homework15/main.cpp at master · otushomework/homework15 · 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
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
#include <cstdlib>
#include <iostream>
#include <vector>

#define _DEBUG

#include <dlib/clustering.h>
#include <dlib/rand.h>

typedef dlib::matrix<double,2,1> sample_type;
typedef dlib::radial_basis_kernel<sample_type> kernel_type;

auto split(const std::string &str, char d)
{
    std::vector<std::string> r;

    std::string::size_type start = 0;
    std::string::size_type stop = str.find_first_of(d);
    while(stop != std::string::npos)
    {
        r.push_back(str.substr(start, stop - start));

        start = stop + 1;
        stop = str.find_first_of(d, start);
    }

    r.push_back(str.substr(start));

    return r;
}

#ifdef _DEBUG
void generateData(std::string filename)
{
    std::ofstream file(filename);

    dlib::rand random_generator;

    const int num_points_in_cluster = 100;
    double x, y;
    double radius = 0.5;

    for (int i = 0; i < num_points_in_cluster; ++i)
    {
        double sign = 1;
        if (random_generator.get_random_double() < 0.5)
            sign = -1;

        x = 2 * radius * random_generator.get_random_double() - radius;
        y = sign * sqrt(radius * radius - x * x);

        file << x << ";" << y << std::endl;
    }

    radius = 10.0;
    for (long i = 0; i < num_points_in_cluster; ++i)
    {
        double sign = 1;
        if (random_generator.get_random_double() < 0.5)
            sign = -1;

        x = 2 * radius * random_generator.get_random_double() - radius;
        y = sign * sqrt(radius * radius - x * x);

        file << x << ";" << y << std::endl;
    }

    radius = 4.0;
    for (long i = 0; i < num_points_in_cluster; ++i)
    {
        double sign = 1;
        if (random_generator.get_random_double() < 0.5)
            sign = -1;
        x = 2 * radius * random_generator.get_random_double() - radius;
        y = sign * sqrt(radius * radius - x * x);

        x += 25;
        y += 25;

        file << x << ";" << y << std::endl;
    }
}
#endif

//cat kkmeans_ex.txt | ./kkmeans 3
int main(int argc, char * argv[])
{
    if (argc != 2)
    {
        return 1;
    }

#ifdef _DEBUG
    // ./kkmeans generate
    if (std::string(argv[1]) == "generate")
    {
        generateData("kkmeans_ex.txt");
        return 0;
    }
#endif

    dlib::kcentroid<kernel_type> kc(kernel_type(0.1),0.01, 8);
    dlib::kkmeans<kernel_type> test(kc);

    std::vector<sample_type> samples;
    std::vector<sample_type> initialCenters;

    int clustersCount = atoi(argv[1]);

    for(std::string line; std::getline(std::cin, line);)
    {
        auto coords = split(line, ';');
        if ( coords.size() != 2 )
            continue;

        samples.push_back(sample_type(std::stod(coords.at(0)), std::stod(coords.at(1))));
#ifdef _DEBUG
        std::cout << "src " << coords.at(0) << " " << coords.at(1) << " " << std::stod(coords.at(0)) << " " << std::stod(coords.at(1)) << std::endl;
#endif
    }

    test.set_number_of_centers(clustersCount);
    dlib::pick_initial_centers(clustersCount, initialCenters, samples, test.get_kernel());
    test.train(samples, initialCenters);

    for (unsigned long i = 0; i < samples.size(); ++i)
    {
        std::cout << samples[i](0) << ";" << samples[i](1) << ";" << test(samples[i]) << std::endl;
    }

    return 0;
}