leaspy.models.obs_models

Attributes

`OBSERVATION_MODELS`
`ObservationModelFactoryInput`

Classes

`ObservationModel`	Base class for valid observation models that may be used in probabilistic models (stateless).
`BernoulliObservationModel`	Base class for valid observation models that may be used in probabilistic models (stateless).
`ObservationModelNames`	Enumeration defining the possible names for observation models.
`FullGaussianObservationModel`	Specialized GaussianObservationModel when all data share the same observation model, with default naming.
`GaussianObservationModel`	Specialized ObservationModel for noisy observations with Gaussian residuals assumption.
`AbstractWeibullRightCensoredObservationModel`	Base class for valid observation models that may be used in probabilistic models (stateless).
`WeibullRightCensoredObservationModel`	Base class for valid observation models that may be used in probabilistic models (stateless).
`WeibullRightCensoredWithSourcesObservationModel`	Base class for valid observation models that may be used in probabilistic models (stateless).

Functions

observation_model_factory(model, **kwargs)

Factory for observation models.

Package Contents

class ObservationModel

Base class for valid observation models that may be used in probabilistic models (stateless).

In particular, it provides data & linked variables regarding observations and their attachment to the model (the negative log-likelihood - nll - to be minimized).

Parameters:

namestr: The name of observed variable (to name the data variable & attachment term related to this observation).
getterfunction Dataset -> WeightedTensor: The way to retrieve the observed values from the Dataset (as a WeightedTensor): e.g. all values, subset of values - only x, y, z features, one-hot encoded features, …
distSymbolicDistribution: The symbolic distribution, parametrized by model variables, for observed values (so to compute attachment).
extra_varsNone (default) or Mapping[VarName, VariableInterface]: Some new variables that are needed to fully define the symbolic distribution or the sufficient statistics. (e.g. “noise_std”, and “y_L2_per_ft” for instance for a Gaussian model)

name: leaspy.variables.specs.VariableName

getter: Callable[[Dataset], WeightedTensor]

dist: SymbolicDistribution

extra_vars: Mapping[leaspy.variables.specs.VariableName, VariableInterface] | None = None

get_nll_attach_var_name(named_attach_vars=True)

Return the name of the negative log likelihood attachement variable.

Parameters:: named_attach_vars (bool)
Return type:: str

get_variables_specs(named_attach_vars=True)

Automatic specifications of variables for this observation model.

Parameters:: named_attach_vars (bool)
Return type:: dict[leaspy.variables.specs.VariableName, VariableInterface]

serialized()

Nice representation of instance without its name (should be JSON exportable).

Return type:: Any

to_dict()

To be implemented…

Return type:: dict

to_string()

method for parameter saving

Return type:: str

class BernoulliObservationModel(**extra_vars)

Bases: leaspy.models.obs_models._base.ObservationModel

Base class for valid observation models that may be used in probabilistic models (stateless).

In particular, it provides data & linked variables regarding observations and their attachment to the model (the negative log-likelihood - nll - to be minimized).

Parameters:

namestr: The name of observed variable (to name the data variable & attachment term related to this observation).
getterfunction Dataset -> WeightedTensor: The way to retrieve the observed values from the Dataset (as a WeightedTensor): e.g. all values, subset of values - only x, y, z features, one-hot encoded features, …
distSymbolicDistribution: The symbolic distribution, parametrized by model variables, for observed values (so to compute attachment).
extra_varsNone (default) or Mapping[VarName, VariableInterface]: Some new variables that are needed to fully define the symbolic distribution or the sufficient statistics. (e.g. “noise_std”, and “y_L2_per_ft” for instance for a Gaussian model)

Parameters:

extra_vars (VariableInterface)

string_for_json = 'bernoulli'

static y_getter(dataset)

Parameters:: dataset (Dataset)
Return type:: WeightedTensor

OBSERVATION_MODELS: Dict[ObservationModelNames, Type[leaspy.models.obs_models._base.ObservationModel]]

ObservationModelFactoryInput

class ObservationModelNames(*args, **kwds)

Bases: enum.Enum

Enumeration defining the possible names for observation models.

