summary refs log tree commit diff stats
path: root/lib/wrappers/libsvm.nim
blob: ac5889410060c9bc6977d5de284a35b3918733bf (plain) (blame)
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
#
#
#            Nim's Runtime Library
#        (c) Copyright 2012 Andreas Rumpf
#
#    See the file "copying.txt", included in this
#    distribution, for details about the copyright.
#

## This module is a low level wrapper for `libsvm`:idx:.

{.deadCodeElim: on.}
const
  LIBSVM_VERSION* = 312

when defined(windows):
  const svmdll* = "libsvm.dll"
elif defined(macosx):
  const svmdll* = "libsvm.dylib"
else:
  const svmdll* = "libsvm.so"

type
  Node*{.pure, final.} = object
    index*: cint
    value*: cdouble

  Problem*{.pure, final.} = object
    L*: cint
    y*: ptr cdouble
    x*: ptr ptr Node

  Type*{.size: sizeof(cint).} = enum
    C_SVC, NU_SVC, ONE_CLASS, EPSILON_SVR, NU_SVR

  KernelType*{.size: sizeof(cint).} = enum
    LINEAR, POLY, RBF, SIGMOID, PRECOMPUTED

  Parameter*{.pure, final.} = object
    typ*: Type
    kernelType*: KernelType
    degree*: cint             # for poly
    gamma*: cdouble           # for poly/rbf/sigmoid
    coef0*: cdouble           # for poly/sigmoid
                              # these are for training only
    cache_size*: cdouble      # in MB
    eps*: cdouble             # stopping criteria
    C*: cdouble               # for C_SVC, EPSILON_SVR and NU_SVR
    nr_weight*: cint          # for C_SVC
    weight_label*: ptr cint   # for C_SVC
    weight*: ptr cdouble      # for C_SVC
    nu*: cdouble              # for NU_SVC, ONE_CLASS, and NU_SVR
    p*: cdouble               # for EPSILON_SVR
    shrinking*: cint          # use the shrinking heuristics
    probability*: cint        # do probability estimates
{.deprecated: [Tnode: Node, Tproblem: Problem, Ttype: Type,
              TKernelType: KernelType, Tparameter: Parameter].}

#
# svm_model
#

type
  Model*{.pure, final.} = object
    param*: Parameter         # parameter
    nr_class*: cint           # number of classes, = 2 in regression/one class svm
    L*: cint                  # total #SV
    SV*: ptr ptr Node         # SVs (SV[l])
    sv_coef*: ptr ptr cdouble # coefficients for SVs in decision functions (sv_coef[k-1][l])
    rho*: ptr cdouble         # constants in decision functions (rho[k*(k-1)/2])
    probA*: ptr cdouble       # pariwise probability information
    probB*: ptr cdouble       # for classification only
    label*: ptr cint          # label of each class (label[k])
    nSV*: ptr cint            # number of SVs for each class (nSV[k])
                              # nSV[0] + nSV[1] + ... + nSV[k-1] = l
                              # XXX
    free_sv*: cint            # 1 if svm_model is created by svm_load_model
                              # 0 if svm_model is created by svm_train
{.deprecated: [TModel: Model].}

proc train*(prob: ptr Problem, param: ptr Parameter): ptr Model{.cdecl,
    importc: "svm_train", dynlib: svmdll.}
proc cross_validation*(prob: ptr Problem, param: ptr Parameter, nr_fold: cint,
                       target: ptr cdouble){.cdecl,
    importc: "svm_cross_validation", dynlib: svmdll.}
proc save_model*(model_file_name: cstring, model: ptr Model): cint{.cdecl,
    importc: "svm_save_model", dynlib: svmdll.}
proc load_model*(model_file_name: cstring): ptr Model{.cdecl,
    importc: "svm_load_model", dynlib: svmdll.}
proc get_svm_type*(model: ptr Model): cint{.cdecl, importc: "svm_get_svm_type",
    dynlib: svmdll.}
proc get_nr_class*(model: ptr Model): cint{.cdecl, importc: "svm_get_nr_class",
    dynlib: svmdll.}
proc get_labels*(model: ptr Model, label: ptr cint){.cdecl,
    importc: "svm_get_labels", dynlib: svmdll.}
proc get_svr_probability*(model: ptr Model): cdouble{.cdecl,
    importc: "svm_get_svr_probability", dynlib: svmdll.}
proc predict_values*(model: ptr Model, x: ptr Node, dec_values: ptr cdouble): cdouble{.
    cdecl, importc: "svm_predict_values", dynlib: svmdll.}
proc predict*(model: ptr Model, x: ptr Node): cdouble{.cdecl,
    importc: "svm_predict", dynlib: svmdll.}
proc predict_probability*(model: ptr Model, x: ptr Node,
                          prob_estimates: ptr cdouble): cdouble{.cdecl,
    importc: "svm_predict_probability", dynlib: svmdll.}
proc free_model_content*(model_ptr: ptr Model){.cdecl,
    importc: "svm_free_model_content", dynlib: svmdll.}
proc free_and_destroy_model*(model_ptr_ptr: ptr ptr Model){.cdecl,
    importc: "svm_free_and_destroy_model", dynlib: svmdll.}
proc destroy_param*(param: ptr Parameter){.cdecl, importc: "svm_destroy_param",
    dynlib: svmdll.}
proc check_parameter*(prob: ptr Problem, param: ptr Parameter): cstring{.
    cdecl, importc: "svm_check_parameter", dynlib: svmdll.}
proc check_probability_model*(model: ptr Model): cint{.cdecl,
    importc: "svm_check_probability_model", dynlib: svmdll.}

proc set_print_string_function*(print_func: proc (arg: cstring) {.cdecl.}){.
    cdecl, importc: "svm_set_print_string_function", dynlib: svmdll.}