diff --git a/mindspore/nn/cell.py b/mindspore/nn/cell.py
index 6fa9b5d331..faa2643ab7 100755
--- a/mindspore/nn/cell.py
+++ b/mindspore/nn/cell.py
@@ -51,6 +51,9 @@ class Cell(Cell_):
     Args:
         auto_prefix (bool): Recursively generate namespaces. Default: True.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> class MyCell(Cell):
         >>>    def __init__(self):
diff --git a/mindspore/nn/layer/activation.py b/mindspore/nn/layer/activation.py
index e9025ea256..53693e9fb9 100644
--- a/mindspore/nn/layer/activation.py
+++ b/mindspore/nn/layer/activation.py
@@ -67,6 +67,9 @@ class Softmax(Cell):
     Outputs:
         Tensor, which has the same type and shape as `x` with values in the range[0,1].
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> input_x = Tensor(np.array([-1, -2, 0, 2, 1]), mindspore.float16)
         >>> softmax = nn.Softmax()
@@ -104,6 +107,9 @@ class LogSoftmax(Cell):
     Outputs:
         Tensor, which has the same type and shape as the input as `x` with values in the range[-inf,0).
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> input_x = Tensor(np.array([[-1.0, 4.0, -8.0], [2.0, -5.0, 9.0]]), mindspore.float32)
         >>> log_softmax = nn.LogSoftmax()
@@ -174,6 +180,9 @@ class ReLU(Cell):
     Outputs:
         Tensor, with the same type and shape as the `input_data`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> input_x = Tensor(np.array([-1, 2, -3, 2, -1]), mindspore.float16)
         >>> relu = nn.ReLU()
@@ -204,6 +213,9 @@ class ReLU6(Cell):
     Outputs:
         Tensor, which has the same type as `input_data`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> input_x = Tensor(np.array([-1, -2, 0, 2, 1]), mindspore.float16)
         >>> relu6 = nn.ReLU6()
@@ -242,6 +254,9 @@ class LeakyReLU(Cell):
     Outputs:
         Tensor, has the same type and shape as the `input_x`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> input_x = Tensor(np.array([[-1.0, 4.0, -8.0], [2.0, -5.0, 9.0]]), mindspore.float32)
         >>> leaky_relu = nn.LeakyReLU()
@@ -287,6 +302,9 @@ class Tanh(Cell):
     Outputs:
         Tensor, with the same type and shape as the `input_data`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> input_x = Tensor(np.array([1, 2, 3, 2, 1]), mindspore.float16)
         >>> tanh = nn.Tanh()
@@ -319,6 +337,9 @@ class GELU(Cell):
     Outputs:
         Tensor, with the same type and shape as the `input_data`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> input_x = Tensor(np.array([[-1.0, 4.0, -8.0], [2.0, -5.0, 9.0]]), mindspore.float32)
         >>> gelu = nn.GELU()
@@ -351,6 +372,9 @@ class Sigmoid(Cell):
     Outputs:
         Tensor, with the same type and shape as the `input_data`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> input_x = Tensor(np.array([-1, -2, 0, 2, 1]), mindspore.float16)
         >>> sigmoid = nn.Sigmoid()
@@ -390,6 +414,9 @@ class PReLU(Cell):
     Outputs:
         Tensor, with the same type and shape as the `input_data`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> input_x = Tensor(np.array([[[[0.1, 0.6], [0.9, 0.9]]]]), mindspore.float32)
         >>> prelu = nn.PReLU()
@@ -444,6 +471,9 @@ class HSwish(Cell):
     Outputs:
         Tensor, with the same type and shape as the `input_data`.
 
+    Supported Platforms:
+        ``GPU``
+
     Examples:
         >>> input_x = Tensor(np.array([-1, -2, 0, 2, 1]), mindspore.float16)
         >>> hswish = nn.HSwish()
@@ -479,6 +509,9 @@ class HSigmoid(Cell):
     Outputs:
         Tensor, with the same type and shape as the `input_data`.
