From 0fa0fd39bbd69edcf919dd0f8770a57dca2148b9 Mon Sep 17 00:00:00 2001
From: lihongkang <[lihongkang1@huawei.com]>
Date: Fri, 27 Nov 2020 17:41:05 +0800
Subject: [PATCH] fix bugs

---
 mindspore/nn/layer/activation.py       |  3 +++
 mindspore/nn/loss/loss.py              |  3 +++
 mindspore/ops/operations/array_ops.py  | 10 ++++++----
 mindspore/ops/operations/math_ops.py   |  2 +-
 mindspore/ops/operations/nn_ops.py     |  3 +++
 mindspore/ops/operations/random_ops.py |  3 +++
 6 files changed, 19 insertions(+), 5 deletions(-)

diff --git a/mindspore/nn/layer/activation.py b/mindspore/nn/layer/activation.py
index f72a44d95b..3d9c4bcb24 100644
--- a/mindspore/nn/layer/activation.py
+++ b/mindspore/nn/layer/activation.py
@@ -374,6 +374,9 @@ class FastGelu(Cell):
     Outputs:
         Tensor, with the same type and shape as the `input_data`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> input_x = Tensor(np.array([[-1.0, 4.0, -8.0], [2.0, -5.0, 9.0]]), mindspore.float32)
         >>> fast_gelu = nn.FastGelu()
diff --git a/mindspore/nn/loss/loss.py b/mindspore/nn/loss/loss.py
index 25d61e8454..fdcbcf5164 100644
--- a/mindspore/nn/loss/loss.py
+++ b/mindspore/nn/loss/loss.py
@@ -309,6 +309,9 @@ class SampledSoftmaxLoss(_Loss):
     Outputs:
         Tensor, a tensor of shape (N) with the per-example sampled softmax losses.
 
+    Supported Platforms:
+        ``GPU``
+
     Examples:
         >>> loss = nn.SampledSoftmaxLoss(num_sampled=4, num_classes=7, num_true=1)
         >>> weights = Tensor(np.random.randint(0, 9, [7, 10]), mindspore.float32)
diff --git a/mindspore/ops/operations/array_ops.py b/mindspore/ops/operations/array_ops.py
index c0d260053b..1d63452595 100644
--- a/mindspore/ops/operations/array_ops.py
+++ b/mindspore/ops/operations/array_ops.py
@@ -554,6 +554,8 @@ class DynamicShape(Primitive):
         >>> input_tensor = Tensor(np.ones(shape=[3, 2, 1]), mindspore.float32)
         >>> shape = ops.DynamicShape()
         >>> output = shape(input_tensor)
+        >>> print(output)
+        [3 2 1]
     """
 
     @prim_attr_register
@@ -709,7 +711,7 @@ class Unique(Primitive):
         containing indices of elements in the input coressponding to the output tensor.
 
     Supported Platforms:
-        ``Ascend`` ``CPU``
+        ``Ascend`` ``GPU`` ``CPU``
 
     Examples:
         >>> x = Tensor(np.array([1, 2, 5, 2]), mindspore.int32)
@@ -779,7 +781,7 @@ class SparseGatherV2(GatherV2):
         Tensor, the shape of tensor is :math:`(z_1, z_2, ..., z_N)`.
 
     Supported Platforms:
-        ``GPU``
+        ``Ascend`` ``GPU``
 
     Examples:
         >>> input_params = Tensor(np.array([[1, 2, 7, 42], [3, 4, 54, 22], [2, 2, 55, 3]]), mindspore.float32)
@@ -2240,7 +2242,7 @@ class Pack(PrimitiveWithInfer):
                     or if the shapes of elements in input_x are not the same.
 
     Supported Platforms:
-        ``Ascend``
+        ``Ascend`` ``GPU``
 
     Examples:
         >>> data1 = Tensor(np.array([0, 1]).astype(np.float32))
@@ -2295,7 +2297,7 @@ class Unpack(PrimitiveWithInfer):
         ValueError: If axis is out of the range [-len(input_x.shape), len(input_x.shape)).
 
     Supported Platforms:
-        ``Ascend``
+        ``Ascend`` ``GPU``
 
     Examples:
         >>> unpack = ops.Unpack()
diff --git a/mindspore/ops/operations/math_ops.py b/mindspore/ops/operations/math_ops.py
index fe13e1ba1e..c6258639ee 100644
--- a/mindspore/ops/operations/math_ops.py
+++ b/mindspore/ops/operations/math_ops.py
@@ -1996,7 +1996,7 @@ class DivNoNan(_MathBinaryOp):
         and the data type is the one with higher precision or higher digits among the two inputs.
 
     Supported Platforms:
-        ``Ascend``
+        ``Ascend`` ``GPU``
 
     Examples:
         >>> input_x = Tensor(np.array([-1.0, 0., 1.0, 5.0, 6.0]), mindspore.float32)
diff --git a/mindspore/ops/operations/nn_ops.py b/mindspore/ops/operations/nn_ops.py
index d5cfcb8904..587499e9fb 100644
--- a/mindspore/ops/operations/nn_ops.py
+++ b/mindspore/ops/operations/nn_ops.py
@@ -2971,6 +2971,9 @@ class FastGelu(PrimitiveWithInfer):
     Outputs:
         Tensor, with the same type and shape as input.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> tensor = Tensor(np.array([[-1.0, 4.0, -8.0], [2.0, -5.0, 9.0]]), mindspore.float32)
         >>> fast_gelu = P.FastGelu()
diff --git a/mindspore/ops/operations/random_ops.py b/mindspore/ops/operations/random_ops.py
index 9ab3ef39a8..32a5fa61ba 100644
--- a/mindspore/ops/operations/random_ops.py
+++ b/mindspore/ops/operations/random_ops.py
@@ -560,6 +560,9 @@ class UniformCandidateSampler(PrimitiveWithInfer):
         - **sampled_expected_count** (Tensor) - The expected counts under the sampling distribution of
             each of sampled_candidates. Shape: (num_sampled, ).
 
+    Supported Platforms:
+        ``GPU``
+
     Examples:
         >>> sampler = ops.UniformCandidateSampler(1, 3, False, 4)
         >>> output1, output2, output3 = sampler(Tensor(np.array([[1],[3],[4],[6],[3]], dtype=np.int32)))