mindspore.Tensor.nan_to_num

Tensor.nan_to_num(nan, posinf, neginf) Tensor

Replace the NaN, positive infinity and negative infinity values of the self with the specified values in nan, posinf and neginf respectively.

Warning

For Ascend, it is only supported on Atlas A2 Training Series Products. This is an experimental API that is subject to change or deletion.

Parameters
  • nan (number, optional) – The replace value of NaN. Default: None.

  • posinf (number, optional) – the value to replace positive infinity values with. Default: None, replacing positive infinity with the maximum value supported by the data type of self.

  • neginf (number, optional) – the value to replace negative infinity values with. Default: None, replacing negative infinity with the minimum value supported by the data type of self.

Returns

Tensor, has the same shape and dtype as self.

Supported Platforms:

Ascend CPU

Examples

>>> import mindspore
>>> import numpy as np
>>> from mindspore import Tensor
>>> input = Tensor(np.array([float('nan'), float('inf'), -float('inf'), 5.0]), mindspore.float32)
>>> output = input.nan_to_num(1.0, 2.0, 3.0)
>>> print(output)
[1.  2.  3.  5.0]