 
+    Supported Platforms:
+        ``GPU``
+
     Examples:
         >>> input_x = Tensor(np.array([-1, -2, 0, 2, 1]), mindspore.float16)
         >>> hsigmoid = nn.HSigmoid()
@@ -514,6 +547,9 @@ class LogSigmoid(Cell):
     Outputs:
         Tensor, with the same type and shape as the `input_data`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = nn.LogSigmoid()
         >>> input_x = Tensor(np.array([1.0, 2.0, 3.0]), mindspore.float32)
diff --git a/mindspore/nn/layer/basic.py b/mindspore/nn/layer/basic.py
index 18033a48c2..696cf58153 100644
--- a/mindspore/nn/layer/basic.py
+++ b/mindspore/nn/layer/basic.py
@@ -72,6 +72,9 @@ class Dropout(Cell):
     Outputs:
         Tensor, output tensor with the same shape as the input.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> x = Tensor(np.ones([2, 2, 3]), mindspore.float32)
         >>> net = nn.Dropout(keep_prob=0.8)
@@ -139,6 +142,9 @@ class Flatten(Cell):
         Tensor, the shape of the output tensor is :math:`(N, X)`, where :math:`X` is
         the product of the remaining dimensions.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> input = Tensor(np.array([[[1.2, 1.2], [2.1, 2.1]], [[2.2, 2.2], [3.2, 3.2]]]), mindspore.float32)
         >>> net = nn.Flatten()
@@ -195,6 +201,9 @@ class Dense(Cell):
     Outputs:
         Tensor of shape :math:`(*, out\_channels)`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> input = Tensor(np.random.randint(0, 255, [2, 3]), mindspore.float32)
         >>> net = nn.Dense(3, 4)
@@ -316,6 +325,9 @@ class ClipByNorm(Cell):
     Outputs:
         Tensor, clipped tensor with the same shape as the input, whose type is float32.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = nn.ClipByNorm()
         >>> input = Tensor(np.random.randint(0, 10, [4, 16]), mindspore.float32)
@@ -387,6 +399,9 @@ class Norm(Cell):
         Tensor, output tensor with dimensions in 'axis' reduced to 1 will be returned if 'keep_dims' is True;
         otherwise a Tensor with dimensions in 'axis' removed is returned.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = nn.Norm(axis=0)
         >>> input = Tensor(np.random.randint(0, 10, [2, 4]), mindspore.float32)
@@ -444,6 +459,9 @@ class OneHot(Cell):
         Tensor, the one-hot tensor of data type 'dtype' with dimension at 'axis' expanded to 'depth' and filled with
         on_value and off_value.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> net = nn.OneHot(depth=4, axis=1)
         >>> indices = Tensor([[1, 3], [0, 2]], dtype=mindspore.int32)
@@ -501,6 +519,9 @@ class Pad(Cell):
           is [[1,2,3],[4,5,6],[7,8,9]] and `paddings` is [[1,1],[2,2]], then the Outputs is
           [[2,1,1,2,3,3,2],[2,1,1,2,3,3,2],[5,4,4,5,6,6,5],[8,7,7,8,9,9,8],[8,7,7,8,9,9,8]].
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> from mindspore import Tensor
         >>> from mindspore.ops import operations as P
@@ -587,6 +608,9 @@ class Interpolate(Cell):
         If scale is set, the result is 4-D tensor with shape:math:'(batch, channels, scale_factor * height,
         scale_factor * width)' in float32
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> from mindspore.ops import operations as P
         >>> tensor = Tensor([[[[1, 2, 3, 4], [5, 6, 7, 8]]]], mindspore.float32)
@@ -630,6 +654,9 @@ class Unfold(Cell):
         Tensor, a 4-D tensor whose data type is same as 'input_x',
         and the shape is [out_batch, out_depth, out_row, out_col], the out_batch is the same as the in_batch.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> net = Unfold(ksizes=[1, 2, 2, 1], strides=[1, 2, 2, 1], rates=[1, 2, 2, 1])
         >>> image = Tensor(np.ones([2, 3, 6, 6]), dtype=mstype.float16)
@@ -675,6 +702,9 @@ class Tril(Cell):
     Outputs:
         Tensor, has the same type as input `x`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> x = Tensor(np.array([[1, 2], [3, 4]]))
         >>> tril = nn.Tril()
@@ -757,6 +787,9 @@ class MatrixDiag(Cell):
     Outputs:
         Tensor, has the same type as input `x`. The shape must be x.shape + (x.shape[-1], ).
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> x = Tensor(np.array([1, -1]), mstype.float32)
         >>> matrix_diag = nn.MatrixDiag()
@@ -789,6 +822,9 @@ class MatrixDiagPart(Cell):
     Outputs:
         Tensor, has the same type as input `x`. The shape must be x.shape[:-2] + [min(x.shape[-2:])].
