mindspore.nn.Tanhshrink

class mindspore.nn.Tanhshrink[source]

Tanhshrink activation function.

The tanhshrink function is evaluated by element and returns a new tensor.

Tanh function is defined as:

\[tanhshrink(x_i) =x_i- \frac{\exp(x_i) - \exp(-x_i)}{\exp(x_i) + \exp(-x_i)} = x_i-\frac{\exp(2x_i) - 1}{\exp(2x_i) + 1},\]

where \(x_i\) is an element of the input Tensor.

Inputs:

x (Tensor) - Tensor of any dimension, input with data type of float16 or float32.

Outputs:

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

Raises

TypeError – If x is not a Tensor.
TypeError – If dtype of x is neither float16 nor float32.

Supported Platforms:: Ascend GPU CPU

Examples

>>> import mindspore as ms
>>> import mindspore.nn as nn
>>> from mindspore import Tensor
>>> import numpy as np
>>> x = Tensor(np.array([1, 2, 3, 2, 1]), ms.float16)
>>> tanhshrink = nn.Tanhshrink()
>>> output = tanhshrink(x)
>>> print(output)
[0.2383 1.036  2.004  1.036  0.2383]