kmeans.py 文件源码

def k_means(data: Iterable[Point], k:int=2, iterations:int=10) -> List[Point]:
    'Return k-centroids for the data'
    data = list(data)
    centroids = sample(data, k)
    for i in range(iterations):
        labeled = assign_data(centroids, data)
        centroids = compute_centroids(labeled.values())
    return centroids