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> x = Tensor([[[-1, 0], [0, 1]], [[-1, 0], [0, 1]], [[-1, 0], [0, 1]]], mindspore.float32)
         >>> matrix_diag_part = nn.MatrixDiagPart()
@@ -823,6 +859,9 @@ class MatrixSetDiag(Cell):
     Outputs:
         Tensor, has the same type and shape as input `x`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> x = Tensor([[[-1, 0], [0, 1]], [[-1, 0], [0, 1]], [[-1, 0], [0, 1]]], mindspore.float32)
         >>> diagonal = Tensor([[-1., 2.], [-1., 1.], [-1., 1.]], mindspore.float32)
diff --git a/mindspore/nn/layer/combined.py b/mindspore/nn/layer/combined.py
index 981b11add4..8475c01122 100644
--- a/mindspore/nn/layer/combined.py
+++ b/mindspore/nn/layer/combined.py
@@ -76,6 +76,9 @@ class Conv2dBnAct(Cell):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, H_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = nn.Conv2dBnAct(120, 240, 4, has_bn=True, activation='relu')
         >>> input = Tensor(np.ones([1, 120, 1024, 640]), mindspore.float32)
@@ -168,6 +171,9 @@ class DenseBnAct(Cell):
     Outputs:
         Tensor of shape :math:`(N, out\_channels)`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> net = nn.DenseBnAct(3, 4)
         >>> input = Tensor(np.random.randint(0, 255, [2, 3]), mindspore.float32)
diff --git a/mindspore/nn/layer/container.py b/mindspore/nn/layer/container.py
index 1e602143f3..711f6522f8 100644
--- a/mindspore/nn/layer/container.py
+++ b/mindspore/nn/layer/container.py
@@ -80,6 +80,9 @@ class SequentialCell(Cell):
     Outputs:
         Tensor, the output Tensor with shape depending on the input and defined sequence of Cells.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> conv = nn.Conv2d(3, 2, 3, pad_mode='valid')
         >>> bn = nn.BatchNorm2d(2)
@@ -184,6 +187,9 @@ class CellList(_CellListBase, Cell):
     Args:
         args (list, optional): List of subclass of Cell.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> conv = nn.Conv2d(100, 20, 3)
         >>> bn = nn.BatchNorm2d(20)
diff --git a/mindspore/nn/layer/conv.py b/mindspore/nn/layer/conv.py
index 3bd3cf6c54..fa93080479 100644
--- a/mindspore/nn/layer/conv.py
+++ b/mindspore/nn/layer/conv.py
@@ -196,6 +196,9 @@ class Conv2d(_Conv):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, H_{out}, W_{out})` or `(N, H_{out}, W_{out}, C_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> net = nn.Conv2d(120, 240, 4, has_bias=False, weight_init='normal')
         >>> input = Tensor(np.ones([1, 120, 1024, 640]), mindspore.float32)
@@ -372,6 +375,9 @@ class Conv1d(_Conv):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = nn.Conv1d(120, 240, 4, has_bias=False, weight_init='normal')
         >>> input = Tensor(np.ones([1, 120, 640]), mindspore.float32)
@@ -543,6 +549,9 @@ class Conv2dTranspose(_Conv):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, H_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = nn.Conv2dTranspose(3, 64, 4, has_bias=False, weight_init='normal', pad_mode='pad')
         >>> input = Tensor(np.ones([1, 3, 16, 50]), mindspore.float32)
@@ -719,6 +728,9 @@ class Conv1dTranspose(_Conv):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = nn.Conv1dTranspose(3, 64, 4, has_bias=False, weight_init='normal', pad_mode='pad')
         >>> input = Tensor(np.ones([1, 3, 50]), mindspore.float32)
diff --git a/mindspore/nn/layer/embedding.py b/mindspore/nn/layer/embedding.py
index 6b0556eb17..ad947d7f31 100755
--- a/mindspore/nn/layer/embedding.py
+++ b/mindspore/nn/layer/embedding.py
@@ -60,6 +60,9 @@ class Embedding(Cell):
     Outputs:
         Tensor of shape :math:`(\text{batch_size}, \text{input_length}, \text{embedding_size})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = nn.Embedding(20000, 768,  True)
         >>> input_data = Tensor(np.ones([8, 128]), mindspore.int32)
@@ -160,6 +163,9 @@ class EmbeddingLookup(Cell):
     Outputs:
         Tensor, the shape of tensor is :math:`(z_1, z_2, ..., z_N)`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> input_indices = Tensor(np.array([[1, 0], [3, 2]]), mindspore.int32)
         >>> result = nn.EmbeddingLookup(4,2)(input_indices)
diff --git a/mindspore/nn/layer/image.py b/mindspore/nn/layer/image.py
index aea86b3658..3a488d69c6 100644
--- a/mindspore/nn/layer/image.py
+++ b/mindspore/nn/layer/image.py
@@ -50,6 +50,9 @@ class ImageGradients(Cell):
         - **dy** (Tensor) - vertical image gradients, the same type and shape as input.
