leaspy.models.utils.attributes.abstract_manifold_model_attributes module

class AbstractManifoldModelAttributes(name: str, dimension: int, source_dimension: int)

Bases: AbstractAttributes

Abstract base class for attributes of leaspy manifold models.

Contains the common attributes & methods of the different attributes classes. Such classes are used to update the models’ attributes.

Parameters:

namestr

dimensionint

source_dimensionint (default None)

Raises:

LeaspyModelInputError

if any inconsistent parameter.

Attributes:

namestr (default None)

Name of the associated leaspy model.

dimensionint

source_dimensionint

univariatebool

Whether model is univariate or not (i.e. dimension == 1)

has_sourcesbool

Whether model has sources or not (not univariate and source_dimension >= 1)

update_possibilitiesset[str]

Contains the available parameters to update. Different models have different parameters.

positionstorch.Tensor [dimension] (default None)

<!> Depending on the model it does not correspond to the same thing.

velocitiestorch.Tensor [dimension] (default None)

Vector of velocities for each feature (positive components). For multivariate models only (except for parallel model as it is useless).

orthonormal_basistorch.Tensor [dimension, dimension - 1] (default None)

For multivariate and multivariate parallel models, with source_dimension >= 1.

betastorch.Tensor [dimension - 1, source_dimension] (default None)

For multivariate and multivariate parallel models, with source_dimension >= 1.

mixing_matrixtorch.Tensor [dimension, source_dimension] (default None)

Matrix A such that w_i = A * s_i. For multivariate and multivariate parallel models, with source_dimension >= 1.

Methods

get_attributes()	Returns the attributes of the model, which is a tuple of three torch tensors.
move_to_device(device)	Move the tensor attributes of this class to the specified device.
update(names_of_changed_values, values)	Update model group average parameter(s).

get_attributes() → Tuple[FloatTensor, FloatTensor, FloatTensor]

Returns the attributes of the model, which is a tuple of three torch tensors.

For the precise definitions of those attributes please refer to the exact attributes class associated to your model.

Returns:

positions: torch.Tensor

velocities: torch.Tensor

mixing_matrix: torch.Tensor

move_to_device(device: device)

Move the tensor attributes of this class to the specified device.

Parameters:

devicetorch.device

abstract update(names_of_changed_values: Set[str], values: Dict[str, Tensor]) → None

Update model group average parameter(s).

Parameters:

names_of_changed_valuesset[str]

Values to be updated

valuesdict [str, torch.Tensor]

New values used to update the model’s group average parameters

Raises:

LeaspyModelInputError

If names_of_changed_values contains unknown values to update.