phlower.nn.DeepSets

class phlower.nn.DeepSets(lambda_config, gamma_config, last_activation_name, pool_operator_name, unbatch_key=None)

Bases: IPhlowerCoreModule, Module

DeepSets is a neural network module that performs permutation invariant / equivariant operation on the input tensor.

Ref: https://arxiv.org/abs/1703.06114

Parameters:

• lambda_config (MLPConfiguration) – Configuration for the lambda network.

• gamma_config (MLPConfiguration) – Configuration for the gamma network.

• last_activation_name (str) – Name of the last activation function.

• pool_operator_name (str) – Name of the pooling operator. “max” or “mean”. Default is “max”.

• unbatch_key (str | None) – Key of the unbatch operation.

Methods

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

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 DeepSets from setting object

Parameters:

setting (DeepSetsSetting) – setting object

Returns:

DeepSets

Return type:

Self

classmethod get_nn_name()

Return name of DeepSets

Returns:

name

Return type:

str