diff --git a/docs/api/api_python/ops/mindspore.ops.CombinedNonMaxSuppression.rst b/docs/api/api_python/ops/mindspore.ops.CombinedNonMaxSuppression.rst
index 0cb4f74f5f9..5aa2471c11c 100644
--- a/docs/api/api_python/ops/mindspore.ops.CombinedNonMaxSuppression.rst
+++ b/docs/api/api_python/ops/mindspore.ops.CombinedNonMaxSuppression.rst
@@ -3,15 +3,24 @@
 
 .. py:class:: mindspore.ops.CombinedNonMaxSuppression(clip_boxes=True, pad_per_class=False)
 
-    根据分数降序，使用非极大抑制法通过遍历所有可能的边界框（Bounding Box）来选择一个最优的结果。
+    使用非极大值抑制法遍历候选边界框列表，从中选择一组子集，其中边界框按其置信度得分降序排列。
 
     参数：
-        - **clip_boxes** (bool, 可选) - 如果为True，则假设边界框坐标在[0,1]之间，如果超出[0,1]则剪辑输出框。如果为False，则不进行剪切并按原样输出框坐标。默认值：True。
-        - **pad_per_class** (bool, 可选) - 如果为True，输出 `nmsed_boxes` 、`nmsed_scores` 和 `nmsed_classes` 将被填充为 `max_output_size_per_class` * num_classes的长度，如果该长度超过 `max_total_size` ，在这种情况下它将被裁剪为 `max_total_size` 。如果为False，则输出 `nmsed_boxes` 、 `nmsed_scores` 和 `nmsed_classes` 将被填充/裁剪到 `max_total_size` 。默认值：False。
+        - **clip_boxes** (bool, 可选) - 确定是否应用边界框归一化，以确保坐标在[0,1]范围内。
+          默认值：True。
+
+          - 如果为True，则剪裁超出此范围的框。
+          - 如果为False，则返回框坐标，而不进行任何修改。
+
+        - **pad_per_class** (bool, 可选) - 确定是否需要对非极大值抑制（NMS）算法的输出进行填充或剪裁，以满足最大尺寸的限制。
+          默认值：False。
+
+          - 如果为False，则将输出剪裁到最大尺寸 `max_total_size` 。
+          - 如果为True，则将输出填充到 `max_size_per_class` * `num_classes` 的最大长度，如果超 `过max_total_size` 则剪裁。
 
     输入：
-        - **boxes** (Tensor) - Shape可表示为：(batch_size, num_boxes, q, 4)。如果q为1，则对所有类都使用相同的边界框；如果q等于类的数量，则对于每一类都使用特定的边界框。数据类型：float32。
-        - **scores** (Tensor) - 表示对应于每个Bounding Boxes( `boxes` 的每一行)的单个分数，数据类型必须为float32，其shape可表示为：(batch_size, num_boxes, num_classes)。
+        - **boxes** (Tensor) - 边界框坐标，是一个float32类型的Tensor，shape为 :math:`(batch_size, num_boxes, q, 4)` 。如果q为1，则所有类别都使用相同的边界框。否则，如果q等于类的数量，则对于每一类都使用特定的边界框。
+        - **scores** (Tensor) - 表示对应于每个Bounding Boxes( `boxes` 的每一行)的单个分数，数据类型必须为float32，其shape可表示为： :math:`(batch_size, num_boxes, num_classes)` 。
         - **max_output_size_per_class** (Tensor) - 0D Tensor，表示每个类中由非极大抑制法（non-max suppression）选择的Bounding Boxes数目的上限。数据类型：int32。
         - **max_total_size** (Tensor) - 0D Tensor，表示在所有类中可保留的Bounding Boxes数目的上限。数据类型：int32。
         - **iou_threshold** (Tensor) - 0D Tensor，判断Bounding Boxes是否与IOU重叠过多的阈值，取值必须在[0,1]区间内。数据类型：float32。
@@ -33,4 +42,4 @@
         - **ValueError** -  `scores` 不是3D Tensor。
         - **ValueError** -  `max_total_size` 小于0。
         - **ValueError** -  `max_output_size_per_class` 小于0。
-        - **ValueError** -  `iou_threshold` 取值不在区间[0,1]中。
\ No newline at end of file
+        - **ValueError** -  `iou_threshold` 取值不在区间[0,1]中。
diff --git a/docs/api/api_python/ops/mindspore.ops.DataFormatVecPermute.rst b/docs/api/api_python/ops/mindspore.ops.DataFormatVecPermute.rst
index d19dbcfa30c..36c318e430b 100644
--- a/docs/api/api_python/ops/mindspore.ops.DataFormatVecPermute.rst
+++ b/docs/api/api_python/ops/mindspore.ops.DataFormatVecPermute.rst
@@ -3,14 +3,14 @@
 
