mindspore.ops.nansum

View Source On Gitee
mindspore.ops.nansum(input, axis=None, keepdims=False, *, dtype=None)[source]

Computes sum of input over a given dimension, ignoring NaN.

Parameters

  • input (Tensor) – The input tensor.

  • axis (Union[int, tuple(int)], optional) – The dimensions to reduce. Supposed the rank of input is r, axis must be in the range [-rank(input), rank(input)). Default None, all dimensions are reduced.

  • keepdims (bool, optional) – Whether the output tensor keeps has dim retained. Default False.

Keyword Arguments

dtype (mindspore.dtype, optional) – The dtype of output tensor. Default None.

Returns

Tensor

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> x = mindspore.tensor([[float("nan"), 2, 3], [1, float("nan"), 3], [1, 2, float("nan")]], mindspore.float32)
>>> # case1: axis is None, keepdims is False,
>>> output1 = mindspore.ops.nansum(x, axis=None, dtype=mindspore.float32)
>>> print(output1)
12.0
>>> # case2: axis is int, set as 0, and keepdims is False
>>> output2 = mindspore.ops.nansum(x, axis=0, dtype=mindspore.float32)
>>> print(output2)
[2. 4. 6.]
>>> # case3: axis is int, set as 0, and keepdims is False
>>> output3 = mindspore.ops.nansum(x, axis=0, keepdims=True, dtype=mindspore.float32)
>>> print(output3)
[[2. 4. 6.]]
>>> # case4: axis is tuple(int) or list(int), set as (0, 1), and keepdims is False
>>> output4 = mindspore.ops.nansum(x, axis=(0, 1), dtype=mindspore.float32)
>>> print(output4)
12.0