         - **dx** (Tensor) - horizontal image gradients, the same type and shape as input.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = nn.ImageGradients()
         >>> image = Tensor(np.array([[[[1,2],[3,4]]]]), dtype=mindspore.int32)
@@ -211,6 +214,9 @@ class SSIM(Cell):
     Outputs:
         Tensor, has the same dtype as img1. It is a 1-D tensor with shape N, where N is the batch num of img1.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = nn.SSIM()
         >>> img1 = Tensor(np.random.random((1,3,16,16)), mindspore.float32)
@@ -289,6 +295,9 @@ class MSSSIM(Cell):
     Outputs:
         Tensor, the value is in range [0, 1]. It is a 1-D tensor with shape N, where N is the batch num of img1.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> net = nn.MSSSIM(power_factors=(0.033, 0.033, 0.033))
         >>> np.random.seed(0)
@@ -380,6 +389,9 @@ class PSNR(Cell):
     Outputs:
         Tensor, with dtype mindspore.float32. It is a 1-D tensor with shape N, where N is the batch num of img1.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = nn.PSNR()
         >>> img1 = Tensor(np.random.random((1,3,16,16)))
@@ -450,6 +462,9 @@ class CentralCrop(Cell):
     Outputs:
         Tensor, 3-D or 4-D float tensor, according to the input.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> net = nn.CentralCrop(central_fraction=0.5)
         >>> image = Tensor(np.random.random((4, 3, 4, 4)), mindspore.float32)
diff --git a/mindspore/nn/layer/lstm.py b/mindspore/nn/layer/lstm.py
index b0d5a72b2a..f038009669 100755
--- a/mindspore/nn/layer/lstm.py
+++ b/mindspore/nn/layer/lstm.py
@@ -95,6 +95,9 @@ class LSTM(Cell):
         - **hx_n** (tuple) - A tuple of two Tensor (h_n, c_n) both of shape
           (num_directions * `num_layers`, batch_size, `hidden_size`).
 
+    Supported Platforms:
+        ``GPU``
+
     Examples:
         >>> net = nn.LSTM(10, 12, 2, has_bias=True, batch_first=True, bidirectional=False)
         >>> input = Tensor(np.ones([3, 5, 10]).astype(np.float32))
@@ -303,6 +306,9 @@ class LSTMCell(Cell):
         - **reserve** - reserved
         - **state** - reserved
 
+    Supported Platforms:
+        ``GPU`` ``CPU``
+
     Examples:
         >>> net = nn.LSTMCell(10, 12, has_bias=True, batch_first=True, bidirectional=False)
         >>> input = Tensor(np.ones([3, 5, 10]).astype(np.float32))
diff --git a/mindspore/nn/layer/math.py b/mindspore/nn/layer/math.py
index f4a86e3e06..c11b943a89 100644
--- a/mindspore/nn/layer/math.py
+++ b/mindspore/nn/layer/math.py
@@ -60,6 +60,9 @@ class ReduceLogSumExp(Cell):
         - If axis is tuple(int), set as (2, 3), and keep_dims is False,
           the shape of output is :math:`(x_1, x_4, ..., x_R)`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> input_x = Tensor(np.random.randn(3, 4, 5, 6).astype(np.float32))
         >>> op = nn.ReduceLogSumExp(1, keep_dims=True)
@@ -98,6 +101,9 @@ class Range(Cell):
     Outputs:
         Tensor, the dtype is int if the dtype of `start`, `limit` and `delta` all are int. Otherwise, dtype is float.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> net = nn.Range(1, 8, 2)
         >>> output = net()
@@ -152,6 +158,9 @@ class LinSpace(Cell):
     Outputs:
         Tensor, With type same as `start`. The shape is 1-D with length of `num`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> linspace = nn.LinSpace(1, 10, 5)
         >>> output = linspace()
@@ -225,6 +234,9 @@ class LGamma(Cell):
     Outputs:
         Tensor, has the same shape and dtype as the `input_x`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> input_x = Tensor(np.array([2, 3, 4]).astype(np.float32))
         >>> op = nn.LGamma()
@@ -758,6 +770,9 @@ class MatMul(Cell):
     Outputs:
         Tensor, the shape of the output tensor depends on the dimension of input tensors.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> net = nn.MatMul()
         >>> input_x1 = Tensor(np.ones(shape=[3, 2, 3]), mindspore.float32)
@@ -830,6 +845,9 @@ class Moments(Cell):
         - **mean** (Tensor) - The mean of input x, with the same date type as input x.