 .. py:class:: mindspore.ops.DataFormatVecPermute(src_format='NHWC', dst_format='NCHW')
 
-    将输入按从 `src_format` 到 `dst_format` 的变化重新排列。
+    将输入Tensor的数据格式从 `src_format` 转换为 `dst_format` ，通过重新排列其维度来实现。
 
     参数：
-        - **src_format** (str, 可选) - 原先的数据排列格式，可以为'NHWC'和'NCHW'之一。默认值：'NHWC'。
+        - **src_format** (str, 可选) - 初始数据排列格式，可以为'NHWC'和'NCHW'之一。默认值：'NHWC'。
         - **dst_format** (str, 可选) - 目标数据排列格式，可以为'NHWC'和'NCHW'之一。默认值：'NCHW'。
 
     输入：
-        - **input_x** (Tensor) - shape为(4, )或(4, 2)的输入Tensor。数据类型为int32或int64。
+        - **input_x** (Tensor) - shape为 :math:`(4, )` 或 :math:`(4, 2)` 的输入Tensor。数据类型为int32或int64。
 
     输出：
         与 `input_x` 的shape和数据类型一致的Tensor。
@@ -18,5 +18,5 @@
     异常：
         - **TypeError** - 输入 `input_x` 不是Tensor。
         - **TypeError** -  `input_x` 的数据类型不是int32或int64。
-        - **ValueError** -  `input_x` 的shape不为(4, )或(4, 2)。
-        - **ValueError** -  `src_format` 或 `dst_format` 不是'NHWC'或'NCHW'之一。
\ No newline at end of file
+        - **ValueError** -  `input_x` 的shape不为 :math:`(4, )` or :math:`(4, 2)` 。
+        - **ValueError** -  `src_format` 或 `dst_format` 不是'NHWC'或'NCHW'之一。
diff --git a/docs/api/api_python/ops/mindspore.ops.DiagPart.rst b/docs/api/api_python/ops/mindspore.ops.DiagPart.rst
index f197e5a326a..a74d6dfb31b 100644
--- a/docs/api/api_python/ops/mindspore.ops.DiagPart.rst
+++ b/docs/api/api_python/ops/mindspore.ops.DiagPart.rst
@@ -3,7 +3,7 @@ mindspore.ops.DiagPart
 
 .. py:class:: mindspore.ops.DiagPart
 
-    返回输入的对角线部分。
+    提取输入Tensor的对角线元素。
 
     假如 `input_x` 有维度 :math:`[D_1,..., D_k, D_1,..., D_k]`，那么输出是一个秩为k的Tensor，维度为 :math:`[D_1,..., D_k]`，其中：
 
diff --git a/docs/api/api_python/ops/mindspore.ops.EuclideanNorm.rst b/docs/api/api_python/ops/mindspore.ops.EuclideanNorm.rst
index 2b8c9cce787..20c675d7835 100644
--- a/docs/api/api_python/ops/mindspore.ops.EuclideanNorm.rst
+++ b/docs/api/api_python/ops/mindspore.ops.EuclideanNorm.rst
@@ -3,15 +3,15 @@
 
 .. py:class:: mindspore.ops.EuclideanNorm(keep_dims=False)
 
-    计算Tensor维度上元素的欧几里得范数，根据给定的轴对输入进行规约操作。
+    计算Tensor沿指定轴的欧几里得范数（又称L2范数）。默认情况下会移除axis中指定的维度。
 
     参数：
         - **keep_dims** (bool，可选) - 如果为True，被规约的轴保留为1，如果为False，不保留给定的这些轴，默认值：False。
 
     输入：
-        - **x** (Tensor) - 输入Tensor，将被规约，数据类型为：float16、float32、float64、int8、int16、
+        - **x** (Tensor) - 输入Tensor，数据类型为：float16、float32、float64、int8、int16、
           int32、int64、complex64、complex128、uint8、uint16、uint32、uint64。
-        - **axes** (Tensor) - 将进行规约的轴。数据类型为：int32、int64。取值范围为： `[-rank(x), rank(x))`。
+        - **axes** (Tensor) - 指定进行规约计算的维度。数据类型为：int32、int64。取值范围为： `[-rank(x), rank(x))` 。
 
