'' ' DBSCAN Algorithm: From a sample of any sample space, given in advance to make the radius of the circle, where the circle is the sample in the sample are treated in the same cluster, these samples were in the circle continue to do as the center circle, repeat the above process, expanding the size of the circle in the sample until no new sample is added so far, so far to obtain a cluster. In the remaining samples, the above process is repeated until all samples in the sample is exhausted space. Characteristics of DBSCAN: 1. previously given radius will affect the final clustering result can be selected by means of the profile coefficients preferred aspect. The formation of the clusters, the sample is subdivided into the following three categories: the outer periphery of the sample: the other samples are collected to a cluster, but not be able to sample a new sample is introduced. Isolated Sample: cluster number of samples below the lower limit of the set, is not called clustering, otherwise referred to as isolated samples. Core sample: sample in addition to the outer periphery of the sample and the isolated samples. '' ' Import numpy AS NP Import matplotlib.pyplot AS MP Import sklearn.cluster SC AS Import sklearn.metrics SM AS # to read data, the drawn image x = np.loadtxt (' ./Ml_data/multiple3.txt ' , the unpack = False, DTYPE = ' F8 ' , DELIMITER = ' , ' ) Print (x.shape) # select the optimal radius epsilons np.linspace = (0.3, 1.0,. 8 ) Models , Scores = [], [] for Epsilon in epsilons: # construct models for each radius DBSCAN model = sc.DBSCAN (EPS = Epsilon, min_samples =. 5 ) model.fit (X) Score = sm.silhouette_score (X, model .labels_, SAMPLE_SIZE = len (X), Metric = ' Euclidean ' ) models.append (Model) scores.append (Score) index = np.array (Scores) .argmax () best_model = Models [index] best_eps = epsilons [index] BEST_SCORE = Scores [index] Print (best_eps, best_model, BEST_SCORE) # -product algorithm acquired DBSAN subscript core samples core_indices = best_model.core_sample_indices_ core_mask = np.zeros (len (X), DTYPE = ' BOOL ' ) core_mask [core_indices] = True Print (core_mask) # obtain sample mask isolated offset_mask = best_model.labels_ = -1 = Print (OFFSET_MASK) # Get the mask outer periphery of the sample p_mask = ~ (core_mask | OFFSET_MASK) Print (p_mask) # outputs the core coordinates, the isolated point coordinates and the outer peripheral point coordinates Print ( ' core point coordinate: ' , X [core_mask], On Sep = ' \ n- ' ) Print ( ' isolated point coordinates: ' , X [OFFSET_MASK], On Sep = ' \ n- ' ) Print ( ' peripheral point coordinates: ' , X [p_mask], On Sep = ' \ n- ' ) # Paint sample display data mp.figure ( ' DBSCAN ' , facecolor ='lightgray') mp.title('DBSCAN', fontsize=16) mp.xlabel('X', fontsize=14) mp.ylabel('Y', fontsize=14) mp.tick_params(labelsize=10) mp.scatter(x[core_mask][:, 0], x[core_mask][:, 1], s=80, c=best_model.labels_[core_mask], label='core') mp.scatter(x[offset_mask][:, 0], x[offset_mask][:, 1], s=80, c=best_model.labels_[offset_mask], label='offset') mp.scatter(x[p_mask][:, 0], x[p_mask][:, 1], s=80, c=best_model.labels_[p_mask], label='p') mp.legend() mp.show() 输出结果： (200, 2) 0.7999999999999999 DBSCAN(algorithm='auto', eps=0.7999999999999999, leaf_size=30, metric='euclidean', metric_params=None, min_samples=5, n_jobs=None, p=None) 0.41779833446624154 [False True True True True True True True True True False True True True True True True True True False True True True True True True False True True True True True True True True True True True True True True True True True True True False True True True True True True True True True False True True True True True True True True True True True True True True True True True True True False True True True True True True True True True True True True True True True True True True True False True True True True True True True True True True False True True True True True True True True False True True True True True True False True True False True True True False False True True True True False True True True True True True False True False True True False True True True False True True True True True False False True True True True False True True False True True True True True True True True True