diff --git a/mindspore/common/parameter.py b/mindspore/common/parameter.py
index 995c981c10d..44f58b7facc 100644
--- a/mindspore/common/parameter.py
+++ b/mindspore/common/parameter.py
@@ -466,7 +466,13 @@ class Parameter(Tensor_):
 
     @property
     def requires_grad(self):
-        """Return whether the parameter requires gradient."""
+        """
+        Return whether the parameter requires gradient.
+
+        The main function of requires_grad is to tell auto grad to start recording operations on a Tensor.
+        If a Tensor has requires_grad=False, then Tensor requires_grad will make auto grad start recording
+        operations on the tensor.
+        """
         return self.param_info.requires_grad
 
     @requires_grad.setter
diff --git a/mindspore/nn/layer/basic.py b/mindspore/nn/layer/basic.py
index eb6baf65d82..9b44b18186e 100644
--- a/mindspore/nn/layer/basic.py
+++ b/mindspore/nn/layer/basic.py
@@ -679,9 +679,9 @@ class Pad(Cell):
                 x = [[1,2,3], [4,5,6], [7,8,9]].
                 # The above can be seen: 1st dimension of `x` is 3, 2nd dimension of `x` is 3.
                 # Substitute into the formula to get:
-                # 1st dimension of output is paddings[0][0] + 3 + paddings[0][1] = 1 + 3 + 1 = 4.
+                # 1st dimension of output is paddings[0][0] + 3 + paddings[0][1] = 1 + 3 + 1 = 5.
                 # 2nd dimension of output is paddings[1][0] + 3 + paddings[1][1] = 2 + 3 + 2 = 7.
-                # So the shape of output is (4, 7).
+                # So the shape of output is (5, 7).
 
         mode (str): Specifies padding mode. The optional values are "CONSTANT", "REFLECT", "SYMMETRIC".
             Default: "CONSTANT".
@@ -1007,6 +1007,13 @@ class Tril(Cell):
     """
     Returns a tensor with elements above the kth diagonal zeroed.
 
+    The lower triangular part of the matrix is defined as the elements on and below the diagonal.
+
+    The parameter `k` controls the diagonal to be considered.
+    If diagonal = 0, all elements on and below the main diagonal are retained.
+    Positive values include as many diagonals above the main diagonal, and similarly,
+    negative values exclude as many diagonals below the main diagonal.
+
     Inputs:
         - **x** (Tensor) - The input tensor. The data type is Number.
           :math:`(N,*)` where :math:`*` means, any number of additional dimensions.
@@ -1094,6 +1101,12 @@ class Triu(Cell):
     """
     Returns a tensor with elements below the kth diagonal zeroed.
 
+    The upper triangular part of the matrix is defined as the elements on and above the diagonal.
+
+    The parameter `k` controls the diagonal to be considered. If `k` = 0, all elements on and above the main diagonal
+    are retained. Positive values do not include as many diagonals above the main diagonal, and similarly,
+    negative values include as many diagonals below the main diagonal.
+
     Inputs:
         - **x** (Tensor) - The input tensor. The data type is Number.
           :math:`(N,*)` where :math:`*` means, any number of additional dimensions.
diff --git a/mindspore/nn/layer/lstm.py b/mindspore/nn/layer/lstm.py
index 8bcf26e5260..f27e6f8104f 100755
--- a/mindspore/nn/layer/lstm.py
+++ b/mindspore/nn/layer/lstm.py
@@ -74,8 +74,8 @@ class LSTM(Cell):
             f_t = \sigma(W_{fx} x_t + b_{fx} + W_{fh} h_{(t-1)} + b_{fh}) \\
             \tilde{c}_t = \tanh(W_{cx} x_t + b_{cx} + W_{ch} h_{(t-1)} + b_{ch}) \\
             o_t = \sigma(W_{ox} x_t + b_{ox} + W_{oh} h_{(t-1)} + b_{oh}) \\
-            c_t = f_t * c_{(t-1)} + i_t * \tilde{c}_t \\
-            h_t = o_t * \tanh(c_t) \\
+            c_t = f_t \odot c_{(t-1)} + i_t \odot \tilde{c}_t \\
+            h_t = o_t \odot \tanh(c_t) \\
         \end{array}
 
     Here :math:`\sigma` is the sigmoid function, and :math:`*` is the Hadamard product. :math:`W, b`
diff --git a/mindspore/ops/operations/array_ops.py b/mindspore/ops/operations/array_ops.py
index 4bbeba85006..92d9469f5fc 100755
--- a/mindspore/ops/operations/array_ops.py
+++ b/mindspore/ops/operations/array_ops.py
@@ -451,6 +451,9 @@ class Reshape(PrimitiveWithInfer):
     """
     Reshapes the input tensor with the same values based on a given shape tuple.
 
+    The 'input_shape' can only have one -1 at most, in which case it’s inferred from the remaining dimensions and
+    the number of elements in the input.
+
     Inputs:
         - **input_x** (Tensor) - The shape of tensor is :math:`(x_1, x_2, ..., x_R)`.
         - **input_shape** (tuple[int]) - The input tuple is constructed by multiple
@@ -701,6 +704,13 @@ class Transpose(Primitive):
     """
     Permutes the dimensions of the input tensor according to input permutation.
 
+    For a 1-D array this has no effect, as a transposed vector is simply the same vector.
+    To convert a 1-D array into a 2D column vecto please refer the class: mindspore.ops.ExpandDims.
+    For a 2-D array, this is a standard matrix transpose. For an n-D array, if axes are given,
+    their order indicates how the axes are permuted (see Examples).
+    If axes are not provided and a.shape = (i[0], i[1], ... i[n-2], i[n-1]),
+    then a.transpose().shape = (i[n-1], i[n-2], ... i[1], i[0]).
+
     Inputs:
         - **input_x** (Tensor) - The shape of tensor is :math:`(x_1, x_2, ..., x_R)`.
         - **input_perm** (tuple[int]) - The permutation to be converted. The elements in `input_perm` are composed of
diff --git a/mindspore/ops/operations/nn_ops.py b/mindspore/ops/operations/nn_ops.py
index c4ccc4431a3..b2fc5d784ae 100755
--- a/mindspore/ops/operations/nn_ops.py
+++ b/mindspore/ops/operations/nn_ops.py
@@ -3435,9 +3435,11 @@ class L2Normalize(PrimitiveWithInfer):
     This operator will normalize the input using the given axis. The function is shown as follows:
 
     .. math::
-        \text{output} = \frac{x}{\sqrt{\text{max}(\text{sum} (\text{x}^2), \epsilon)}},
+        \displaylines{{\text{output} = \frac{x}{\sqrt{\text{max}(\parallel x_i \parallel^p , \epsilon)} } } \\
+        {\parallel x_i \parallel^p = (\sum_{i}^{}\left | x_i  \right | ^p  )^{1/p}} }
 
-    where :math:`\epsilon` is epsilon.
+    where :math:`\epsilon` is epsilon amd :math:`\sum_{i}^{}\left | x_i  \right | ^p` calculate
+    along the dimension `axis`.
 
     Args:
         axis (Union[list(int), tuple(int), int]): The starting axis for the input to apply the L2 Normalization.