         - **variance** (Tensor) - The variance of input x, with the same date type as input x.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> net = nn.Moments(axis=3, keep_dims=True)
         >>> input_x = Tensor(np.array([[[[1, 2, 3, 4], [3, 4, 5, 6]]]]), mindspore.float32)
diff --git a/mindspore/nn/layer/normalization.py b/mindspore/nn/layer/normalization.py
index 5dc7da7183..cc477bc23a 100644
--- a/mindspore/nn/layer/normalization.py
+++ b/mindspore/nn/layer/normalization.py
@@ -283,6 +283,9 @@ class BatchNorm1d(_BatchNorm):
     Outputs:
         Tensor, the normalized, scaled, offset tensor, of shape :math:`(N, C_{out})`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> net = nn.BatchNorm1d(num_features=4)
         >>> np.random.seed(0)
@@ -369,6 +372,9 @@ class BatchNorm2d(_BatchNorm):
     Outputs:
         Tensor, the normalized, scaled, offset tensor, of shape :math:`(N, C_{out}, H_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> net = nn.BatchNorm2d(num_features=3)
         >>> np.random.seed(0)
@@ -458,6 +464,9 @@ class GlobalBatchNorm(_BatchNorm):
     Outputs:
         Tensor, the normalized, scaled, offset tensor, of shape :math:`(N, C_{out}, H_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> # This example should be run with multiple processes. Refer to the run_distribute_train.sh
         >>> import os
@@ -557,6 +566,9 @@ class LayerNorm(Cell):
     Outputs:
         Tensor, the normalized and scaled offset tensor, has the same shape and data type as the `input_x`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> x = Tensor(np.ones([20, 5, 10, 10]), mindspore.float32)
         >>> shape1 = x.shape[1:]
@@ -630,6 +642,9 @@ class GroupNorm(Cell):
     Outputs:
         Tensor, the normalized and scaled offset tensor, has the same shape and data type as the `input_x`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> goup_norm_op = nn.GroupNorm(2, 2)
         >>> x = Tensor(np.ones([1, 2, 4, 4], np.float32))
diff --git a/mindspore/nn/layer/pooling.py b/mindspore/nn/layer/pooling.py
index 08828e34df..c28330e7de 100644
--- a/mindspore/nn/layer/pooling.py
+++ b/mindspore/nn/layer/pooling.py
@@ -104,6 +104,9 @@ class MaxPool2d(_PoolNd):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, H_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> pool = nn.MaxPool2d(kernel_size=3, stride=1)
         >>> x = Tensor(np.random.randint(0, 10, [1, 2, 4, 4]), mindspore.float32)
@@ -171,6 +174,9 @@ class MaxPool1d(_PoolNd):
     Outputs:
         Tensor of shape :math:`(N, C, L_{out}))`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> max_pool = nn.MaxPool1d(kernel_size=3, stride=1)
         >>> x = Tensor(np.random.randint(0, 10, [1, 2, 4]), mindspore.float32)
@@ -257,6 +263,9 @@ class AvgPool2d(_PoolNd):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, H_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> pool = nn.AvgPool2d(kernel_size=3, stride=1)
         >>> x = Tensor(np.random.randint(0, 10, [1, 2, 4, 4]), mindspore.float32)
@@ -322,6 +331,9 @@ class AvgPool1d(_PoolNd):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, L_{out})`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> pool = nn.AvgPool1d(kernel_size=6, stride=1)
         >>> x = Tensor(np.random.randint(0, 10, [1, 3, 6]), mindspore.float32)
diff --git a/mindspore/nn/layer/quant.py b/mindspore/nn/layer/quant.py
index fead4d6ed9..4baaf38066 100644
--- a/mindspore/nn/layer/quant.py
+++ b/mindspore/nn/layer/quant.py
@@ -587,6 +587,9 @@ class Conv2dBnFoldQuant(Cell):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, H_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> qconfig = compression.quant.create_quant_config()
         >>> conv2d_bnfold = nn.Conv2dBnFoldQuant(1, 6, kernel_size=(2, 2), stride=(1, 1), pad_mode="valid",
@@ -772,6 +775,9 @@ class Conv2dBnWithoutFoldQuant(Cell):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, H_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> qconfig = compression.quant.create_quant_config()
         >>> conv2d_no_bnfold = nn.Conv2dBnWithoutFoldQuant(1, 6, kernel_size=(2, 2), stride=(1, 1), pad_mode="valid",
@@ -893,6 +899,9 @@ class Conv2dQuant(Cell):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, H_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> qconfig = compression.quant.create_quant_config()
         >>> conv2d_quant = nn.Conv2dQuant(1, 6, kernel_size= (2, 2), stride=(1, 1), pad_mode="valid",
@@ -994,6 +1003,9 @@ class DenseQuant(Cell):
     Outputs:
         Tensor of shape :math:`(N, C_{out}, H_{out}, W_{out})`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> qconfig = compression.quant.create_quant_config()
