refine code

Author: MayYouBeProsperous

python/paddle/distribution/exponential.py

@@ -84,7 +84,8 @@ def sample(self, shape=()):
             Tensor, A tensor with prepended dimensions shape.The data type is float32.
 
         """
-        return self.rsample(shape)
+        with paddle.no_grad():
+            return self.rsample(shape)
 
     def rsample(self, shape=()):
         """Generate reparameterized samples of the specified shape.

python/paddle/distribution/gamma.py

@@ -139,25 +139,11 @@ def entropy(self):
         )
 
     def sample(self, shape=()):
-        """Generate samples of the specified shape.
-
-        Args:
-            shape (Sequence[int], optional): Shape of the generated samples.
-
-        Returns:
-            Tensor, A tensor with prepended dimensions shape.The data type is float32.
-        """
+        """Generate samples of the specified shape."""
         raise NotImplementedError
 
     def rsample(self, shape=()):
-        """Generate reparameterized samples of the specified shape.
-
-        Args:
-            shape (Sequence[int], optional): Shape of the generated samples.
-
-        Returns:
-            Tensor: A tensor with prepended dimensions shape. The data type is float32.
-        """
+        """Generate reparameterized samples of the specified shape."""
         raise NotImplementedError
 
     def kl_divergence(self, other):

test/distribution/test_distribution_gamma.py

@@ -170,11 +170,6 @@ def test_entropy(self):
 @parameterize.parameterize_cls(
     (parameterize.TEST_CASE_NAME, 'concentration', 'rate'),
     [
-        (
-            '0-dim',
-            0.5,
-            0.5,
-        ),
         (
             'one-dim',
             parameterize.xrand(
@@ -201,19 +196,6 @@ def test_entropy(self):
                 min=np.finfo(dtype='float32').tiny,
             ),
         ),
-        (
-            'broadcast',
-            parameterize.xrand(
-                (2, 1),
-                dtype='float32',
-                min=np.finfo(dtype='float32').tiny,
-            ),
-            parameterize.xrand(
-                (2, 3),
-                dtype='float32',
-                min=np.finfo(dtype='float32').tiny,
-            ),
-        ),
     ],
 )
 class TestGammaSample(unittest.TestCase):