leaspy.models.utils.attributes

Submodules

• leaspy.models.utils.attributes.abstract_attributes
• leaspy.models.utils.attributes.abstract_manifold_model_attributes
• leaspy.models.utils.attributes.attributes_factory
• leaspy.models.utils.attributes.linear_attributes
• leaspy.models.utils.attributes.logistic_attributes
• leaspy.models.utils.attributes.logistic_parallel_attributes

Classes

LinearAttributes	Attributes of leaspy linear models.
LogisticAttributes	Attributes of leaspy logistic models.
LogisticParallelAttributes	Attributes of leaspy logistic parallel models.
AttributesFactory	Return an Attributes class object based on the given parameters.

Package Contents

class LinearAttributes(name, dimension, source_dimension)

Bases: leaspy.models.utils.attributes.abstract_manifold_model_attributes.AbstractManifoldModelAttributes

Attributes of leaspy linear models.

Contains the common attributes & methods to update the linear model’s attributes.

Parameters:

namestr

dimensionint

source_dimensionint

Attributes:

namestr (default ‘linear’)

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] (default {‘all’, ‘g’, ‘v0’, ‘betas’} )

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

positionstorch.Tensor [dimension] (default None)

positions = realizations[‘g’] such that “p0” = positions

velocitiestorch.Tensor [dimension] (default None)

Always positive: exp(realizations[‘v0’])

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

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

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

Matrix A such that w_i = A * s_i.

See also

UnivariateModel

MultivariateModel

update(names_of_changed_values, values)

Update model group average parameter(s).

Parameters:

names_of_changed_valuesset[str]

Elements of set must be either:

• all (update everything)

• g correspond to the attribute positions.

• v0 (only for multivariate models) correspond to the attribute velocities. When we are sure that the v0 change is only a scalar multiplication (in particular, when we reparametrize log(v0) <- log(v0) + mean(xi)), we may update velocities using v0_collinear, otherwise we always assume v0 is NOT collinear to previous value (no need to perform the verification it is - would not be really efficient)

• betas correspond to the linear combination of columns from the orthonormal basis so to derive the mixing_matrix.

valuesdict [str, torch.Tensor]

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

Raises:

LeaspyModelInputError

If names_of_changed_values contains unknown parameters.

class LogisticAttributes(name, dimension, source_dimension)

Bases: leaspy.models.utils.attributes.abstract_manifold_model_attributes.AbstractManifoldModelAttributes

Attributes of leaspy logistic models.

Contains the common attributes & methods to update the logistic model’s attributes.

Parameters:

namestr

dimensionint

source_dimensionint

Attributes:

namestr (default ‘logistic’)

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] (default {‘all’, ‘g’, ‘v0’, ‘betas’} )

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

positionstorch.Tensor [dimension] (default None)

positions = exp(realizations[‘g’]) such that “p0” = 1 / (1 + positions)

velocitiestorch.Tensor [dimension] (default None)

Always positive: exp(realizations[‘v0’])

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

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

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

Matrix A such that w_i = A * s_i.

See also

UnivariateModel

MultivariateModel

update(names_of_changed_values, values)

Update model group average parameter(s).

Parameters:

names_of_changed_valuesset[str]

Elements of set must be either:

• all (update everything)

• g correspond to the attribute positions.

• v0 (only for multivariate models) correspond to the attribute velocities. When we are sure that the v0 change is only a scalar multiplication (in particular, when we reparametrize log(v0) <- log(v0) + mean(xi)), we may update velocities using v0_collinear, otherwise we always assume v0 is NOT collinear to previous value (no need to perform the verification it is - would not be really efficient)

• betas correspond to the linear combination of columns from the orthonormal basis so to derive the mixing_matrix.

valuesdict [str, torch.Tensor]

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

Raises:

LeaspyModelInputError

If names_of_changed_values contains unknown parameters.

class LogisticParallelAttributes(name, dimension, source_dimension)

Bases: leaspy.models.utils.attributes.abstract_manifold_model_attributes.AbstractManifoldModelAttributes

Attributes of leaspy logistic parallel models.

Contains the common attributes & methods of the logistic parallel models’ attributes.

Parameters:

namestr

dimensionint

source_dimensionint

Attributes:

namestr (default ‘logistic_parallel’)

Name of the associated leaspy model.

dimensionint

source_dimensionint

has_sourcesbool

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

update_possibilitiesset[str] (default {‘all’, ‘g’, ‘deltas’, ‘betas’} )

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

positionstorch.Tensor (scalar) (default None)

positions = exp(realizations[‘g’]) such that “p0” = 1 / (1 + positions * exp(-deltas))

deltastorch.Tensor [dimension] (default None)

deltas = [0, delta_2_realization, …, delta_n_realization]

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

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

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

Matrix A such that w_i = A * s_i.

Raises:

LeaspyModelInputError

if any inconsistent parameters for the model.

See also

MultivariateParallelModel

deltas: torch.FloatTensor = None

update_possibilities

get_attributes()

Returns the following attributes: positions, deltas & mixing_matrix.

Returns:

positions: torch.Tensor

deltas: torch.Tensor

mixing_matrix: torch.Tensor

update(names_of_changed_values, values)

Update model group average parameter(s).

Parameters:

names_of_changed_valuesset[str]

Elements of set must be either:

• all (update everything)

• g correspond to the attribute positions.

• deltas correspond to the attribute deltas.

• betas correspond to the linear combination of columns from the orthonormal basis so to derive the mixing_matrix.

valuesdict [str, torch.Tensor]

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

Raises:

LeaspyModelInputError

If names_of_changed_values contains unknown parameters.

class AttributesFactory

Return an Attributes class object based on the given parameters.

classmethod attributes(name, dimension, source_dimension=None)

Class method to build correct model attributes depending on model name.

Parameters:

namestr

dimensionint

source_dimensionint, optional (default None)

Returns:

AbstractAttributes

Raises:

LeaspyModelInputError

if any inconsistent parameter.

Parameters:

• name (str)

• dimension (int)

• source_dimension (int)

Return type:

AbstractAttributes