         >>> dense_quant = nn.DenseQuant(3, 6, quant_config=qconfig)
@@ -1100,6 +1112,9 @@ class ActQuant(_QuantActivation):
     Outputs:
         Tensor, with the same type and shape as the `input`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> qconfig = compression.quant.create_quant_config()
         >>> act_quant = nn.ActQuant(nn.ReLU(), quant_config=qconfig)
@@ -1162,6 +1177,9 @@ class TensorAddQuant(Cell):
     Outputs:
         Tensor, with the same type and shape as the `input_x1`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> qconfig = compression.quant.create_quant_config()
         >>> add_quant = nn.TensorAddQuant(quant_config=qconfig)
@@ -1210,6 +1228,9 @@ class MulQuant(Cell):
     Outputs:
         Tensor, with the same type and shape as the `input_x1`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> qconfig = compression.quant.create_quant_config()
         >>> mul_quant = nn.MulQuant(quant_config=qconfig)
diff --git a/mindspore/nn/loss/loss.py b/mindspore/nn/loss/loss.py
index 36fb500ccd..25d61e8454 100644
--- a/mindspore/nn/loss/loss.py
+++ b/mindspore/nn/loss/loss.py
@@ -91,6 +91,9 @@ class L1Loss(_Loss):
     Outputs:
         Tensor, loss float tensor.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> loss = nn.L1Loss()
         >>> input_data = Tensor(np.array([1, 2, 3]), mindspore.float32)
@@ -133,6 +136,9 @@ class MSELoss(_Loss):
     Outputs:
         Tensor, weighted loss float tensor.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> loss = nn.MSELoss()
         >>> input_data = Tensor(np.array([1, 2, 3]), mindspore.float32)
@@ -180,6 +186,9 @@ class SmoothL1Loss(_Loss):
     Outputs:
         Tensor, loss float tensor.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> loss = nn.SmoothL1Loss()
         >>> input_data = Tensor(np.array([1, 2, 3]), mindspore.float32)
@@ -234,6 +243,9 @@ class SoftmaxCrossEntropyWithLogits(_Loss):
         Tensor, a tensor of the same shape as logits with the component-wise
         logistic losses.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> loss = nn.SoftmaxCrossEntropyWithLogits(sparse=True)
         >>> logits = Tensor(np.random.randint(0, 9, [1, 10]), mindspore.float32)
@@ -511,6 +523,9 @@ class BCELoss(_Loss):
         Tensor or Scalar, if `reduction` is 'none', then output is a tensor and has the same shape as `inputs`.
         Otherwise, the output is a scalar.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> weight = Tensor(np.array([[1.0, 2.0, 3.0], [4.0, 3.3, 2.2]]), mindspore.float32)
         >>> loss = nn.BCELoss(weight=weight, reduction='mean')
@@ -570,6 +585,9 @@ class CosineEmbeddingLoss(_Loss):
         - **loss** (Tensor) - If `reduction` is "none", its shape is the same as `y`'s shape, otherwise a scalar value
           will be returned.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> x1 = Tensor(np.array([[0.3, 0.8], [0.4, 0.3]]), mindspore.float32)
         >>> x2 = Tensor(np.array([[0.4, 1.2], [-0.4, -0.9]]), mindspore.float32)
diff --git a/mindspore/nn/optim/adam.py b/mindspore/nn/optim/adam.py
index bd1cd16319..d5d2d2d4e0 100755
--- a/mindspore/nn/optim/adam.py
+++ b/mindspore/nn/optim/adam.py
@@ -272,6 +272,9 @@ class Adam(Optimizer):
     Outputs:
         Tensor[bool], the value is True.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> net = Net()
         >>> #1) All parameters use the same learning rate and weight decay
@@ -407,6 +410,9 @@ class AdamWeightDecay(Optimizer):
     Outputs:
         tuple[bool], all elements are True.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = Net()
         >>> #1) All parameters use the same learning rate and weight decay
@@ -535,6 +541,9 @@ class AdamOffload(Optimizer):
     Outputs:
         Tensor[bool], the value is True.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> net = Net()
         >>> #1) All parameters use the same learning rate and weight decay
diff --git a/mindspore/nn/optim/ftrl.py b/mindspore/nn/optim/ftrl.py
index 013dcea198..5e9ebd1934 100644
--- a/mindspore/nn/optim/ftrl.py
+++ b/mindspore/nn/optim/ftrl.py
@@ -126,6 +126,9 @@ class FTRL(Optimizer):
     Outputs:
         tuple[Parameter], the updated parameters, the shape is the same as `params`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = Net()
         >>> #1) All parameters use the same learning rate and weight decay
diff --git a/mindspore/nn/optim/lamb.py b/mindspore/nn/optim/lamb.py
index b283125f81..99f4b17978 100755
--- a/mindspore/nn/optim/lamb.py
+++ b/mindspore/nn/optim/lamb.py
