-
Notifications
You must be signed in to change notification settings - Fork 0
/
trainer.h
122 lines (94 loc) · 3.12 KB
/
trainer.h
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
#ifndef TRAINER_H
#define TRAINER_H
#include <boost/serialization/optional.hpp>
#include <boost/serialization/shared_ptr.hpp>
#include <boost/serialization/deque.hpp>
#include <boost/numeric/ublas/vector.hpp>
#include <forward_list>
#include <memory>
#include <boost/optional/optional.hpp>
#include <deque>
#include <string>
#include <vector>
#include "division_plane.h"
#include "shape.h"
#include "vector_entry.h"
class MatrixBranch;
class VectorLoader;
class Trainer
{
public:
Trainer();
Trainer(int vector_size_, std::deque<VectorEntry> dataset_, int depth_ = 0,
MatrixBranch* matrix_branch_ = NULL);
Trainer(VectorLoader* loader);
VectorEntry generate_random(Shape shape) const;
void populate(int fakes, Shape shape);
void subdivide(int max_depth, bool delete_data);
int count_trues() const;
void show_tree(std::string prefix = "");
bool is_pure() const;
bool is_counted() const;
int get_true_count() const;
int get_fake_count() const;
void recalculate_minmax();
double get_volume() const;
int get_vector_size() const;
std::vector<std::deque<bool> > get_leaves() const;
std::shared_ptr<Trainer> cut_leaf(std::deque<bool> history);
void fill_leaf(std::deque<bool> history, int count, Shape shape);
void give_matrix_stock(std::forward_list<std::shared_ptr<MatrixBranch> >* ms);
void normalise_dataset();
protected:
Trainer(int vector_size_, int depth_, MatrixBranch* matrix_branch_,
std::forward_list<std::shared_ptr<MatrixBranch> >* ms);
static int trainer_count;
std::deque<VectorEntry> dataset;
void calculate_centres();
void calculate_division(bool delete_data);
void purify();
bool get_random_half() const;
void add_fake(VectorEntry& vec);
void categorise(VectorEntry& vec) const;
std::vector<std::deque<bool> > get_leaves_recursive(std::deque<bool>& history) const;
std::shared_ptr<Trainer> cut_leaf_recursive(std::deque<bool>& history);
void fill_leaf_recursive(std::deque<bool>& history, VectorEntry& fake);
private:
int ID;
int vector_size;
int depth;
boost::numeric::ublas::vector<double> minimum;
boost::numeric::ublas::vector<double> maximum;
boost::numeric::ublas::vector<double> true_centre;
boost::numeric::ublas::vector<double> fake_centre;
int true_count;
int fake_count;
boost::optional<DivisionPlane> division;
std::shared_ptr<Trainer> positive;
std::shared_ptr<Trainer> negative;
MatrixBranch* matrix_branch;
std::forward_list<std::shared_ptr<MatrixBranch> >* matrix_stock;
std::string spaces() const;
void conservate();
private:
friend class boost::serialization::access;
template<class Archive>
void serialize(Archive& ar, const unsigned int version) {
ar & dataset;
ar & ID;
ar & vector_size;
ar & depth;
ar & minimum;
ar & maximum;
ar & true_centre;
ar & fake_centre;
ar & true_count;
ar & fake_count;
ar & division;
ar & positive;
ar & negative;
ar & matrix_branch;
//ar & matrix_stock;
}
};
#endif // TRAINER_H