Reenable old MatrixNormal test

Author: ricardoV94

pymc/tests/test_distributions_random.py

@@ -1756,6 +1756,7 @@ class TestMatrixNormal(BaseTestDistributionRandom):
         "check_pymc_params_match_rv_op",
         "check_draws",
         "check_errors",
+        "check_random_variable_prior",
     ]
 
     def check_draws(self):
@@ -1824,6 +1825,28 @@ def check_errors(self):
                     shape=15,
                 )
 
+    def check_random_variable_prior(self):
+        """
+        This test checks for shape correctness when using MatrixNormal distribution
+        with parameters as random variables.
+        Originally reported - https://github.com/pymc-devs/pymc/issues/3585
+        """
+        K = 3
+        D = 15
+        mu_0 = np.zeros((D, K))
+        lambd = 1.0
+        with pm.Model() as model:
+            sd_dist = pm.HalfCauchy.dist(beta=2.5, size=D)
+            packedL = pm.LKJCholeskyCov("packedL", eta=2, n=D, sd_dist=sd_dist, compute_corr=False)
+            L = pm.expand_packed_triangular(D, packedL, lower=True)
+            Sigma = pm.Deterministic("Sigma", L.dot(L.T))  # D x D covariance
+            mu = pm.MatrixNormal(
+                "mu", mu=mu_0, rowcov=(1 / lambd) * Sigma, colcov=np.eye(K), shape=(D, K)
+            )
+            prior = pm.sample_prior_predictive(2, return_inferencedata=False)
+
+        assert prior["mu"].shape == (2, D, K)
+
 
 class TestInterpolated(BaseTestDistributionRandom):
     def interpolated_rng_fn(self, size, mu, sigma, rng):
@@ -2435,30 +2458,6 @@ def generate_shapes(include_params=False):
     return data
 
 
-@pytest.mark.xfail(reason="This distribution has not been refactored for v4")
-def test_matrix_normal_random_with_random_variables():
-    """
-    This test checks for shape correctness when using MatrixNormal distribution
-    with parameters as random variables.
-    Originally reported - https://github.com/pymc-devs/pymc/issues/3585
-    """
-    K = 3
-    D = 15
-    mu_0 = np.zeros((D, K))
-    lambd = 1.0
-    with pm.Model() as model:
-        sd_dist = pm.HalfCauchy.dist(beta=2.5)
-        packedL = pm.LKJCholeskyCov("packedL", eta=2, n=D, sd_dist=sd_dist)
-        L = pm.expand_packed_triangular(D, packedL, lower=True)
-        Sigma = pm.Deterministic("Sigma", L.dot(L.T))  # D x D covariance
-        mu = pm.MatrixNormal(
-            "mu", mu=mu_0, rowcov=(1 / lambd) * Sigma, colcov=np.eye(K), shape=(D, K)
-        )
-        prior = pm.sample_prior_predictive(2)
-
-    assert prior["mu"].shape == (2, D, K)
-
-
 @pytest.mark.xfail(reason="This distribution has not been refactored for v4")
 class TestMvGaussianRandomWalk(SeededTest):
     @pytest.mark.parametrize(

pymc/tests/test_idata_conversion.py

@@ -331,14 +331,14 @@ def test_missing_data_model(self):
         # See https://github.com/pymc-devs/pymc/issues/5255
         assert inference_data.log_likelihood["y_observed"].shape == (2, 100, 3)
 
-    @pytest.mark.xfal(reason="Multivariate partial observed RVs not implemented for V4")
+    @pytest.mark.xfail(reason="Multivariate partial observed RVs not implemented for V4")
     def test_mv_missing_data_model(self):
         data = ma.masked_values([[1, 2], [2, 2], [-1, 4], [2, -1], [-1, -1]], value=-1)
 
         model = pm.Model()
         with model:
             mu = pm.Normal("mu", 0, 1, size=2)
-            sd_dist = pm.HalfNormal.dist(1.0)
+            sd_dist = pm.HalfNormal.dist(1.0, size=2)
             # pylint: disable=unpacking-non-sequence
             chol, *_ = pm.LKJCholeskyCov("chol_cov", n=2, eta=1, sd_dist=sd_dist, compute_corr=True)
             # pylint: enable=unpacking-non-sequence