@@ -235,6 +235,9 @@ class Lamb(Optimizer):
     Outputs:
         tuple[bool], all elements are True.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = Net()
         >>> #1) All parameters use the same learning rate and weight decay
diff --git a/mindspore/nn/optim/lars.py b/mindspore/nn/optim/lars.py
index 65187c4cc9..4f2dfaf2c2 100755
--- a/mindspore/nn/optim/lars.py
+++ b/mindspore/nn/optim/lars.py
@@ -68,6 +68,9 @@ class LARS(Optimizer):
     Outputs:
         Union[Tensor[bool], tuple[Parameter]], it depends on the output of `optimizer`.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> net = Net()
         >>> loss = nn.SoftmaxCrossEntropyWithLogits()
diff --git a/mindspore/nn/optim/lazyadam.py b/mindspore/nn/optim/lazyadam.py
index 92dc0ecd53..7f755b1392 100644
--- a/mindspore/nn/optim/lazyadam.py
+++ b/mindspore/nn/optim/lazyadam.py
@@ -183,6 +183,9 @@ class LazyAdam(Optimizer):
     Outputs:
         Tensor[bool], the value is True.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> net = Net()
         >>> #1) All parameters use the same learning rate and weight decay
diff --git a/mindspore/nn/optim/momentum.py b/mindspore/nn/optim/momentum.py
index 390324414c..2e74f03264 100755
--- a/mindspore/nn/optim/momentum.py
+++ b/mindspore/nn/optim/momentum.py
@@ -104,6 +104,9 @@ class Momentum(Optimizer):
     Raises:
         ValueError: If the momentum is less than 0.0.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU`` ``CPU``
+
     Examples:
         >>> net = Net()
         >>> #1) All parameters use the same learning rate and weight decay
diff --git a/mindspore/nn/optim/optimizer.py b/mindspore/nn/optim/optimizer.py
index 3835c30bd3..e4322a4072 100755
--- a/mindspore/nn/optim/optimizer.py
+++ b/mindspore/nn/optim/optimizer.py
@@ -83,6 +83,9 @@ class Optimizer(Cell):
     Raises:
         ValueError: If the learning_rate is a Tensor, but the dimension of tensor is greater than 1.
         TypeError: If the learning_rate is not any of the three types: float, Tensor, nor Iterable.
+
+    Supported Platforms:
+        ``Ascend`` ``GPU``
     """
 
     def __init__(self, learning_rate, parameters, weight_decay=0.0, loss_scale=1.0):
diff --git a/mindspore/nn/optim/proximal_ada_grad.py b/mindspore/nn/optim/proximal_ada_grad.py
index 923c6feb73..bf458fbb65 100644
--- a/mindspore/nn/optim/proximal_ada_grad.py
+++ b/mindspore/nn/optim/proximal_ada_grad.py
@@ -107,6 +107,9 @@ class ProximalAdagrad(Optimizer):
     Outputs:
         Tensor[bool], the value is True.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> net = Net()
         >>> #1) All parameters use the same learning rate and weight decay
diff --git a/mindspore/nn/optim/rmsprop.py b/mindspore/nn/optim/rmsprop.py
index 658a14443e..5d364cd9eb 100644
--- a/mindspore/nn/optim/rmsprop.py
+++ b/mindspore/nn/optim/rmsprop.py
@@ -130,6 +130,9 @@ class RMSProp(Optimizer):
     Outputs:
         Tensor[bool], the value is True.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = Net()
         >>> #1) All parameters use the same learning rate and weight decay
diff --git a/mindspore/nn/optim/sgd.py b/mindspore/nn/optim/sgd.py
index d3d438b334..f916638b56 100755
--- a/mindspore/nn/optim/sgd.py
+++ b/mindspore/nn/optim/sgd.py
@@ -103,6 +103,9 @@ class SGD(Optimizer):
     Raises:
         ValueError: If the momentum, dampening or weight_decay value is less than 0.0.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = Net()
         >>> #1) All parameters use the same learning rate and weight decay
diff --git a/mindspore/nn/wrap/cell_wrapper.py b/mindspore/nn/wrap/cell_wrapper.py
index c795ba2b05..37b730eb90 100644
--- a/mindspore/nn/wrap/cell_wrapper.py
+++ b/mindspore/nn/wrap/cell_wrapper.py
@@ -78,6 +78,9 @@ class WithLossCell(Cell):
     Outputs:
         Tensor, a scalar tensor with shape :math:`()`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = Net()
         >>> loss_fn = nn.SoftmaxCrossEntropyWithLogits(sparse=False)
@@ -134,6 +137,9 @@ class WithGradCell(Cell):
     Outputs:
         list, a list of Tensors with identical shapes as trainable weights.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> # For a defined network Net without loss function
         >>> net = Net()
@@ -187,6 +193,9 @@ class TrainOneStepCell(Cell):
     Outputs:
         Tensor, a scalar Tensor with shape :math:`()`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> net = Net()
         >>> loss_fn = nn.SoftmaxCrossEntropyWithLogits()
@@ -249,6 +258,9 @@ class GetNextSingleOp(Cell):
 
     For detailed information, refer to `ops.operations.GetNext`.
