def test_ecoc_probability():
features,labels = load()
features = features[labels < 5]
labels = labels[labels < 5]
raw = svm.svm_raw(kernel=svm.dot_kernel(), C=1.)
base = ctransforms(raw, svm.svm_sigmoidal_correction())
learner = ecoc_learner(base, probability=True)
model = learner.train(features[::2], labels[::2])
results = list(map(model.apply, features[1::2]))
results = np.array(results)
assert results.shape[1] == len(set(labels))
assert np.mean(results.argmax(1) == labels[1::2]) > .5
def test_platt_correction_class():
F=numpy.ones(10)
L=numpy.ones(10)
F[:5] *= -1
L[:5] *= 0
corrector = svm_sigmoidal_correction()
A,B = learn_sigmoid_constants(F,L)
model = corrector.train(F,L)
assert model.A == A
assert model.B == B
assert model.apply(10) > .99
assert model.apply(-10) < .01
def test_platt_correction_class():
F = numpy.ones(10)
L = numpy.ones(10)
F[:5] *= -1
L[:5] *= 0
corrector = svm_sigmoidal_correction()
A, B = learn_sigmoid_constants(F, L)
model = corrector.train(F, L)
assert model.A == A
assert model.B == B
assert model.apply(10) > .99
assert model.apply(-10) < .01