     输出：
         Tensor，与输入 `x` 具有相同的数据类型。
diff --git a/docs/api/api_python/ops/mindspore.ops.Expand.rst b/docs/api/api_python/ops/mindspore.ops.Expand.rst
index a8d19755747..dfa6b40aef5 100644
--- a/docs/api/api_python/ops/mindspore.ops.Expand.rst
+++ b/docs/api/api_python/ops/mindspore.ops.Expand.rst
@@ -3,6 +3,6 @@
 
 .. py:class:: mindspore.ops.Expand
 
-    返回一个当前张量的新视图，其中单维度扩展到更大的尺寸。
+    沿着Tensor单维度进行扩展，以匹配给定的目标shape。
 
     更多细节请参考 :func:`mindspore.ops.expand`。
diff --git a/docs/api/api_python/ops/mindspore.ops.FillDiagonal.rst b/docs/api/api_python/ops/mindspore.ops.FillDiagonal.rst
index 6f590b5c1e0..08fbc8d16a9 100644
--- a/docs/api/api_python/ops/mindspore.ops.FillDiagonal.rst
+++ b/docs/api/api_python/ops/mindspore.ops.FillDiagonal.rst
@@ -3,17 +3,17 @@ mindspore.ops.FillDiagonal
 
 .. py:class:: mindspore.ops.FillDiagonal(fill_value, wrap=False)
 
-    填充至少具有二维的Tensor的主对角线。当 `dims>2` 时，输入的所有维度必须等长。此函数就地修改输入Tensor，并返回输入Tensor。
+    填充至少具有二维的Tensor的主对角线。当输入的维度大于2时，所有维度必须等长。此函数原地修改输入Tensor，并返回输入Tensor。
 
     参数：
-        - **fill_value** (float) - 填充值。
-        - **wrap** (bool，可选) - 如果设置为True, 表示高矩阵N列之后的对角线 `wrapped` ，否则不处理，默认值：False。
+        - **fill_value** (float) - `input_x` 对角线的填充值。
+        - **wrap** (bool，可选) - 控制对于一个高矩阵（即矩阵的行数大于列数），对角线元素是否继续延伸到剩余的行。下面的例子说明产生怎样的效果。默认值：False。
 
     输入：
         - **input_x** (Tensor) - shape为 :math:`(x_1, x_2, ..., x_R)` ，其数据类型必须为：float32、int32或者int64。
 
     输出：
-        - **y** (Tensor) - Tensor，和输入 `x` 具有相同的shape和dtype。
+        - **y** (Tensor) - Tensor，和输入 `input_x` 具有相同的shape和dtype。
 
     异常：
         - **TypeError** - 如果 `input_x` 的dtype不是：float32、int32或者int64。
diff --git a/mindspore/python/mindspore/ops/operations/array_ops.py b/mindspore/python/mindspore/ops/operations/array_ops.py
index 37fbe5137c6..32992be69a9 100755
--- a/mindspore/python/mindspore/ops/operations/array_ops.py
+++ b/mindspore/python/mindspore/ops/operations/array_ops.py
@@ -176,7 +176,7 @@ def _check_infer_attr_reduce(axis, keep_dims, prim_name):
 
 class Expand(Primitive):
     """
-    Returns a new view of the self tensor with singleton dimensions expanded to a larger size.
+    Expands the Tensor along singleton dimensions to match given desired shape.
 
     Refer to :func:`mindspore.ops.expand` for more details.
 
@@ -3571,10 +3571,10 @@ class Diag(PrimitiveWithCheck):
 class DiagPart(PrimitiveWithCheck):
     r"""
 
-    Extracts the diagonal part from given tensor.
+    Extracts the diagonal elements from the given Tensor.
 
-    Assume input has dimensions :math:`[D_1,..., D_k, D_1,..., D_k]`, the output is a tensor
-    of rank k with dimensions :math:`[D_1,..., D_k]` where:
+    If the input is a Tensor of shape :math:`[D_1,..., D_k, D_1,..., D_k]`, then the
+    output will be a Tensor of rank k of shape :math:`[D_1,..., D_k]` where:
     :math:`output[i_1,..., i_k] = input[i_1,..., i_k, i_1,..., i_k]`.
 
     Inputs:
@@ -7586,20 +7586,23 @@ class LeftShift(Primitive):
 
 class FillDiagonal(Primitive):
     """
-    Fill the main diagonal of a tensor that has at least 2-dimensions.
-    When dims>2, all dimensions of input must be of equal length.
-    This function modifies the input tensor in-place, and returns the input tensor.
+    Fills the main diagonal of a Tensor in-place with a specified value and returns the result.
+    The input has at least 2 dimensions, and all dimensions of input must be equal in length
+    when the dimension of input is greater than 2.
 
