phlower.nn.Dirichlet

class phlower.nn.Dirichlet(activation, dirichlet_name, nodes=None)

Bases: IPhlowerCoreModule, Module

Dirichlet is a neural network module that overwrites values with that of dirichlet field.

Parameters:

• activation (str) – Name of the activation function to apply to the output.

• dirichlet_name (str) – Name of the dirichlet field.

• nodes (list[int] | None (optional)) – List of feature dimension sizes (The last value of tensor shape). Defaults to None.

Examples

>>> dirichlet = Dirichlet(activation="relu", dirichlet_name="dirichlet")
>>> dirichlet(data)

Methods

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

Attributes

T_destination
call_super_init
dump_patches
training

forward(data, *, field_data=None, **kwards)

forward function which overloads torch.nn.Module

Parameters:

• data (IPhlowerTensorCollections) – IPhlowerTensorCollections data which receives from predecessors

• supports – dict[str, PhlowerTensor] | None Graph object. Defaults to None. Dirichlet will not use it.

• field_data (ISimulationField | None)

Returns:

Tensor object

Return type:

PhlowerTensor

classmethod from_setting(setting)

Create Dirichlet from setting object

Parameters:

setting (DirichletSetting) – setting object

Returns:

Dirichlet

Return type:

Self

classmethod get_nn_name()

Return name of Dirichlet

Returns:

name

Return type:

str