!3345 Fix minor bugs in denoting and test cases

Merge pull request !3345 from peixu_ren/custom_gpu

mindspore/ops/composite/__init__.py

@@ -27,7 +27,7 @@ from .clip_ops import clip_by_value
 from .multitype_ops.add_impl import hyper_add
 from .multitype_ops.ones_like_impl import ones_like
 from .multitype_ops.zeros_like_impl import zeros_like
-from .random_ops import normal
+from .random_ops import set_seed, normal
 
 
 __all__ = [
@@ -48,5 +48,6 @@ __all__ = [
     'zeros_like',
     'ones_like',
     'zip_operation',
+    'set_seed',
     'normal',
     'clip_by_value',]

mindspore/ops/composite/random_ops.py

@@ -15,8 +15,11 @@
 
 """Operations for random number generatos."""
 
-from mindspore.ops.primitive import constexpr
 from .. import operations as P
+from .. import functional as F
+from ..primitive import constexpr
+from .multitype_ops import _constexpr_utils as const_utils
+from ...common import dtype as mstype
 
 # set graph-level RNG seed
 _GRAPH_SEED = 0
@@ -31,17 +34,17 @@ def get_seed():
     return _GRAPH_SEED
 
 
-def normal(shape, mean, stddev, seed):
+def normal(shape, mean, stddev, seed=0):
     """
     Generates random numbers according to the Normal (or Gaussian) random number distribution.
     It is defined as:
 
     Args:
-        - **shape** (tuple) - The shape of random tensor to be generated.
+        shape (tuple): The shape of random tensor to be generated.
-        - **mean** (Tensor) - The mean μ distribution parameter, which specifies the location of the peak.
+        mean (Tensor): The mean μ distribution parameter, which specifies the location of the peak.
           With float32 data type.
-        - **stddev** (Tensor) - The deviation σ distribution parameter. With float32 data type.
+        stddev (Tensor): The deviation σ distribution parameter. With float32 data type.
-        - **seed** (int): Seed is used as entropy source for Random number engines generating pseudo-random numbers.
+        seed (int): Seed is used as entropy source for Random number engines generating pseudo-random numbers.
           Default: 0.
 
     Returns:
@@ -52,9 +55,13 @@ def normal(shape, mean, stddev, seed):
         >>> shape = (4, 16)
         >>> mean = Tensor(1.0, mstype.float32)
         >>> stddev = Tensor(1.0, mstype.float32)
+        >>> C.set_seed(10)
         >>> output = C.normal(shape, mean, stddev, seed=5)
     """
-    set_seed(10)
+    mean_dtype = F.dtype(mean)
+    stddev_dtype = F.dtype(stddev)
+    const_utils.check_tensors_dtype_same(mean_dtype, mstype.float32, "normal")
+    const_utils.check_tensors_dtype_same(stddev_dtype, mstype.float32, "normal")
     seed1 = get_seed()
     seed2 = seed
     stdnormal = P.StandardNormal(seed1, seed2)

tests/st/ops/ascend/test_aicpu_ops/test_standard_normal.py

@@ -29,7 +29,7 @@ class Net(nn.Cell):
         self.stdnormal = P.StandardNormal(seed, seed2)
 
     def construct(self):
-        return self.stdnormal(self.shape, self.seed, self.seed2)
+        return self.stdnormal(self.shape)
 
 
 def test_net():

tests/st/ops/gpu/test_standard_normal.py

@@ -29,7 +29,7 @@ class Net(nn.Cell):
         self.stdnormal = P.StandardNormal(seed, seed2)
 
     def construct(self):
-        return self.stdnormal(self.shape, self.seed, self.seed2)
+        return self.stdnormal(self.shape)
 
 
 def test_net():