sciai.architecture.AdaptActivation

查看源文件
class sciai.architecture.AdaptActivation(activation, a, scale)[源代码]

具有可训练参数和固定尺度的自适应激活函数。 自适应激活函数详情请查看: Adaptive activation functions accelerate convergence in deep and physics-informed neural networksLocally adaptive activationfunctions with slope recoveryfor deep and physics-informedneural network

参数:
  • activation (Union[str, Cell, Primitive, function]) - 激活函数。

  • a (Union[Number, Tensor, Parameter]) - 可训练参数 a

  • scale (Union[Number, Tensor]) - 固定比例参数。

输入:
  • x (Tensor) - AdaptActivation的输入。

输出:

Tensor,shape与 x 一致的被激活的输出。

异常:
  • TypeError - 如果输入类型不正确。

支持平台:

GPU CPU Ascend

样例:

>>> import mindspore as ms
>>> from mindspore import ops, nn
>>> from sciai.architecture import AdaptActivation
>>> a = ms.Tensor(0.1, ms.float32)
>>> net = AdaptActivation(nn.Tanh(), a=a, scale=10)
>>> x = ops.ones((2, 3), ms.float32)
>>> y = net(x)
>>> print(y)
[[0.7615942 0.7615942 0.7615942]
[0.7615942 0.7615942 0.7615942]]