 
+    Supported Platforms:
+        ``GPU``
+
     Examples:
         >>> # Refer to dataset_helper.py for detail usage.
         >>> data_set = get_dataset()
@@ -346,6 +358,9 @@ class WithEvalCell(Cell):
         Tuple, containing a scalar loss Tensor, a network output Tensor of shape :math:`(N, \ldots)`
         and a label Tensor of shape :math:`(N, \ldots)`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> # For a defined network Net without loss function
         >>> net = Net()
@@ -380,6 +395,9 @@ class ParameterUpdate(Cell):
     Raises:
         KeyError: If parameter with the specified name does not exist.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> network = Net()
         >>> param = network.parameters_dict()['learning_rate']
diff --git a/mindspore/nn/wrap/grad_reducer.py b/mindspore/nn/wrap/grad_reducer.py
index a19ad30c5f..b3cd4c1e4c 100644
--- a/mindspore/nn/wrap/grad_reducer.py
+++ b/mindspore/nn/wrap/grad_reducer.py
@@ -245,6 +245,9 @@ class DistributedGradReducer(Cell):
     Raises:
         ValueError: If degree is not a int or less than 0.
 
+    Supported Platforms:
+        ``Ascend``
+
     Examples:
         >>> # This example should be run with multiple processes. Refer to the run_distribute_train.sh
         >>> import os
diff --git a/mindspore/nn/wrap/loss_scale.py b/mindspore/nn/wrap/loss_scale.py
index 9a53602c85..cb1f0b4844 100644
--- a/mindspore/nn/wrap/loss_scale.py
+++ b/mindspore/nn/wrap/loss_scale.py
@@ -75,6 +75,9 @@ class DynamicLossScaleUpdateCell(Cell):
     Outputs:
         Tensor, a scalar Tensor with shape :math:`()`.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> import numpy as np
         >>> from mindspore import Tensor, Parameter, nn
@@ -155,6 +158,9 @@ class FixedLossScaleUpdateCell(Cell):
     Args:
         loss_scale_value (float): Initializes loss scale.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> import numpy as np
         >>> from mindspore import Tensor, Parameter, nn
@@ -220,6 +226,9 @@ class TrainOneStepWithLossScaleCell(TrainOneStepCell):
         - **overflow** (Tensor) -  Tensor with shape :math:`()`, type is bool.
         - **loss scaling value** (Tensor) -  Tensor with shape :math:`()`
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> import numpy as np
         >>> from mindspore import Tensor, Parameter, nn
diff --git a/mindspore/ops/operations/array_ops.py b/mindspore/ops/operations/array_ops.py
index bb7a591915..085b3fcdaa 100644
--- a/mindspore/ops/operations/array_ops.py
+++ b/mindspore/ops/operations/array_ops.py
@@ -1135,6 +1135,9 @@ class Ones(PrimitiveWithInfer):
     Outputs:
         Tensor, has the same type and shape as input shape value.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> from mindspore.ops import operations as P
         >>> ones = P.Ones()
@@ -1185,6 +1188,9 @@ class Zeros(PrimitiveWithInfer):
     Outputs:
         Tensor, has the same type and shape as input shape value.
 
+    Supported Platforms:
+        ``Ascend`` ``GPU``
+
     Examples:
         >>> from mindspore.ops import operations as P
         >>> zeros = P.Zeros()