|
1
|
|
|
import sklearn.svm as skl_svm |
|
|
|
|
|
|
2
|
|
|
|
|
3
|
|
|
from Orange.classification import SklLearner, SklModel |
|
4
|
|
|
from Orange.base import SklLearner as SklLearnerBase |
|
5
|
|
|
from Orange.preprocess import Normalize |
|
6
|
|
|
|
|
7
|
|
|
__all__ = ["SVMLearner", "LinearSVMLearner", "NuSVMLearner", |
|
8
|
|
|
"OneClassSVMLearner"] |
|
9
|
|
|
|
|
10
|
|
|
|
|
11
|
|
|
svm_pps = SklLearner.preprocessors + [Normalize()] |
|
12
|
|
|
|
|
13
|
|
|
|
|
14
|
|
|
class SVMClassifier(SklModel): |
|
15
|
|
|
|
|
16
|
|
|
def predict(self, X): |
|
17
|
|
|
value = self.skl_model.predict(X) |
|
18
|
|
|
if self.skl_model.probability: |
|
19
|
|
|
prob = self.skl_model.predict_proba(X) |
|
20
|
|
|
return value, prob |
|
21
|
|
|
return value |
|
22
|
|
|
|
|
23
|
|
|
|
|
24
|
|
|
class SVMLearner(SklLearner): |
|
25
|
|
|
__wraps__ = skl_svm.SVC |
|
26
|
|
|
__returns__ = SVMClassifier |
|
27
|
|
|
name = 'svm' |
|
28
|
|
|
preprocessors = svm_pps |
|
29
|
|
|
|
|
30
|
|
|
def __init__(self, C=1.0, kernel='rbf', degree=3, gamma=0.0, |
|
|
|
|
|
|
31
|
|
|
coef0=0.0, shrinking=True, probability=False, |
|
|
|
|
|
|
32
|
|
|
tol=0.001, cache_size=200, max_iter=-1, |
|
|
|
|
|
|
33
|
|
|
preprocessors=None): |
|
34
|
|
|
super().__init__(preprocessors=preprocessors) |
|
35
|
|
|
self.params = vars() |
|
36
|
|
|
self.supports_multiclass = True |
|
37
|
|
|
self.supports_weights = True |
|
38
|
|
|
|
|
39
|
|
|
|
|
40
|
|
|
class LinearSVMLearner(SklLearner): |
|
41
|
|
|
__wraps__ = skl_svm.LinearSVC |
|
42
|
|
|
name = 'linear svm' |
|
43
|
|
|
preprocessors = svm_pps |
|
44
|
|
|
|
|
45
|
|
|
def __init__(self, penalty='l2', loss='squared_hinge', dual=True, |
|
|
|
|
|
|
46
|
|
|
tol=0.0001, C=1.0, multi_class='ovr', fit_intercept=True, |
|
|
|
|
|
|
47
|
|
|
intercept_scaling=True, random_state=None, |
|
|
|
|
|
|
48
|
|
|
preprocessors=None): |
|
49
|
|
|
super().__init__(preprocessors=preprocessors) |
|
50
|
|
|
self.params = vars() |
|
51
|
|
|
self.supports_multiclass = True |
|
52
|
|
|
|
|
53
|
|
|
|
|
54
|
|
|
class NuSVMClassifier(SklModel): |
|
55
|
|
|
|
|
56
|
|
|
def predict(self, X): |
|
57
|
|
|
value = self.skl_model.predict(X) |
|
58
|
|
|
if self.skl_model.probability: |
|
59
|
|
|
prob = self.skl_model.predict_proba(X) |
|
60
|
|
|
return value, prob |
|
61
|
|
|
return value |
|
62
|
|
|
|
|
63
|
|
|
|
|
64
|
|
|
class NuSVMLearner(SklLearner): |
|
65
|
|
|
__wraps__ = skl_svm.NuSVC |
|
66
|
|
|
__returns__ = NuSVMClassifier |
|
67
|
|
|
name = 'nu svm' |
|
68
|
|
|
preprocessors = svm_pps |
|
69
|
|
|
|
|
70
|
|
|
def __init__(self, nu=0.5, kernel='rbf', degree=3, gamma=0.0, coef0=0.0, |
|
|
|
|
|
|
71
|
|
|
shrinking=True, probability=False, tol=0.001, cache_size=200, |
|
|
|
|
|
|
72
|
|
|
max_iter=-1, preprocessors=None): |
|
|
|
|
|
|
73
|
|
|
super().__init__(preprocessors=preprocessors) |
|
74
|
|
|
self.params = vars() |
|
75
|
|
|
self.supports_multiclass = True |
|
76
|
|
|
self.supports_weights = True |
|
77
|
|
|
|
|
78
|
|
|
|
|
79
|
|
|
class OneClassSVMLearner(SklLearnerBase): |
|
80
|
|
|
__wraps__ = skl_svm.OneClassSVM |
|
81
|
|
|
name = 'one class svm' |
|
82
|
|
|
preprocessors = svm_pps |
|
83
|
|
|
|
|
84
|
|
|
def __init__(self, kernel='rbf', degree=3, gamma=0.0, coef0=0.0, |
|
|
|
|
|
|
85
|
|
|
tol=0.001, nu=0.5, shrinking=True, cache_size=200, |
|
|
|
|
|
|
86
|
|
|
max_iter=-1, preprocessors=None): |
|
|
|
|
|
|
87
|
|
|
super().__init__(preprocessors=preprocessors) |
|
88
|
|
|
self.params = vars() |
|
89
|
|
|
self.supports_weights = True |
|
90
|
|
|
|
|
91
|
|
|
def fit(self, X, Y=None, W=None): |
|
92
|
|
|
clf = self.__wraps__(**self.params) |
|
93
|
|
|
if W is not None: |
|
94
|
|
|
return self.__returns__(clf.fit(X, W.reshape(-1))) |
|
95
|
|
|
return self.__returns__(clf.fit(X)) |
|
96
|
|
|
|
|
97
|
|
|
|
|
98
|
|
|
if __name__ == '__main__': |
|
99
|
|
|
import Orange |
|
100
|
|
|
|
|
101
|
|
|
data = Orange.data.Table('iris') |
|
102
|
|
|
learners = [SVMLearner(), NuSVMLearner(), LinearSVMLearner()] |
|
103
|
|
|
res = Orange.evaluation.CrossValidation(data, learners) |
|
104
|
|
|
for l, ca in zip(learners, Orange.evaluation.CA(res)): |
|
105
|
|
|
print("learner: {}\nCA: {}\n".format(l, ca)) |
|
106
|
|
|
|
hugobuddel /
orange3
Loading history...
This can be caused by one of the following:
1. Missing Dependencies
This error could indicate a configuration issue of Pylint. Make sure that your libraries are available by adding the necessary commands.
2. Missing __init__.py files
This error could also result from missing
__init__.pyfiles in your module folders. Make sure that you place one file in each sub-folder.