OPTICS-Python/main.py at master · BigTailFox/OPTICS-Python · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
# coding=utf-8

# created by leoherz, 2020/04/17
# an example of OPTICS, on data set:

import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from optics import OPTICS

COLORS = ['plum', 'palevioletred', 'sandybrown', 'lightcoral', 'crimson',
          'mediumaquamarine', 'steelblue', 'slateblue', "cadetblue", 'darkcyan'
          'cornflowerblue', 'seagreen', 'lightsalmon', 'burlywood', 'forestgreen']

if __name__ == "__main__":
    df = pd.read_table("cluster_data_set/spiral.txt",
                       delim_whitespace=True)
    print(df.describe())
    fig0 = plt.figure("ground truth")
    for i in range(len(df)):
        plt.scatter(df.iloc[i, 0], df.iloc[i, 1],
                    color=COLORS[df.iloc[i, -1] % len(COLORS)])

    # remove label for clustering task.
    df1 = df.iloc[:, :-1]
    # run optics cluster.
    model = OPTICS(df1)
    model.optics(Eps=10000, MinPts=2)
    # show ordered reachable distance and core distance.
    fig1 = plt.figure("ordered rd with ground truth label")
    index = np.linspace(0, model.len, model.len)
    rd = []
    cd = []
    labels = []
    for i in range(model.len):
        rd.append(model.reachable_distances[model.result_queue[i]])
        cd.append(model.core_distances[model.result_queue[i]])
        labels.append(
            COLORS[df.iloc[model.result_queue[i], -1] % len(COLORS)])
    plt.bar(index, rd, color=labels)
    plt.plot(index, rd, color='black', linewidth=0.3)
    plt.ylim(0, 5)
    plt.xlim(-5, model.len+5)
    fig2 = plt.figure("ordered rd and cd")
    plt.fill_between(index, 0, cd, alpha=0.25, color='crimson')
    plt.fill_between(index, 0, rd, alpha=0.2, color='cornflowerblue')
    plt.xlim(-5, model.len+5)
    plt.ylim(0, 5)

    # extract cluster result and visualize
    model.cluster_extract(1.5)
    fig3 = plt.figure("clustering result")
    cluster_labels = []
    for i in range(model.len):
        c = model.cluster_labels[i]
        if c == -1:
            cluster_labels.append("darkgrey")
        else:
            cluster_labels.append(
                COLORS[c % len(COLORS)])
    plt.scatter(df.iloc[:, 0], df.iloc[:, 1], color=cluster_labels)

    fig4 = plt.figure("rd with clustering labels")
    plt.bar(index, rd, color=[cluster_labels[i] for i in model.result_queue])
    plt.plot(index, rd, color='black', linewidth=0.3)
    plt.xlim(-5, model.len + 5)
    plt.ylim(0, 5)

    plt.show()