def test_platt_correction(): F=numpy.ones(10) L=numpy.ones(10) F[:5] *= -1 L[:5] *= 0 A,B=learn_sigmoid_constants(F,L) assert 1./(1.+numpy.exp(+10*A+B)) > .99 assert 1./(1.+numpy.exp(-10*A+B)) <.01
def test_platt_correction(): F = numpy.ones(10) L = numpy.ones(10) F[:5] *= -1 L[:5] *= 0 A, B = learn_sigmoid_constants(F, L) assert 1. / (1. + numpy.exp(+10 * A + B)) > .99 assert 1. / (1. + numpy.exp(-10 * A + B)) < .01
def test_learn_sigmoid_contants(): Y = np.repeat((0,1),100) np.random.seed(3) for i in xrange(10): F = np.random.rand(200)-.3 F[100:] *= -1 old = old_learn_sigmoid_constants(F,Y) new = svm.learn_sigmoid_constants(F,Y) assert np.allclose(old, new)
def test_learn_sigmoid_contants(): Y = np.repeat((0, 1), 100) np.random.seed(3) for i in xrange(10): F = np.random.rand(200) - .3 F[100:] *= -1 old = old_learn_sigmoid_constants(F, Y) new = svm.learn_sigmoid_constants(F, Y) assert np.allclose(old, new)
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