apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-06-05 14:55:15 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

[coverage] test predict sparse gp

Browse source
This commit is contained in:
Max Zwiessele 2015-09-04 13:28:59 +01:00
parent eddebd2ad9
commit e766eb6de9
1 changed files with 33 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

33
GPy/testing/bgplvm_minibatch_tests.py
View file

@ -28,6 +28,29 @@ class BGPLVMTest(unittest.TestCase):
np.testing.assert_allclose(m.gradient, self.m_full.gradient)
assert(m.checkgrad())
def test_predict_missing_data(self):
m = GPy.models.bayesian_gplvm_minibatch.BayesianGPLVMMiniBatch(self.Y, self.Q, missing_data=True, stochastic=True, batchsize=self.Y.shape[1])
m[:] = self.m_full[:]
np.testing.assert_almost_equal(m.log_likelihood(), self.m_full.log_likelihood(), 7)
np.testing.assert_allclose(m.gradient, self.m_full.gradient)
self.assertRaises(NotImplementedError, m.predict, m.X, full_cov=True)
mu1, var1 = m.predict(m.X, full_cov=False)
mu2, var2 = self.m_full.predict(self.m_full.X, full_cov=False)
np.testing.assert_allclose(mu1, mu2)
np.testing.assert_allclose(var1, var2)
mu1, var1 = m.predict(m.X.mean, full_cov=True)
mu2, var2 = self.m_full.predict(self.m_full.X.mean, full_cov=True)
np.testing.assert_allclose(mu1, mu2)
np.testing.assert_allclose(var1[:,:,0], var2)
mu1, var1 = m.predict(m.X.mean, full_cov=False)
mu2, var2 = self.m_full.predict(self.m_full.X.mean, full_cov=False)
np.testing.assert_allclose(mu1, mu2)
np.testing.assert_allclose(var1[:,[0]], var2)
def test_lik_comparisons_m0_s0(self):
# Test if the different implementations give the exact same likelihood as the full model.
# All of the following settings should give the same likelihood and gradients as the full model:
@ -71,6 +94,16 @@ class BGPLVMTest(unittest.TestCase):
m = GPy.models.bayesian_gplvm_minibatch.BayesianGPLVMMiniBatch(self.Y, self.Q, missing_data=False, stochastic=True, batchsize=4)
assert(m.checkgrad())
def test_predict(self):
# Test if the different implementations give the exact same likelihood as the full model.
# All of the following settings should give the same likelihood and gradients as the full model:
m = GPy.models.bayesian_gplvm_minibatch.BayesianGPLVMMiniBatch(self.Y, self.Q, missing_data=True, stochastic=True, batchsize=self.Y.shape[1])
m[:] = self.m_full[:]
np.testing.assert_almost_equal(m.log_likelihood(), self.m_full.log_likelihood(), 7)
np.testing.assert_allclose(m.gradient, self.m_full.gradient)
assert(m.checkgrad())
if __name__ == "__main__":
#import sys;sys.argv = ['', 'Test.testName']
unittest.main()