phlower.nn.LayerScaler¶

class phlower.nn.LayerScaler(scaling_method)[source]¶

Bases: IGenericPhlowerCoreModule[IPhlowerLayerParameters, PhlowerTensor], Module

LayerScaler is a neural network module that performs a scaling operation on the input tensor.

Parameters:: scaling_method (str) – The method of scaling to apply. Supported methods are: - “signed_log”: Applies the signed logarithm function to the input. - “asinh”: Applies the inverse hyperbolic sine function to the input.

Examples

>>> layer_scaler = LayerScaler(
...     scaling_method="signed_log"
... )
>>> layer_scaler(data)

Methods

`forward`(data, *[, field_data])	forward function which overloads torch.nn.Module
`from_setting`(setting)	Create LayerScaler from setting object
`get_nn_name`()	Return name of LayerScaler

Attributes

forward(data, *, field_data=None, **kwards)[source]¶

forward function which overloads torch.nn.Module

Parameters:

data (IPhlowerTensorCollections) – IPhlowerTensorCollections data which receives from predecessors
field_data (ISimulationField | None) – ISimulationField | None Constant information through training or prediction

Returns:

Tensor object

Return type:

PhlowerTensor

classmethod from_setting(setting)[source]¶

Create LayerScaler from setting object

Parameters:: setting (LayerScalerSetting) – LayerScalerSetting setting object
Returns:: LayerScaler
Return type:: Self

classmethod get_nn_name()[source]¶

Return name of LayerScaler