GAUSSIAN_DIAGONAL = 'gaussian-diagonal'

GAUSSIAN_SCALAR = 'gaussian-scalar'

BERNOULLI = 'bernoulli'

WEIBULL_RIGHT_CENSORED = 'weibull-right-censored'

WEIBULL_RIGHT_CENSORED_WITH_SOURCES = 'weibull-right-censored-with-sources'

classmethod from_string(model_name)

Parameters:: model_name (str)

observation_model_factory(model, **kwargs)

Factory for observation models.

Parameters:

modelstr or ObservationModel or dict [ str, …]

If an instance of a subclass of ObservationModel, returns the instance.
If a string, then returns a new instance of the appropriate class (with optional parameters kws).
If a dictionary, it must contain the ‘name’ key and other initialization parameters.

**kwargs

Optional parameters for initializing the requested observation model when a string.

Returns:

ObservationModel: The desired observation model.

Raises:

LeaspyModelInputError: If model is not supported.

Parameters:

model (ObservationModelFactoryInput)

Return type:

leaspy.models.obs_models._base.ObservationModel

class FullGaussianObservationModel(noise_std, **extra_vars)

Bases: GaussianObservationModel

Specialized GaussianObservationModel when all data share the same observation model, with default naming.

The default naming is:

‘y’ for observations
‘model’ for model predictions
‘noise_std’ for scale of residuals

We also provide a convenient factory default for most common case, which corresponds to noise_std directly being a ModelParameter (it could also be a PopulationLatentVariable with positive support). Whether scale of residuals is scalar or diagonal depends on the dimension argument of this method.

Parameters:

noise_std (VariableInterface)
extra_vars (VariableInterface)

tol_noise_variance = 1e-05

static y_getter(dataset)

Parameters:: dataset (Dataset)
Return type:: WeightedTensor

classmethod noise_std_suff_stats()

Dictionary of sufficient statistics needed for noise_std (when directly a model parameter).

Return type:: dict[leaspy.variables.specs.VariableName, LinkedVariable]

classmethod scalar_noise_std_update(*, state, y_x_model, model_x_model)

Update rule for scalar noise_std (when directly a model parameter), from state & sufficient statistics.

Parameters:

state (State)
y_x_model (WeightedTensor[float])
model_x_model (WeightedTensor[float])

Return type:

Tensor

classmethod diagonal_noise_std_update(*, state, y_x_model, model_x_model)

Update rule for feature-wise noise_std (when directly a model parameter), from state & sufficient statistics.

Parameters:

state (State)
y_x_model (WeightedTensor[float])
model_x_model (WeightedTensor[float])

Return type:

Tensor

classmethod noise_std_specs(dimension)

Default specifications of noise_std variable when directly modelled as a parameter (no latent population variable).

Parameters:: dimension (int)
Return type:: ModelParameter

classmethod with_noise_std_as_model_parameter(dimension)

Default instance of FullGaussianObservationModel with noise_std (scalar or diagonal depending on dimension) being a ModelParameter.

Parameters:: dimension (int)

classmethod compute_rmse(*, y, model)

Compute root mean square error.

Parameters:

y (WeightedTensor[float])
model (WeightedTensor[float])

Return type:

Tensor

classmethod compute_rmse_per_ft(*, y, model)

Compute root mean square error, per feature.

Parameters:

y (WeightedTensor[float])
model (WeightedTensor[float])

Return type:

Tensor

to_string()

method for parameter saving

Return type:: str

class GaussianObservationModel(name, getter, loc, scale, **extra_vars)

Bases: leaspy.models.obs_models._base.ObservationModel

Specialized ObservationModel for noisy observations with Gaussian residuals assumption.

Parameters:

name (leaspy.variables.specs.VariableName)
getter (Callable[[Dataset], WeightedTensor])
loc (leaspy.variables.specs.VariableName)
scale (leaspy.variables.specs.VariableName)
extra_vars (VariableInterface)

class AbstractWeibullRightCensoredObservationModel

Bases: leaspy.models.obs_models._base.ObservationModel

Base class for valid observation models that may be used in probabilistic models (stateless).

