mindspore.mint.nn.ELU

class mindspore.mint.nn.ELU(alpha=1.0, inplace=False)

Exponential Linear Unit activation function

Applies the exponential linear unit function element-wise.The activation function is defined as:

$$EL{U}_i=\{\begin{array}{ll}{x}_i,& \text{if }{x}_i\ge 0;\\ \alpha \ast (\exp (x_i)-1),& \text{otherwise.}\end{array}$$

where $x_i$ represents the element of the input and $\alpha$ represents the alpha parameter, and alpha represents the smoothness of the ELU.

ELU Activation Function Graph:

Warning

This is an experimental API that is subject to change or deletion.

Parameters

• alpha (float, optional) – The alpha value of ELU, the data type is float. Default: 1.0.

• inplace (bool, optional) – Whether to use inplace mode, the data type is bool. Default: False.

Inputs:

• input (Tensor) - The input of ELU is a Tensor of any dimension.

Outputs:

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

Raises

• RuntimeError – If the dtype of input is not float16, float32 or bfloat16.

• TypeError – If the dtype of alpha is not float.

Supported Platforms:

Ascend

Examples

>>> import mindspore
>>> from mindspore import Tensor, mint
>>> import numpy as np
>>> input = Tensor(np.array([-1, -2, 0, 2, 1]), mindspore.float32)
>>> elu = mint.nn.ELU()
>>> result = elu(input)
>>> print(result)
[-0.63212055  -0.86466473  0.  2.  1.]