add matrix inputs to uniform; add matrix input unit tests

edward/stats/distributions.py

@@ -25,19 +25,19 @@ def logpmf(self, x):
         Arguments
         ---------
         x: np.array or tf.Tensor
-            If univariate distribution, can be a scalar or vector.
+            If univariate distribution, can be a scalar, vector, or matrix.
             If multivariate distribution, can be a vector or matrix.
 
         params: np.array or tf.Tensor
 
         Returns
         -------
         tf.Tensor
-            For univariate distributions, scalar if scalar input and
-            vector if vector input. For multivariate distributions,
-            scalar if vector input and vector if matrix input, where
-            the each element in the vector evaluates a row in the
-            matrix.
+            For univariate distributions, returns a scalar, vector, or
+            matrix corresponding to the size of input. For
+            multivariate distributions, returns a scalar if vector
+            input and vector if matrix input, where each element in
+            the vector evaluates a row in the matrix.
 
         Note
         ----
@@ -357,7 +357,7 @@ def rvs(self, loc=0, scale=1, size=1):
     def logpdf(self, x, loc=0, scale=1):
         # Note there is no error checking if x is outside domain.
         scale = tf.cast(tf.squeeze(scale), dtype=tf.float32)
-        return tf.squeeze(tf.ones([get_dims(x)[0]]) * -tf.log(scale))
+        return tf.squeeze(tf.ones(get_dims(x)) * -tf.log(scale))
 
 bernoulli = Bernoulli()
 beta = Beta()

tests/test_stats_bernoulli_logpmf.py

@@ -32,3 +32,11 @@ def test_logpmf_int_1d():
 def test_logpmf_float_1d():
     _test_logpmf([0.0, 1.0, 0.0], 0.5)
     _test_logpmf([1.0, 0.0, 0.0], 0.75)
+
+def test_logpmf_int_2d():
+    _test_logpmf(np.array([[0, 1, 0],[0, 1, 0]]), 0.5)
+    _test_logpmf(np.array([[1, 0, 0],[0, 1, 0]]), 0.75)
+
+def test_logpmf_float_2d():
+    _test_logpmf(np.array([[0.0, 1.0, 0.0],[0.0, 1.0, 0.0]]), 0.5)
+    _test_logpmf(np.array([[1.0, 0.0, 0.0],[0.0, 1.0, 0.0]]), 0.75)

tests/test_stats_beta_logpdf.py

@@ -36,3 +36,6 @@ def test_logpdf_scalar():
 
 def test_logpdf_1d():
     _test_logpdf([0.5, 0.3, 0.8, 0.1], a=0.5, b=0.5)
+
+def test_logpdf_2d():
+    _test_logpdf(np.array([[0.5, 0.3, 0.8, 0.1],[0.5, 0.3, 0.8, 0.1]]))

tests/test_stats_binom_logpmf.py

@@ -35,3 +35,11 @@ def test_logpmf_int_1d():
 def test_logpmf_float_1d():
     _test_logpmf([0.0, 1.0, 0.0], 1, 0.5)
     _test_logpmf([1.0, 0.0, 0.0], 1, 0.75)
+
+def test_logpmf_int_2d():
+    _test_logpmf(np.array([[0, 1, 0],[1, 0, 0]]), 1, 0.5)
+    _test_logpmf(np.array([[1, 0, 0],[0, 1, 0]]), 1, 0.75)
+
+def test_logpmf_float_2d():
+    _test_logpmf(np.array([[0.0, 1.0, 0.0],[1.0, 0.0, 0.0]]), 1, 0.5)
+    _test_logpmf(np.array([[1.0, 0.0, 0.0],[0.0, 1.0, 0.0]]), 1, 0.75)

tests/test_stats_chi2_logpdf.py

@@ -26,3 +26,6 @@ def test_logpdf_scalar():
 
 def test_logpdf_1d():
     _test_logpdf([0.1, 1.0, 0.58, 2.3], df=3)
+
+def test_logpdf_2d():
+    _test_logpdf(np.array([[0.1, 1.0, 0.58, 2.3],[0.3, 1.1, 0.68, 1.2]]), df=3)

tests/test_stats_expon_logpdf.py

@@ -32,3 +32,6 @@ def test_logpdf_scalar():
 
 def test_logpdf_1d():
     _test_logpdf([0.5, 2.3, 5.8, 10.1], scale=5.0)
+
+def test_logpdf_2d():
+    _test_logpdf(np.array([[0.5, 2.3, 5.8, 10.1],[0.5, 2.3, 5.8, 10.1]]), scale=5.0)

tests/test_stats_gamma_logpdf.py

@@ -36,3 +36,7 @@ def test_logpdf_scalar():
 
 def test_logpdf_1d():
     _test_logpdf([0.5, 1.2, 5.3, 8.7], a=0.5, b=0.5)
+
+def test_logpdf_2d():
+    _test_logpdf(np.array([[0.5, 1.2, 5.3, 8.7],[0.5, 1.2, 5.3, 8.7]]),
+                 a=0.5, b=0.5)

