Experiment-kNN/function.py

import numpy as np
import os
from sklearn.neighbors import KDTree

"""
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。

请在这里填写学号和姓名，不填写即成绩无效。
"""

id = "202210423213"
name = "熊浩燃"


def classify(input_vector, dataSet, labels, k=1, distance_metric='euclidean', weighted=False):
    """
    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，默认值为1。
    - distance_metric (str): 距离度量方式，支持'euclidean'和'manhattan'。
    - weighted (bool): 是否使用加权投票。

    Returns:
    - int: 预测的分类标签。
    """
    # 选择距离度量
    if distance_metric == 'euclidean':
        distances = np.linalg.norm(dataSet - input_vector, axis=1)
    elif distance_metric == 'manhattan':
        distances = np.sum(np.abs(dataSet - input_vector), axis=1)
    else:
        raise ValueError(f"Unsupported distance metric: {distance_metric}")

    # 获取距离最近的 k 个样本的索引
    nearest_neighbors_indices = distances.argsort()[:k]
    # 获取这些样本的标签
    nearest_labels = np.array(labels)[nearest_neighbors_indices]

    if weighted:
        # 使用加权投票：距离越近，权重越大
        nearest_distances = distances[nearest_neighbors_indices]
        weights = 1 / (nearest_distances + 1e-5)  # 防止距离为0
        weighted_vote = np.bincount(nearest_labels, weights=weights)
        predicted_label = weighted_vote.argmax()
    else:
        predicted_label = np.bincount(nearest_labels).argmax()

    return predicted_label


def classify_with_kdtree(input_vector, dataSet, labels, k=1):
    """
    使用 KD-tree 进行加速的 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，默认值为1。

    Returns:
    - int: 预测的分类标签。
    """
    tree = KDTree(dataSet)  # 构建 KD-tree
    dist, indices = tree.query(input_vector.reshape(1, -1), k=k)  # 查询最近邻
    nearest_labels = np.array(labels)[indices[0]]
    predicted_label = np.bincount(nearest_labels).argmax()
    return predicted_label


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 = []
    data_list = []
    for filename in os.listdir(directory):
        if filename.endswith(".txt"):  # 确保读取的是 txt 文件
            label = int(filename.split('_')[0])
            labels.append(label)
            vector = img_to_vector(os.path.join(directory, filename))
            data_list.append(vector)
    return np.array(data_list), np.array(labels)


def img_to_vector(filename):
    """
    将二值图像转换为向量。
    Args:
    - filename (str): 图像文件路径。

    Returns:
    - np.array: 二值图像向量。
    """
    with open(filename, 'r') as f:  # 确保使用文本模式读取
        lines = f.readlines()
    vector = np.zeros((1024,), dtype=int)  # 将其改为 (1024,) 的一维数组
    for i in range(32):
        for j in range(32):
            vector[32 * i + j] = int(lines[i][j])
    return vector


if __name__ == "__main__":
    # 测试代码
    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"
    )

    test_vector = np.array([int(char) for char in test_data]).reshape(1, 1024)

    # 使用默认的欧氏距离和不加权的投票机制
    predicted_label = classify(test_vector, train_data, train_labels)
    print(f"预测的标签（普通分类）: {predicted_label}")

    # 使用 KD-tree 进行加速的分类
    predicted_label_kdtree = classify_with_kdtree(test_vector, train_data, train_labels)
    print(f"预测的标签KD-tree 加速: {predicted_label_kdtree}")