!11097 Fix some api comments.

From: @zhang_yi2020
Reviewed-by: @gemini524,@wuxuejian,@liangchenghui
Signed-off-by: @wuxuejian

mindspore/nn/layer/conv.py

@@ -191,10 +191,10 @@ class Conv2d(_Conv):
 
     Inputs:
         - **input** (Tensor) - Tensor of shape :math:`(N, C_{in}, H_{in}, W_{in})` \
-          or `(N, H_{in}, W_{in}, C_{in})`.
+          or :math:`(N, H_{in}, W_{in}, C_{in})`.
 
     Outputs:
-        Tensor of shape :math:`(N, C_{out}, H_{out}, W_{out})` or `(N, H_{out}, W_{out}, C_{out})`.
+        Tensor of shape :math:`(N, C_{out}, H_{out}, W_{out})` or :math:`(N, H_{out}, W_{out}, C_{out})`.
 
     Supported Platforms:
         ``Ascend`` ``GPU`` ``CPU``

mindspore/nn/layer/embedding.py

@@ -375,7 +375,7 @@ class MultiFieldEmbeddingLookup(EmbeddingLookup):
           Specifies the field id of elements of the input_indices. Type is Int32.
 
     Outputs:
-        Tensor, the shape of tensor is :math:`(batch_size, field_size, embedding_size)`. Type is Float32.
+        Tensor, the shape of tensor is :math:`(batch\_size, field\_size, embedding\_size)`. Type is Float32.
 
     Supported Platforms:
         ``Ascend`` ``GPU``