Fix testval of ObservedRV

Author: aseyboldt

pymc3/model.py

@@ -915,7 +915,7 @@ def as_tensor(data, name, model, distribution):
 
     if hasattr(data, 'mask'):
         from .distributions import NoDistribution
-        testval = distribution.testval or data.mean().astype(dtype)
+        testval = distribution.default()
         fakedist = NoDistribution.dist(shape=data.mask.sum(), dtype=dtype,
                                        testval=testval, parent_dist=distribution)
         missing_values = FreeRV(name=name + '_missing', distribution=fakedist,
@@ -959,13 +959,14 @@ def __init__(self, type=None, owner=None, index=None, name=None, data=None,
             data = pandas_to_array(data)
             type = TensorType(distribution.dtype, data.shape)
 
+        self.observations = data
+
         super(ObservedRV, self).__init__(type, owner, index, name)
 
         if distribution is not None:
             data = as_tensor(data, name, model, distribution)
 
             self.missing_values = data.missing_values
-
             self.logp_elemwiset = distribution.logp(data)
             self.total_size = total_size
             self.model = model

pymc3/tests/test_model.py

@@ -1,5 +1,8 @@
 import pytest
 from theano import theano, tensor as tt
+import numpy as np
+import pandas as pd
+import numpy.testing as npt
 
 import pymc3 as pm
 from pymc3.distributions import HalfCauchy, Normal, transforms
@@ -165,4 +168,14 @@ def test_duplicate_vars():
         with pm.Model():
             pm.Binomial('a', 10, .5)
             pm.Normal('a', transform=transforms.log)
-    err.match('already exists')
+    err.match('already exists')
+
+
+def test_empty_observed():
+    data = pd.DataFrame(np.ones((2, 3)) / 3)
+    data.values[:] = np.nan
+    with pm.Model():
+        a = pm.Normal('a', observed=data)
+        npt.assert_allclose(a.tag.test_value, np.zeros((2, 3)))
+        b = pm.Beta('b', alpha=1, beta=1, observed=data)
+        npt.assert_allclose(b.tag.test_value, np.ones((2, 3)) / 2)