True True True True True True True True True True True True True True True True True False True True True True True] [False False False False False False False False False False False False False False False False False False False False False False False False False False True False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False True False False False False False False False False False False False True False False False False False False False False False False False False False False False True False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False False] [ True False False False False False False False False False True False False False False False False False False True False False False False False False False False False False False False False False False False False False False False False False False False False False True False False False False False False False False False True False False False False False False False False False False False False False False False False False False False True False False False False False False False False False False False False False False False False False False False True False False False False False False False False False False True False False False False False False False False True False False False False False False True False False True False False False True True False False False False False False False False False False False True False True False False False False False False True False False False False False True True False False False False False False False True False False False False False False False False False False False False False False false false false false false false false false false false false false false false false false True false] 核心点坐标: [[ 2.84 3.16 ] [ 4.74 1.84 ] [ 6:36 4.89 ] [ 1.77 1:55 ] [ 4:29 6.74 ] [ 5.78 0.95 ] [ 8:36 5.2 ] [ 1.76 2.9 ] [3.2 4.98] [6.02 6.12] [2.91 1.74] [3.81 6.2 ] [5.67 2.19] [6.93 5.8 ] [1.84 1.36] [3.5 5.6 ] [5.77 1.54] [2.1 1.68] [3.98 4.85] [5.64 3.35] [7.59 6.79] [1.92 3.93] [3.93 5.7 ] [7.24 6.23] [1.36 1.47] [1.78 5.39] [5.93 1.19] [7.34 5.58] [1.86 1.99] [3.2 5.19] [5.62 2.45] [6.31 5.37] [1.49 1.8 ] [2.13 5.29] [5.57 2.77] [6.77 5.8 ] [1.79 2.13] [3.71 4.65] [5.46 1.92] [6.53 5.14] [2.06 2.56] [4.03 4.89] [7.25 4.27] [2.77 2.91] [2.98 4.4 ] [5.21 3.07] [6.54 4.05] [1.59 1.04] [1.65 5.41] [6.53 1.76] [6.77 5.59] [2.21 4.57] [5.58 1.88] [7.61 4.95] [1.8 1.25] [2.77 5.71] [5.13 2.4 ] [7.14 5.6 ] [1.83 1.26] [2.78 5.67] [5.76 2.64] [6.71 6.01] [2.64 1.86] [3.21 5.09] [7.23 1.22] [6.47 5.5 ] [2.46 2.51] [1.92 3.94] [5.64 2.76] [7.16 6.08] [3.52 4.25] [5.41 3.55] [7.9 5.22] [1.99 2.64] [3.46 4.66] [6.56 2.25] [6.89 4.43] [1.4 1.08] [3.43 6.08] [6.64 2.19] [6.26 5.91] [2.15 3.11] [3.18 5.41] [6.73 3.67] [7.36 5.97] [1.55 2.96] [3.99 5.63] [6.68 3.51] [6.34 5.99] [3.91 4.54] [6.21 3.2 ] [6.25 5.33] [2.13 1.91] [3.75 6.23] [4.35 1.49] [7.14 5.64] [1.68 2.67] [3.85 4.84] [6.04 2.68] [1.55 2.26] [1.86 5.29] [6.06 2.12] [7.77 5.17] [1.35 2.16] [3.53 6.11] [6.8 0.61] [6.92 6.02] [3.07 5.09] [6.44 2.07] [7.15 6.15] [1.01 2.56] [2.45 6.24] [6.86 2.54] [2.73 1.7 ] [2.25 5.62] [7.56 5.93] [3.01 1.68] [2.74 4.45] [1.99 0.39] [4.05 6.16] [5.34 2.66] [6.09 5.58] [3.14 5.85] [6.04 0.98] [6.02 5.82] [1.56 1.65] [3.08 6.49] [5.64 3.26] [2.75 1.9 ] [6.47 0.7 ] [7.04 4.83] [2.8 3.42] [7.14 1.72] [6.9 6.86] [4.55 5.17] [6.29 1.99] [7.41 4.88] [2.83 2.29] [3.24 5.83] [1.51 2.95] [2.55 5.51] [6.5 1.72] [7.97 4.62] [2.82 4.83] [5.02 3.53] [1.57 2.67] [1.72 4.05] [6.44 3.28] [7.41 6.85] [1.67 1.5 ] [1.77 5.59] [5.4 2.64] [7.18 5.94] [1.45 1.47] [3.77 5.01] [6.31 0.48] [6.65 5.65] [1.37 2.51] [2.54 4.42] [6.17 1.8 ] [7.29 4.68] [2.44 1.57] [3.46 4.5 ] [5.58 2.19] [6.46 6.23] [2.17 2.4 ] [3.29 5.66] [6.99 1.71] [6.54 4.71] [0.99 2.14] [3.67 4.47] [6.64 6.66] [2.15 0.75] [ 3.26 6.08 ] [ 5.72 1.98 ] [ 7.26 5.35 ]] isolated coordinates: [[ 5.66 -0.33 ] [ 0.16 1.55 ] [ -0.27 2.22 ] [ -0.39 2.28 ]] peripheral point coordinates: [[ 1.96 -0.09 ] [ 5.56 0.34 ] [ 8.68 4.33 ] [ 7.54 0.52 ] [ 0.42 2.75 ] [ 3.29 1.26 ] [ 3.96 1.83 ] [ 5.26 5.89 ] [1.11 3.74] [ 8.17 6.19] [ 7.93 1.84] [ 4.53 3.2 ] [ 8.95 5.7 ] [ 8.26 6.04] [ 3.46 3.57] [ 2.6 0.77] [ 4.26 1.67] [ 4.78 7.25] [ 7.77 7.38] [ 4.98 1.02]]
