写一个不使用kdtree或balltree的DBSCAN聚类的python代码

下面是一个简单的实现DBSCAN聚类的Python代码：

class DBSCAN:
    def __init__(self, eps, min_samples):
        self.eps = eps
        self.min_samples = min_samples

    def fit(self, X):
        self.labels_ = np.zeros(len(X))
        self.visited_ = np.zeros(len(X))

        label = 0

        for i in range(len(X)):
            if not self.visited_[i]:
                self.visited_[i] = 1

                neighbors = self.region_query(X[i])

                if len(neighbors) < self.min_samples:
                    self.labels_[i] = -1  # noise
                else:
                    self.expand_cluster(X, i, neighbors, label)
                    label += 1

        return self.labels_

    def expand_cluster(self, X, i, neighbors, label):
        self.labels_[i] = label

        for j in neighbors:
            if not self.visited_[j]:
                self.visited_[j] = 1

                new_neighbors = self.region_query(X[j])

                if len(new_neighbors) >= self.min_samples:
                    neighbors = np.concatenate((neighbors, new_neighbors))

            if self.labels_[j] == 0:
                self.labels_[j] = label

    def region_query(self, X):
        return np.where(np.linalg.norm(self.eps - X, axis=1) <= self.eps)[0]

该代码中的DBSCAN类具有fit方法，该方法接受数据矩阵作为输入，并返回一个长度为n的标签数组，其中n是输入数据的样本数。标签为-1表示噪声点，而其他标签表示相应的聚类。该代码使用欧几里得距离进行距离计算。