phlower.nn.Reducer

class phlower.nn.Reducer(activation, operator, nodes=None)

Bases: IPhlowerCoreModule, Module

Reducer is a neural network module that performs a reduction operation on the input tensor.

Parameters:

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

• operator (str) – Name of the operator to apply to the input tensors. “add” or “mul”. Default is “add”.

• nodes (list[int]) – List of feature dimension sizes (The last value of tensor shape).

Examples

>>> reducer = Reducer(activation="relu", operator="add", nodes=[10, 20, 30])
>>> reducer(data)

Methods

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

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

• field_data (ISimulationField | None) – ISimulationField | None Constant information through training or prediction

Returns:

Tensor object

Return type:

PhlowerTensor

classmethod from_setting(setting)

Create Reducer from setting object

Parameters:

setting (ReducerSetting) – ReducerSetting setting object

Returns:

Reducer

Return type:

Self

classmethod get_nn_name()

Return name of Reducer

Returns:

name

Return type:

str