fix tensor_scatter_mul doc

docs/api/api_python/mindspore.ops.rst

@@ -459,6 +459,7 @@ Array操作
     mindspore.ops.TensorScatterDiv
     mindspore.ops.TensorScatterMax
     mindspore.ops.TensorScatterMin
+    mindspore.ops.TensorScatterMul
     mindspore.ops.TensorScatterSub
     mindspore.ops.TensorScatterUpdate
     mindspore.ops.TensorScatterElements

docs/api/api_python/mindspore/mindspore.Tensor.rst

@@ -1193,11 +1193,12 @@ mindspore.Tensor
         - **ValueError** - Tensor的shape长度小于 `indices` 的shape的最后一个维度。
 
     .. py:method:: scatter_mul(indices, updates)
+        根据指定的索引，通过乘法进行计算，将结果赋值到输出Tensor中。更新后的结果是通过算子output返回，而不是直接原地更新当前Tensor。
 
-        根据指定的索引，通过乘法进行计算，将输出赋值到输出Tensor中。
+        `indices` 的最后一个轴是每个索引向量的深度。对于每个索引向量， `updates` 中必须有相应的值。 `updates` 的shape应该等于 `input_x[indices]` 的shape。其中 `input_x` 指当前Tensor。 有关更多详细信息，请参见使用用例。
 
         .. note::
-            - 如果 `indices` 的某些值超出当前Tensor的维度范围，在 `CPU` 后端会抛出错误，在 `GPU` 后端则忽略错误且更新值不可信任。
+            - 如果 `indices` 的某些值超出范围，则相应的 `updates` 不会更新为当前Tensor，而不是抛出索引错误。
 
         **参数：**
 

docs/api/api_python/ops/mindspore.ops.TensorScatterMul.rst

@@ -0,0 +1,8 @@
+mindspore.ops.TensorScatterMul
+==============================
+
+.. py:class:: mindspore.ops.TensorScatterMul
+
+    根据指定的更新值和输入索引，进行乘法运算更新输入Tensor的值。当同一索引有不同更新值时，更新的结果将是累积乘法的结果。此操作与 :class:`mindspore.ops.ScatterNdMul` 类似，只是更新后的结果是通过算子output返回，而不是直接原地更新input。
+
+    更多参考相见 :func:`mindspore.ops.tensor_scatter_mul`。

docs/api/api_python/ops/mindspore.ops.func_tensor_scatter_mul.rst

@@ -8,7 +8,7 @@ mindspore.ops.tensor_scatter_mul
     `indices` 的最后一个轴是每个索引向量的深度。对于每个索引向量， `updates` 中必须有相应的值。 `updates` 的shape应该等于 `input_x[indices]` 的shape。有关更多详细信息，请参见使用用例。
 
     .. note::
-        - 如果 `indices` 的某些值超出 `input_x` 的维度范围，在 `CPU` 后端会抛出错误，在 `GPU` 后端则忽略错误且更新值不可信任。
+        - 如果 `indices` 的某些值超出 `input_x` 的维度范围，则相应的 `updates` 不会更新为 `input_x` ，而不是抛出索引错误。
 
     **参数：**
 

docs/api/api_python_en/mindspore.ops.rst

@@ -458,6 +458,7 @@ Array Operation
     mindspore.ops.TensorScatterDiv
     mindspore.ops.TensorScatterMax
     mindspore.ops.TensorScatterMin
+    mindspore.ops.TensorScatterMul
     mindspore.ops.TensorScatterSub
     mindspore.ops.TensorScatterUpdate
     mindspore.ops.TensorShape

mindspore/python/mindspore/common/tensor.py

@@ -739,15 +739,13 @@ class Tensor(Tensor_):
         equal to the shape of `input_x[indices]`. For more details, see use cases.
 
         Note:
-            - If some values of the `indices` are out of bound, CPU backend will raise an index error.
-              GPU backend will not raise and index error
-              and the corresponding `updates` will not be updated to self tensor.
+            - If some values of the `indices` are out of bound, instead of raising an index error,
+              the corresponding `updates` will not be updated to `input_x`.
 
         Args:
-            indices (Tensor): The index of input tensor whose data type is int32 or int64.
-                The rank must be at least 2.
+            indices (Tensor): The index of input tensor whose data type is int32 or int64. The rank must be at least 2.
             updates (Tensor): The tensor to update the input tensor, has the same type as input,
-                and updates.shape should be equal to indices.shape[:-1] + input_x.shape[indices.shape[-1]:].
+                and updates shape should be equal to indices.shape[:-1] + input_x.shape[indices.shape[-1]:].
 
         Returns:
             Tensor, has the same shape and type as self tensor.

mindspore/python/mindspore/ops/function/array_func.py

@@ -3196,7 +3196,7 @@ def tensor_scatter_mul(input_x, indices, updates):
     """
     Creates a new tensor by multiplying the values from the positions in `input_x` indicated by
     `indices`, with values from `updates`. When divided values are provided for the same
-    index, the result of the update will be to divided these values respectively. Except that
+    index, the result of the update will multiply these values respectively. Except that
     the updates are applied on output `Tensor` instead of input `Parameter`.
 
     The last axis of `indices` is the depth of each index vectors. For each index vector,
@@ -3204,16 +3204,14 @@ def tensor_scatter_mul(input_x, indices, updates):
     equal to the shape of `input_x[indices]`. For more details, see use cases.
 
     Note:
-        - If some values of the `indices` are out of bound, CPU backend will raise an index error.
-          GPU backend will not raise and index error
-          and the corresponding `updates` will not be updated to `input_x`.
+        - If some values of the `indices` are out of bound, instead of raising an index error,
+          the corresponding `updates` will not be updated to `input_x`.
 
     Args:
         input_x (Tensor): The target tensor. The dimension of input_x must be no less than indices.shape[-1].
-        indices (Tensor): The index of input tensor whose data type is int32 or int64.
-            The rank must be at least 2.
+        indices (Tensor): The index of input tensor whose data type is int32 or int64. The rank must be at least 2.
         updates (Tensor): The tensor to update the input tensor, has the same type as input,
-            and updates.shape should be equal to indices.shape[:-1] + input_x.shape[indices.shape[-1]:].
+            and updates shape should be equal to indices.shape[:-1] + input_x.shape[indices.shape[-1]:].
 
     Returns:
         Tensor, has the same shape and type as `input_x`.