FeatureSpace.hpp 4.04 KB
Newer Older
Thomas Purcell's avatar
Thomas Purcell committed
1
2
3
#ifndef FEATURE_SPACE
#define FEATURE_SPACE

Thomas Purcell's avatar
Thomas Purcell committed
4
#include <mpi_interface/MPI_Interface.hpp>
Thomas Purcell's avatar
Thomas Purcell committed
5
6
#include <feature_creation/node/FeatureNode.hpp>
#include <feature_creation/node/operator_nodes/allowed_ops.hpp>
Thomas Purcell's avatar
Thomas Purcell committed
7
#include <feature_creation/node/value_storage/nodes_value_containers.hpp>
Thomas Purcell's avatar
Thomas Purcell committed
8

Thomas Purcell's avatar
Thomas Purcell committed
9
10
#include <boost/serialization/shared_ptr.hpp>

Thomas Purcell's avatar
Thomas Purcell committed
11
#include <iostream>
Thomas Purcell's avatar
Thomas Purcell committed
12
#include <iomanip>
Thomas Purcell's avatar
Thomas Purcell committed
13

Thomas Purcell's avatar
Thomas Purcell committed
14
// namespace mpi = boost::mpi;
15
16
17
18
19
/**
 * @brief Feature Space for SISSO calculations
 * @details Stores and performs all feature calculations for SIS
 *
 */
Thomas Purcell's avatar
Thomas Purcell committed
20
21
class FeatureSpace
{
22
23
24
25
26
    std::shared_ptr<MPI_Interface> _mpi_comm; //!< MPi communicator
    int _max_phi; //!< Maximum rung for the feature creation
    int _n_sis_select; //!< Number of features to select for each dimensions
    int _n_samp; //!< Number of samples
    int _n_feat; //!< Total number of features
Thomas Purcell's avatar
Thomas Purcell committed
27
    int _n_rung_store; //!< Total rungs stored
Thomas Purcell's avatar
Thomas Purcell committed
28

29
    double _max_abs_feat_val; //!< Maximum absolute value for any feature
Thomas Purcell's avatar
Thomas Purcell committed
30

31
    std::vector<int> _start_gen; //!< list of starting index for each generation
Thomas Purcell's avatar
Thomas Purcell committed
32

33
34
    std::vector<double> _prop; //!< property to learn
    std::vector<double> _scores; //!< projection scores for each feature
Thomas Purcell's avatar
Thomas Purcell committed
35

36
37
38
39
    std::vector<std::string> _allowed_ops; //!< list of all allowed operators strings
    std::vector<un_op_node_gen> _un_operators; //!< list of all unary operators
    std::vector<bin_op_node_gen> _bin_operators; //!< list of all binary operators
    std::vector<bin_op_node_gen> _com_bin_operators; //!< list of all commutable binary operators
Thomas Purcell's avatar
Thomas Purcell committed
40

41
42
43
    std::vector<node_ptr> _phi_selected; //!< selected features
    std::vector<node_ptr> _phi; //!< all features
    std::vector<node_ptr> _phi_0; //!< initial feature space
Thomas Purcell's avatar
Thomas Purcell committed
44
public:
45
46
47
48
49
50
51
52
53
54
    /**
     * @brief Constructor for the feature space
     * @details constructs the feature space from an initial set of features and a list of allowed operatiors
     *
     * @param mpi_comm MPI communicator for the calculations
     * @param allowed_ops list of allowed operators
     * @param max_phi highest rung value for the calculation
     * @param n_sis_select number of features to select during each SIS step
     * @param max_abs_feat_val maximum absolute feature value
     */
Thomas Purcell's avatar
Thomas Purcell committed
55
    FeatureSpace(
Thomas Purcell's avatar
Thomas Purcell committed
56
        std::shared_ptr<MPI_Interface> mpi_comm,
Thomas Purcell's avatar
Thomas Purcell committed
57
58
59
60
        std::vector<node_ptr> phi_0,
        std::vector<std::string> allowed_ops,
        int max_phi=1,
        int n_sis_select=1,
Thomas Purcell's avatar
Thomas Purcell committed
61
        int max_store_rung=2,
Thomas Purcell's avatar
Thomas Purcell committed
62
63
64
        double max_abs_feat_val=1e27
    );

65
66
67
68
    /**
     * @brief Generate the full feature set from the allowed operators and initial feature set
     * @details populates phi with all features from an initial set and the allowed operators
     */
Thomas Purcell's avatar
Thomas Purcell committed
69
70
    void generate_feature_space();

71
72
73
    /**
     * @brief Accessor function for _phi_selected
     */
Thomas Purcell's avatar
Thomas Purcell committed
74
    inline std::vector<node_ptr> phi_selected(){return _phi_selected;};
75
76
77
78

    /**
     * @brief Accessor function for _phi
     */
Thomas Purcell's avatar
Thomas Purcell committed
79
    inline std::vector<node_ptr> phi(){return _phi;};
80
81
82
83

    /**
     * @brief Accessor function for _phi_0
     */
Thomas Purcell's avatar
Thomas Purcell committed
84
    inline std::vector<node_ptr> phi0(){return _phi_0;};
85
86
87
88

    /**
     * @brief Accessor function for _scores
     */
Thomas Purcell's avatar
Thomas Purcell committed
89
    inline std::vector<double>& scores(){return _scores;};
90
91
92
93

    /**
     * @brief Accessor function for _mpi_comm
     */
Thomas Purcell's avatar
Thomas Purcell committed
94
    inline std::shared_ptr<MPI_Interface> mpi_comm(){return _mpi_comm;}
95
96
97
98
99
100
101

    /**
     * @brief calculate the projection scores for all features for a given property
     * @details Calculate the projection score based on the Pearson correlation
     *
     * @param prop [description]
     */
102
    void project_r(double* prop, int size);
Thomas Purcell's avatar
Thomas Purcell committed
103

104
105
106
107
108
109
    /**
     * @brief Perform SIS on a feature set with a specified property
     * @details Perform sure-independence screening with either the correct property
     *
     * @param prop The property to calculate SIS from
     */
Thomas Purcell's avatar
Thomas Purcell committed
110
    void sis(std::vector<double>& prop);
111
112
113
114
115
116
117
118
119

    /**
     * @brief Is a feature in this process' _phi?
     *
     * @param ind index
     * @return True if feature is in this _phi
     */
    inline bool feat_in_phi(int ind){return (ind >= _phi[0]->feat_ind()) && (ind <= _phi.back()->feat_ind());}

Thomas Purcell's avatar
Thomas Purcell committed
120
121
122
};

#endif