k-medoids without numpy, 无第三方库的实现

k-medoids.py

from math import dist
from pathlib import Path
from random import choice, sample
from typing import List, Tuple

Point = Tuple[float, ...]


def kmedoids(点集: List[Point], k: int):
    初始点集合 = sample(点集, k)
    结果字典 = {i: [x] for i, x in enumerate(初始点集合)}
    代表字典 = dict(enumerate(初始点集合))
    样本点的集合 = set(点集) - set(初始点集合)
    for 点 in 样本点的集合:
        最近的编号 = min(range(k), key=lambda i: dist(点, 代表字典[i]))
        结果字典[最近的编号].append(点)
        替换后的点 = min(
            结果字典[最近的编号],
            key=lambda p: sum(dist(p, pi) for pi in 结果字典[最近的编号]),
        )
        代表字典[最近的编号] = 替换后的点
    return 结果字典


if __name__ == "__main__":

    def 读取文件():
        with open(Path(__file__).parent.absolute() / "data.txt") as 文件:
            yield from (tuple(map(float, 每一行.split())) for 每一行 in 文件)

    print(kmedoids(list(读取文件()), 5))