     Args:
-        fill_value (float): The fill value.
-        wrap (bool, optional): the diagonal `wrapped` after N columns for tall matrices. Default: False.
+        fill_value (float): The value to fill the diagonal of `input_x`.
+        wrap (bool, optional): Controls whether the diagonal elements continue onto the
+            remaining rows in case of a tall matrix(A matrix has more rows than columns).
+            Examples blow demonstrates how it works on a tall matrix if `wrap` is set True.
+            Default: False.
 
     Inputs:
         - **input_x** (Tensor) - The shape of tensor is :math:`(x_1, x_2, ..., x_R)`.
           The data type must be float32, int32 or int64.
 
     Outputs:
-        - **y** (Tensor) - Tensor, has the same shape and data type as the input `x`.
+        - **y** (Tensor) - Tensor, has the same shape and data type as the input `input_x`.
 
     Raises:
         TypeError: If data type of `input_x` is not one of the following: float32, int32, int64.
diff --git a/mindspore/python/mindspore/ops/operations/image_ops.py b/mindspore/python/mindspore/ops/operations/image_ops.py
index d12476fcbc1..5a3ea696878 100644
--- a/mindspore/python/mindspore/ops/operations/image_ops.py
+++ b/mindspore/python/mindspore/ops/operations/image_ops.py
@@ -1066,33 +1066,40 @@ class ScaleAndTranslate(Primitive):
 
 class CombinedNonMaxSuppression(Primitive):
     r"""
-    Greedily selects a subset of bounding boxes in descending order of score.
+    Applies a greedy approach to select a subset of bounding boxes from a list of
+    candidates using NonMaxSuppression, where the boxes are sorted in descending order of their confidence score.
 
     Args:
-        clip_boxes (bool, optional): If true, assume the box coordinates are between [0, 1] and clip the output boxes
-            if they fall beyond [0, 1]. If false, do not do clipping and output the box coordinates as it is.
-            Defaults to true.
-        pad_per_class (bool, optional): If false, the output nmsed boxes,
-            scores and classes are padded/clipped to max_total_size.
-            If true, the output nmsed boxes, scores and classes are padded to be of length
-            max_size_per_class * num_classes, unless it exceeds max_total_size in which case it is clipped to
-            max_total_size. Defaults to false.
+        clip_boxes (bool, optional): Determines whether to apply bounding box normalization to ensure the
+            coordinates are within [0, 1] range. Default: True.
+
+            - If True, clip the boxes that fall outside this range.
+            - If False, return the box coordinates as they are without any modifications.
+
+        pad_per_class (bool, optional): Determines whether the output of the non-maximum suppression (NMS)
+            algorithm should be padded or clipped to meet the maximum size constraints. Default: False.
+
+            - If False, the output is clipped to the maximum size of `max_total_size`.
+            - If True, the output is padded up to `max_size_per_class` * `num_classes` and clipped if
+              it exceeds `max_total_size`.
 
     Inputs:
-        - **boxes** (Tensor) - A Tensor of type float32 and shape (batch_size, num_boxes, q, 4).
-          If q is 1 then same boxes are used for all classes otherwise,
-          if q is equal to number of classes, class-specific boxes are used.
-        - **scores** (Tensor) - A Tensor of type float32 and shape (batch_size, num_boxes, num_classes)
-          representing a single score corresponding to each box (each row of boxes).
-        - **max_output_size_per_class** (Tensor) - A 0D Tensor of type int32, representing the max number of boxes to be
-          selected by non max suppression per class.
-        - **max_total_size** (Tensor) - A 0D Tensor of type int32, representing the maximum number of boxes retained
-          over all classes.
-        - **iou_threshold** (Tensor) - A 0D float32 tensor representing the threshold for deciding whether
-          boxes overlap too much with respect to IOU, and iou_threshold must be equal or greater
+        - **boxes** (Tensor) - A float32 Tensor with shape :math:`(batch_size, num_boxes, q, 4)`
+          representing the bounding box coordinates.
+          `q` can be 1, indicating that all classes use the same bounding box, or the number of classes,
+          indicating that class-specific boxes are applied.
+        - **scores** (Tensor) - A 3-D Tensor of float32 type with the shape
+          :math:`(batch_size, num_boxes, num_classes)`. It contains a score value for each box,
+          with each row of `boxes` represented by a single score.
+        - **max_output_size_per_class** (Tensor) - The maximum number of boxes that can be selected for each class
+          by the non-maximum suppression algorithm, represented by a scalar Tensor of type int32.
+        - **max_total_size** (Tensor) - A scalar Tensor of type int32 that represents the
+          maximum number of boxes that are kept for all classes.
+        - **iou_threshold** (Tensor) - A scalar Tensor of float32 type that represents the threshold for
+          determining if the IOU overlap between boxes is too high. `iou_threshold` must be equal or greater
           than 0 and be equal or smaller than 1.
-        - **score_threshold** (Tensor) - A 0D float32 tensor representing the threshold for deciding when to remove
-          boxes based on score.
+        - **score_threshold** (Tensor) - A scalar Tensor of type float32 that represents the threshold
+          for determining when to remove boxes based on their scores.
 
