Allow calculation of full predictive covariance matrices with multiple outputs and normalization

GPy/testing/model_tests.py

@@ -118,6 +118,24 @@ class MiscTests(unittest.TestCase):
         from scipy.stats import norm
         np.testing.assert_allclose((mu+(norm.ppf(qs/100.)*np.sqrt(var))).flatten(), np.array(q95).flatten())
 
+    def test_multioutput_regression_with_normalizer(self):
+        """
+        Test that normalizing works in multi-output case
+        """
+
+        # Create test inputs
+        X1 = (np.random.rand(50, 1) * 8)
+        Y1 = np.sin(X1) + np.random.randn(*X1.shape) * 0.05
+        Y2 = -np.sin(X1) + np.random.randn(*X1.shape) * 0.05
+        Y = np.hstack((Y1, Y2))
+
+        m = GPy.models.GPRegression(X1, Y, normalizer=True)
+
+        n_test_point = 10
+        mean, covaraince = m.predict(np.random.rand(n_test_point, 1),
+                                     full_cov=True)
+        self.assertTrue(covaraince.shape == (n_test_point, n_test_point, 2))
+
     def check_jacobian(self):
         try:
             import autograd.numpy as np, autograd as ag, GPy, matplotlib.pyplot as plt

GPy/testing/util_tests.py

@@ -28,7 +28,8 @@
 # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 #===============================================================================
 
-import unittest, numpy as np
+import unittest
+import numpy as np
 import GPy
 
 class TestDebug(unittest.TestCase):
@@ -225,3 +226,17 @@ class TestUnivariateGaussian(unittest.TestCase):
         for i in range(len(pySols)):
           diff += abs(derivLogCdfNormal(self.zz[i]) - pySols[i])
         self.assertTrue(diff  < 1e-8)
+
+class TestStandardize(unittest.TestCase):
+    def setUp(self):
+        self.normalizer = GPy.util.normalizer.Standardize()
+        y = np.stack([np.random.randn(10), 2*np.random.randn(10)], axis=1)
+        self.normalizer.scale_by(y)
+    
+    def test_inverse_covariance(self):
+        """
+        Test inverse covariance outputs correct size
+        """
+        covariance = np.random.rand(100, 100)
+        output = self.normalizer.inverse_covariance(covariance)
+        self.assertTrue(output.shape == (100, 100, 2))