mindspore.ops.sum

mindspore.ops.sum(input, dim=None, keepdim=False, *, dtype=None)[source]

Calculate sum of tensor elements over a given dim.

Note

The dim with tensor type is only used for compatibility with older versions and is not recommended.

Parameters

input (Tensor) – The input tensor.
dim (Union[None, int, tuple(int), list(int), Tensor]) – Dimensions along which the sum is calculated. Default None .
keepdim (bool) – Whether the output tensor has dim retained or not. If True , keep these reduced dimensions and the length is 1. If False , will not keep these dimensions. Default False .

Note

If dim is None , sum is calculated on all the elements of the input tensor.
If dim is a tuple or list of ints or tensor, sum is calculated on all dimensions specified in dim .

Keyword Arguments: dtype (mindspore.dtype, optional) – The data type returned.
Returns: Tensor

Supported Platforms:: Ascend GPU CPU

Examples

>>> import mindspore
>>> x = mindspore.tensor([[[1, 1, 1, 1, 1, 1], [2, 2, 2, 2, 2, 2], [3, 3, 3, 3, 3, 3]],
...                      [[4, 4, 4, 4, 4, 4], [5, 5, 5, 5, 5, 5], [6, 6, 6, 6, 6, 6]],
...                      [[7, 7, 7, 7, 7, 7], [8, 8, 8, 8, 8, 8], [9, 9, 9, 9, 9, 9]]], mindspore.float32)
>>> out = mindspore.ops.sum(input=x)
>>> print(out)
270.0
>>> out = mindspore.ops.sum(input=x, dim=1)
>>> print(out)
[[ 6.  6.  6.  6.  6.  6.]
 [15. 15. 15. 15. 15. 15.]
 [24. 24. 24. 24. 24. 24.]]
>>> out = mindspore.ops.sum(input=x, dim=2)
>>> print(out)
[[ 6. 12. 18.]
 [24. 30. 36.]
 [42. 48. 54.]]
>>> out = mindspore.ops.sum(input=x, dim=[1, 2])
>>> print(out)
[ 36.  90. 144.]
>>> out = mindspore.ops.sum(input=x, dim=2, keepdim=True)
>>> print(out)
[[[ 6.]
 [12.]
 [18.]]
[[24.]
 [30.]
 [36.]]
[[42.]
 [48.]
 [54.]]]
>>> print(out.ndim)
3