     Outputs:
         - **nmsed_boxes** - A Tensor of float32 with shape of (batch_size, num_detection, 4), which contains
diff --git a/mindspore/python/mindspore/ops/operations/math_ops.py b/mindspore/python/mindspore/ops/operations/math_ops.py
index 4b4481aeba8..425cfb40d7e 100644
--- a/mindspore/python/mindspore/ops/operations/math_ops.py
+++ b/mindspore/python/mindspore/ops/operations/math_ops.py
@@ -538,18 +538,17 @@ class _Reduce(PrimitiveWithCheck):
 
 class EuclideanNorm(Primitive):
     """
-    Computes the euclidean norm of elements across dimensions of a tensor.
-    Reduces input along the dimensions given in axis.
+    Calculates the Euclidean norm(aka L2 norm) of a Tensor along the specified axes.
+    The specified `axes` are removed by default.
 
     Args:
-        keep_dims (bool, optional): If true, the reduceed dimensions are retained with length 1.
-                          If false, don't keep these dimensions. Default: False.
+        keep_dims (bool, optional): whether to retain the reduced dimensions. If true, retains them with length 1.
+            If false, these dimensions are removed. Default: False.
 
     Inputs:
-        - **x** (Tensor) - The input tensor. Must be one of the following types :float16, float32, float64, int8, int16,
-          int32, int64, complex64, complex128, uint8, uint16, uint32, uint64. The tensor to reduce.
-        - **axes** (Tensor) - The dimensions to reduce. Must be one of the following types: int32, int64.
-          Must be in the range [-rank(x), rank(x)).
+        - **x** (Tensor) - The input Tensor to reduce. Must be one of the following types:
+          float16, float32, float64, int8, int16, int32, int64, complex64, complex128, uint8, uint16, uint32, uint64.
+        - **axes** (Tensor) - The axes to perform reduction on. Must be one of the following types: int32, int64.
 
     Outputs:
         Tensor, has the same type as the 'x'.
diff --git a/mindspore/python/mindspore/ops/operations/nn_ops.py b/mindspore/python/mindspore/ops/operations/nn_ops.py
index cc8e09c1d31..b933771615c 100644
--- a/mindspore/python/mindspore/ops/operations/nn_ops.py
+++ b/mindspore/python/mindspore/ops/operations/nn_ops.py
@@ -1395,16 +1395,15 @@ class Conv2D(Primitive):
 
 class DataFormatVecPermute(Primitive):
     r"""
-    Permute input tensor from src_format to dst_format.
+    Converts the input tensor from the `src_format` to the `dst_format` by permuting its dimensions.
 
     Args:
-        src_format (str, optional): An optional value for source data format. The format can be 'NHWC' and 'NCHW'.
-            Default: 'NHWC'.
-        dst_format (str, optional): An optional value for destination data format. The format can be 'NHWC' and 'NCHW'.
-            Default: 'NCHW'.
+        src_format (str, optional): the source data format, which can be 'NHWC' and 'NCHW'. Default: 'NHWC'.
+        dst_format (str, optional): the target data format, which can be 'NHWC' and 'NCHW'. Default: 'NCHW'.
 
     Inputs:
-        - **input_x** (Tensor) - A Tensor of shape (4, ) or (4, 2) in source data format. Only supports int32 and int64.
+        - **input_x** (Tensor) - A Tensor of shape :math:`(4, )` or :math:`(4, 2)` in source data format.
+          Supports int32 and int64 datatype.
 
     Outputs:
         Tensor, has the same data type and shape as the `input_x`.
@@ -1413,7 +1412,7 @@ class DataFormatVecPermute(Primitive):
         TypeError: If `input_x` is not a Tensor.
         TypeError: If dtype of `input_x` is neither int32 nor int64.
         ValueError: If `src_format` or `dst_format` is not a str in ['NHWC', 'NCHW'].
-        ValueError: If input_x shape is not (4, ) or (4, 2).
+        ValueError: If `input_x` shape is not :math:`(4, )` or :math:`(4, 2).
 
     Supported Platforms:
         ``Ascend`` ``GPU`` ``CPU``