DBSCAN

import copy

from sklearn import datasets
import numpy as np
import matplotlib.pyplot as plt
import random


class DBSCAN:
    def __init__(self, dataset, eps, min_point):
        self.dataset = dataset
        self.eps = eps
        self.min_point = min_point
        self.N, self.M = self.dataset.shape

    def get_neighbour_num(self, i):
        N = list()
        temp = np.sum((self.dataset - self.dataset[i]) ** 2, axis=1) ** 0.5
        N = np.argwhere(temp <= eps).flatten().tolist()
        return N

    def train(self):
        '''存放每一个点是否被遍历过'''
        visited = [0] * self.N
        '''存放每一个点属于的簇'''
        cluster = [0] * self.N
        k = 1
        '''存放每一个点的邻居'''
        neighbour_list = []
        for i in range(self.N):
            neighbour = self.get_neighbour_num(i)
            neighbour_list.append(neighbour)
        while True:
            if visited.count(0) == 0:
                break
            '''随机选取中心簇'''
            index = random.choice([i for i in range(len(visited)) if visited[i] == 0])
            if len(neighbour_list[index]) < self.min_point:
                visited[index] = -1
                cluster[index] = -1
            else:
                Queen = [index]
                cluster_list = [index]
                while Queen:
                    start = Queen[0]
                    if len(neighbour_list[start]) >= self.min_point:
                        for i in range(len(neighbour_list[start])):
                            if neighbour_list[start][i] not in cluster_list and neighbour_list[start][i] not in Queen:
                                Queen.append(neighbour_list[start][i])
                    cluster_list.append(start)
                    visited[start] = -1
                    Queen.pop(0)
                for i in range(len(cluster_list)):
                    cluster[cluster_list[i]] = k
                k += 1
        return cluster


if __name__ == '__main__':
    x1, y1 = datasets.make_circles(n_samples=2000, factor=.6, noise=.02)
    x2, y2 = datasets.make_blobs(n_samples=400, n_features=2, centers=[[1.2, 1.2]], cluster_std=[[.1]], random_state=9)
    dataset = np.concatenate((x1, x2))
    target = np.concatenate((y1, y2))
    eps = 0.08
    min_point = 10
    model = DBSCAN(dataset, eps, min_point)
    clusters = model.train()
    print(list(target))
    print(clusters)
    plt.scatter(np.array([dataset[i][0] for i in range(len(clusters)) if clusters[i] == 1]),
                np.array([dataset[i][1] for i in range(len(clusters)) if clusters[i] == 1]))
    plt.scatter(np.array([dataset[i][0] for i in range(len(clusters)) if clusters[i] == 2]),
                np.array([dataset[i][1] for i in range(len(clusters)) if clusters[i] == 2]))
    plt.scatter(np.array([dataset[i][0] for i in range(len(clusters)) if clusters[i] == 3]),
                np.array([dataset[i][1] for i in range(len(clusters)) if clusters[i] == 3]))
    plt.show()