聚类算法学习----之----sklearn.cluster.DBSCAN

 

 

class DBSCAN(BaseEstimator, ClusterMixin):
    """Perform DBSCAN clustering from vector array or distance matrix.

    DBSCAN - Density-Based Spatial Clustering of Applications with Noise.
    Finds core samples of high density and expands clusters from them.
    Good for data which contains clusters of similar density.

    Read more in the :ref:`User Guide `.

    Parameters
    ----------
    eps : float, optional 同一个簇中样本的最大距离 默认：0.5

    min_samples : int, optional 一个簇中的至少需要包含的样本数 默认：5

    metric : string, or callable 最距离公式，可以用默认的欧式距离，还可以自己定义距离函数 默认：euclidean

    metric_params : dict, optional 默认：None
        Additional keyword arguments for the metric function.

    algorithm : {'auto', 'ball_tree', 'kd_tree', 'brute'}, optional 最近邻搜索算法参数 默认：auto
        brute是蛮力实现，
        kd_tree是KD树实现，
        ball_tree是球树实现,
        auto则会在三种算法中做权衡，选择一个拟合最好的最优算法

    leaf_size : int, optional (default = 30) 使用KD树或者球树时,停止建子树的叶子节点数量的阈值 默认：30
    （最近邻搜索算法的参数）

    p : float, optional 只用于闵可夫斯基距离和带权重闵可夫斯基距离中p值的选择 默认：None
        p=1为曼哈顿距离， p=2为欧式距离

    n_jobs : int, optional (default = 1) 使用的进程数量，默认为：1
        若值为 -1，则用所有的CPU进行运算

      Attributes
    ----------
    core_sample_indices_ : 核心点的索引
        因为labels_不能区分核心点还是边界点，所以需要用这个索引确定核心点

    components_ : array, shape = [n_core_samples, n_features]#核心点

    labels_ : array, shape = [n_samples] 每个点所属集群的标签，-1代表噪声
    """

    def __init__(self, eps=0.5, min_samples=5, metric='euclidean',
                 metric_params=None, algorithm='auto', leaf_size=30, p=None,
                 n_jobs=1):
        self.eps = eps
        self.min_samples = min_samples
        self.metric = metric
        self.metric_params = metric_params
        self.algorithm = algorithm
        self.leaf_size = leaf_size
        self.p = p
        self.n_jobs = n_jobs

    def fit(self, X, y=None, sample_weight=None):
        """Perform DBSCAN clustering from features or distance matrix.

        Parameters
        ----------
        X : 需要分类的数据
        sample_weight : 样本点的权重
        y : Ignored

        """
        X = check_array(X, accept_sparse='csr')
        clust = dbscan(X, sample_weight=sample_weight,
                       **self.get_params())
        self.core_sample_indices_, self.labels_ = clust
        if len(self.core_sample_indices_):
            # fix for scipy sparse indexing issue
            self.components_ = X[self.core_sample_indices_].copy()
        else:
            # no core samples
            self.components_ = np.empty((0, X.shape[1]))
        return self

    def fit_predict(self, X, y=None, sample_weight=None):
        """Performs clustering on X and returns cluster labels.

        Parameters
        ----------
        X : array or sparse (CSR) matrix of shape (n_samples, n_features), or \
                array of shape (n_samples, n_samples)
            A feature array, or array of distances between samples if
            ``metric='precomputed'``.
        sample_weight : array, shape (n_samples,), optional
            Weight of each sample, such that a sample with a weight of at least
            ``min_samples`` is by itself a core sample; a sample with negative
            weight may inhibit its eps-neighbor from being core.
            Note that weights are absolute, and default to 1.

        y : Ignored

        Returns
        -------
        y : ndarray, shape (n_samples,)
            cluster labels
        """
        self.fit(X, sample_weight=sample_weight)
        return self.labels_

 

 

 

测试：

 

import pandas as pd
from sklearn.cluster import DBSCAN
def main():
    stopList= [{'id': '105792','lat': 28.571906,'lng': 112.337788},
               {'id': '55792','lat': 28.573678,'lng': 112.381103},
               { 'id': '500792','lat': 28.571915,'lng': 112.337533},
               { 'id': '5000105792','lat': 28.573978,'lng': 112.35765},
                { 'id': '0105792','lat': 28.572656,'lng': 112.3366},
               {'id': '50005792', 'lat': 28.578011, 'lng': 112.330688},
               {'id': '5000105792', 'lat': 28.572228, 'lng': 112.335841},
               {'id': '500105792', 'lat': 28.57849, 'lng': 112.3338},
               {'id': '5005792', 'lat': 28.57239, 'lng': 112.336491},
               {'id': '105792', 'lat': 28.577943, 'lng': 112.330995},
               {'id': '792', 'lat': 28.571921, 'lng': 112.337783},
               {'id': '505792', 'lat': 28.572401, 'lng': 112.3359},
               {'id': '500092', 'lat': 28.569629, 'lng': 112.34005},
               {'id': '50092', 'lat': 28.588048, 'lng': 112.337783},
               {'id': '505792', 'lat': 28.572035, 'lng': 112.335683},
               {'id': '05792', 'lat': 28.560938, 'lng': 112.378183},
               {'id': '55792', 'lat': 28.544781, 'lng': 112.494936},
               {'id': '505792', 'lat': 28.572296, 'lng': 112.336288},
               {'id': '505792', 'lat': 28.571951, 'lng': 112.337806},
               {'id': '55792', 'lat': 28.571551, 'lng': 112.32685}]

    print('共有%d个点'%len(stopList))
    initdata = pd.DataFrame(stopList)
    scatterData = initdata[['lat', 'lng']]  ## 选择需要显示的字段:经纬度
    modle = DBSCAN(eps=0.0003,min_samples=2)  # dbscan
    resluts = modle.fit(scatterData)  # 聚类
    labels = resluts.labels_  # 每个点所属簇的索引构成的列表
    print('labels\n', labels)
    print('core_sample_indices_\n',resluts.core_sample_indices_)#核心点的索引
    print('components_\n', resluts.components_)  # 核心点


if __name__ == '__main__':
    main()
    

 

 

 

运行结果：

共有20个点
labels
 [ 0 -1  0 -1  1 -1  2 -1  1 -1  0  2 -1 -1  2 -1 -1  1  0 -1]
core_sample_indices_
 [ 0  2  4  6  8 10 11 14 17 18]
components_
 [[  28.571906  112.337788]
 [  28.571915  112.337533]
 [  28.572656  112.3366  ]
 [  28.572228  112.335841]
 [  28.57239   112.336491]
 [  28.571921  112.337783]
 [  28.572401  112.3359  ]
 [  28.572035  112.335683]
 [  28.572296  112.336288]
 [  28.571951  112.337806]]