tests/test_stats_geom_logpmf.py

@@ -33,3 +33,11 @@ def test_logpmf_int_1d():
 def test_logpmf_float_1d():
     _test_logpmf([1.0, 5.0, 3.0], 0.5)
     _test_logpmf([2.0, 8.0, 2.0], 0.75)
+
+def test_logpmf_int_2d():
+    _test_logpmf(np.array([[1, 5, 3],[2, 8, 2]]), 0.5)
+    _test_logpmf(np.array([[2, 8, 2],[1, 5, 3]]), 0.75)
+
+def test_logpmf_float_2d():
+    _test_logpmf(np.array([[1.0, 5.0, 3.0],[2.0, 8.0, 2.0]]), 0.5)
+    _test_logpmf(np.array([[2.0, 8.0, 2.0],[1.0, 5.0, 3.0]]), 0.75)

tests/test_stats_invgamma_logpdf.py

@@ -36,3 +36,7 @@ def test_logpdf_scalar():
 
 def test_logpdf_1d():
     _test_logpdf([0.5, 1.2, 5.3, 8.7], a=0.5, scale=0.5)
+
+def test_logpdf_2d():
+    _test_logpdf(np.array([[0.5, 1.2, 5.3, 8.7],[0.5, 1.2, 5.3, 8.7]]),
+                 a=0.5, scale=0.5)

tests/test_stats_lognorm_logpdf.py

@@ -24,3 +24,6 @@ def test_logpdf_scalar():
 
 def test_logpdf_1d():
     _test_logpdf([2.0, 1.0, 0.58, 2.3])
+
+def test_logpdf_2d():
+    _test_logpdf(np.array([[2.0, 1.0, 0.58, 2.3],[2.1, 1.3, 1.58, 0.3]]))

tests/test_stats_nbinom_logpmf.py

@@ -37,3 +37,11 @@ def test_logpmf_int_1d():
 def test_logpmf_float_1d():
     _test_logpmf([1.0, 5.0, 3.0], 5, 0.5)
     _test_logpmf([2.0, 8.0, 2.0], 5, 0.75)
+
+def test_logpmf_int_2d():
+    _test_logpmf(np.array([[1, 5, 3],[2, 8, 2]]), 5, 0.5)
+    _test_logpmf(np.array([[2, 8, 2],[1, 5, 3]]), 5, 0.75)
+
+def test_logpmf_float_2d():
+    _test_logpmf(np.array([[1.0, 5.0, 3.0],[2.0, 8.0, 2.0]]), 5, 0.5)
+    _test_logpmf(np.array([[2.0, 8.0, 2.0],[1.0, 5.0, 3.0]]), 5, 0.75)

tests/test_stats_norm_logpdf.py

@@ -26,3 +26,6 @@ def test_logpdf_scalar():
 
 def test_logpdf_1d():
     _test_logpdf([0.0, 1.0, 0.58, 2.3])
+
+def test_logpdf_2d():
+    _test_logpdf(np.array([[0.0, 1.0, 0.58, 2.3], [0.1, 1.5, 4.18, 0.3]]))

tests/test_stats_poisson_logpmf.py

@@ -35,3 +35,11 @@ def test_logpmf_int_1d():
 def test_logpmf_float_1d():
     _test_logpmf([0.0, 1.0, 3.0], 0.5)
     _test_logpmf([1.0, 8.0, 2.0], 0.75)
+
+def test_logpmf_int_2d():
+    _test_logpmf(np.array([[0, 1, 3],[1, 8, 2]]), 0.5)
+    _test_logpmf(np.array([[1, 8, 2],[0, 1, 3]]), 0.75)
+
+def test_logpmf_float_2d():
+    _test_logpmf(np.array([[0.0, 1.0, 3.0],[0.0, 1.0, 3.0]]), 0.5)
+    _test_logpmf(np.array([[1.0, 8.0, 2.0],[1.0, 8.0, 2.0]]), 0.75)

tests/test_stats_t_logpdf.py

@@ -28,3 +28,6 @@ def test_logpdf_scalar():
 
 def test_logpdf_1d():
     _test_logpdf([0.0, 1.0, 0.58, 2.3], df=3)
+
+def test_logpdf_2d():
+    _test_logpdf(np.array([[0.0, 1.0, 0.58, 2.3],[0.0, 1.0, 0.58, 2.3]]), df=3)

tests/test_stats_truncnorm_logpdf.py

@@ -26,3 +26,7 @@ def test_logpdf_scalar():
 
 def test_logpdf_1d():
     _test_logpdf([0.0, 1.0, 0.58, 2.3], a=-1.0, b=3.0)
+
+def test_logpdf_2d():
+    _test_logpdf(np.array([[0.0, 1.0, 0.58, 2.3],[0.0, 1.0, 0.58, 2.3]]),
+                 a=-1.0, b=3.0)

tests/test_stats_uniform_logpdf.py

@@ -35,3 +35,7 @@ def test_logpdf_scalar():
 
 def test_logpdf_1d():
     _test_logpdf([0.5, 0.3, 0.8, 0.2], loc=0.1, scale=0.9)
+
+def test_logpdf_2d():
+    _test_logpdf(np.array([[0.5, 0.3, 0.8, 0.2],[0.5, 0.3, 0.8, 0.2]]),
+                 loc=0.1, scale=0.9)