classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1
+    # 选择票数最多的类别作为预测标签
+    sortedClassCount = sorted(classCount.items(), key=lambda x: x[1], reverse=True)
+    return sortedClassCount[0][0]

 def load_dataset(directory):
-    fileList = os.listdir(directory)
-    m = len(fileList)
-    dataSet = np.zeros((m, 1024))
-    labels = []
-    for i, file in enumerate(fileList):
-        path = os.path.join(directory, file)
-        vector = img_to_vector(path)
-        dataSet[i, :] = vector
-        label = int(file.split('_')[0])
-        labels.append(label)
-    return dataSet, np.array(labels)
+    fileList = os.listdir(directory)
+    m = len(fileList)
+    dataSet = np.zeros((m, 1024))
+    labels = []
+    for i, file in enumerate(fileList):
+        path = os.path.join(directory, file)
+        vector = img_to_vector(path)
+        dataSet[i, :] = vector
+        label = int(file.split('_')[0])
+        labels.append(label)
+    return dataSet, np.array(labels)

 def img_to_vector(filename):
-    returnVect = np.zeros(1024)
-    with open(filename, 'r') as fr:
-        lines = fr.read().replace('\n', '')
-        for i in range(32):
-            for j in range(32):
-                returnVect[i*32 + j] = int(lines[i*32 + j])
-    return returnVect
+    returnVect = np.zeros(1024)
+    with open(filename, 'r') as fr:
+        lines = fr.read().replace('\n', '')
+        for i in range(32):
+            for j in range(32):
+                returnVect[i*32 + j] = int(lines[i*32 + j])
+    return returnVect
 if __name__ == "__main__":
-    # 使用绝对路径加载数据集
-    directory = r"C:\Users\86199\Desktop\trainingDigits"
-    train_data, train_labels = load_dataset(directory)
+    # 使用绝对路径加载数据集
+    directory = r"C:\Users\HP\Desktop\trainingDigits"
+    train_data, train_labels = load_dataset(directory)
+
    # 确保 test_data 字符串长度为1024
-    test_data = (
-        "00000000000000110000000000000000"
-        "00000000000011111000000000000000"
-        "00000000000011111100000000000000"
-        "00000000001111111110000000000000"
-        "00000000001111111110000000000000"
-        "00000000001111111100000000000000"
-        "00000000011111111000000000000000"
-        "00000000111111111000000000000000"
-        "00000000111111110000000000000000"
-        "00000000111111110000000000000000"
-        "00000000111111110000000000000000"
-        "00000000111111110000000000000000"
-        "00000000111111100000000000000000"
-        "00000001111111100000000000000000"
-        "00000001111111100000000000000000"
-        "00000001111111100000000000000000"
-        "00000001111111111111000000000000"
-        "00000001111111111111111000000000"
-        "00000001111111111111111100000000"
-        "00000011111111111111111100000000"
-        "00000011111111111111111100000000"
-        "00000001111111111111111110000000"
-        "00000001111111111111111110000000"
-        "00000000111111111100111111100000"
-        "00000000111111111000111111100000"
-        "00000000011111111100111111110000"
-        "00000000011111111111111111000000"
-        "00000000001111111111111111000000"
-        "00000000000111111111111111000000"
-        "00000000000011111111111100000000"
-        "00000000000000111111111000000000"
-        "00000000000000000111100000000000"
-    ).replace("\n", "")  # 移除换行符
+    test_data = (
+        "00000000000000110000000000000000"
+        "00000000000011111000000000000000"
+        "00000000000011111100000000000000"
+        "00000000001111111110000000000000"
+        "00000000001111111110000000000000"
+        "00000000001111111100000000000000"
+        "00000000011111111000000000000000"
+        "00000000111111111000000000000000"
+        "00000000111111110000000000000000"
+        "00000000111111110000000000000000"
+        "00000000111111110000000000000000"
+        "00000000111111110000000000000000"
+        "00000000111111100000000000000000"
+        "00000001111111100000000000000000"
+        "00000001111111100000000000000000"
+        "00000001111111100000000000000000"
+        "00000001111111111111000000000000"
+        "00000001111111111111111000000000"
+        "00000001111111111111111100000000"
+        "00000011111111111111111100000000"
+        "00000011111111111111111100000000"
+        "00000001111111111111111110000000"
+        "00000001111111111111111110000000"
+        "00000000111111111100111111100000"
+        "00000000111111111000111111100000"
+        "00000000011111111100111111110000"
+        "00000000011111111111111111000000"
+        "00000000001111111111111111000000"
+        "00000000000111111111111111000000"
+        "00000000000011111111111100000000"
+        "00000000000000111111111000000000"
+        "00000000000000000111100000000000"
+    ).replace("\n", "")  # 移除换行符

-    # 将 test_data 转换为 NumPy 数组，并确保其长度为1024
-    test_vector = np.array([int(char) for char in test_data])[:1024].reshape(1, 1024)
+    # 将 test_data 转换为 NumPy 数组，并确保其长度为1024
+    test_vector = np.array([int(char) for char in test_data])[:1024].reshape(1, 1024)

-    # 确保数组大小为1024后再进行分类
-    predicted_label = classify(test_vector, train_data, train_labels)
+    # 确保数组大小为1024后再进行分类
+    predicted_label = classify(test_vector, train_data, train_labels)

-    print(f"预测的标签: {predicted_label}")
+    print(f"预测的标签: {predicted_label}")