mindspore.mint.nn.functional.softplus

mindspore.mint.nn.functional.softplus(input, beta=1, threshold=20)[source]

Applies softplus function to input element-wise.

The softplus function is shown as follows, x is the element of input :

\[\text{output} = \frac{1}{beta}\log(1 + \exp(\text{beta * x}))\]

where \(input * beta > threshold\), the implementation converts to the linear function to ensure numerical stability.

Parameters
  • input (Tensor) –

    Tensor of any dimension. Supported dtypes:

    • Ascend: float16, float32, bfloat16.

  • beta (number.Number, optional) – Scaling parameters in the softplus function. Default: 1 .

  • threshold (number.Number, optional) – For numerical stability, the softplus function is converted to a threshold parameter of a linear function. Default: 20 .

Returns

Tensor, with the same type and shape as the input.

Raises
  • TypeError – If input is not a Tensor.

  • TypeError – If dtype of input is not float16, float32, bfloat16.

Supported Platforms:

Ascend

Examples

>>> import mindspore
>>> import numpy as np
>>> from mindspore import Tensor, mint
>>> input = Tensor(np.array([0.1, 0.2, 30, 25]), mindspore.float32)
>>> output = mint.nn.functional.softplus(input)
>>> print(output)
[0.74439657 0.7981388 30. 25.]