In particular, it provides data & linked variables regarding observations and their attachment to the model (the negative log-likelihood - nll - to be minimized).

Parameters:

namestr: The name of observed variable (to name the data variable & attachment term related to this observation).
getterfunction Dataset -> WeightedTensor: The way to retrieve the observed values from the Dataset (as a WeightedTensor): e.g. all values, subset of values - only x, y, z features, one-hot encoded features, …
distSymbolicDistribution: The symbolic distribution, parametrized by model variables, for observed values (so to compute attachment).
extra_varsNone (default) or Mapping[VarName, VariableInterface]: Some new variables that are needed to fully define the symbolic distribution or the sufficient statistics. (e.g. “noise_std”, and “y_L2_per_ft” for instance for a Gaussian model)

static getter(dataset)

Parameters:: dataset (Dataset)
Return type:: WeightedTensor

get_variables_specs(named_attach_vars=True)

Automatic specifications of variables for this observation model.

Parameters:: named_attach_vars (bool)
Return type:: dict[leaspy.variables.specs.VariableName, VariableInterface]

class WeibullRightCensoredObservationModel(nu, rho, xi, tau, **extra_vars)

Bases: AbstractWeibullRightCensoredObservationModel

Base class for valid observation models that may be used in probabilistic models (stateless).

In particular, it provides data & linked variables regarding observations and their attachment to the model (the negative log-likelihood - nll - to be minimized).

Parameters:

namestr: The name of observed variable (to name the data variable & attachment term related to this observation).
getterfunction Dataset -> WeightedTensor: The way to retrieve the observed values from the Dataset (as a WeightedTensor): e.g. all values, subset of values - only x, y, z features, one-hot encoded features, …
distSymbolicDistribution: The symbolic distribution, parametrized by model variables, for observed values (so to compute attachment).
extra_varsNone (default) or Mapping[VarName, VariableInterface]: Some new variables that are needed to fully define the symbolic distribution or the sufficient statistics. (e.g. “noise_std”, and “y_L2_per_ft” for instance for a Gaussian model)

Parameters:

nu (leaspy.variables.specs.VariableName)
rho (leaspy.variables.specs.VariableName)
xi (leaspy.variables.specs.VariableName)
tau (leaspy.variables.specs.VariableName)
extra_vars (VariableInterface)

string_for_json = 'weibull-right-censored'

classmethod default_init(**kwargs)

class WeibullRightCensoredWithSourcesObservationModel(nu, rho, xi, tau, survival_shifts, **extra_vars)

Bases: AbstractWeibullRightCensoredObservationModel

Base class for valid observation models that may be used in probabilistic models (stateless).

In particular, it provides data & linked variables regarding observations and their attachment to the model (the negative log-likelihood - nll - to be minimized).

Parameters:

namestr: The name of observed variable (to name the data variable & attachment term related to this observation).
getterfunction Dataset -> WeightedTensor: The way to retrieve the observed values from the Dataset (as a WeightedTensor): e.g. all values, subset of values - only x, y, z features, one-hot encoded features, …
distSymbolicDistribution: The symbolic distribution, parametrized by model variables, for observed values (so to compute attachment).
extra_varsNone (default) or Mapping[VarName, VariableInterface]: Some new variables that are needed to fully define the symbolic distribution or the sufficient statistics. (e.g. “noise_std”, and “y_L2_per_ft” for instance for a Gaussian model)

Parameters:

nu (leaspy.variables.specs.VariableName)
rho (leaspy.variables.specs.VariableName)
xi (leaspy.variables.specs.VariableName)
tau (leaspy.variables.specs.VariableName)
survival_shifts (leaspy.variables.specs.VariableName)
extra_vars (VariableInterface)

string_for_json = 'weibull-right-censored-with-sources'

classmethod default_init(**kwargs)