GWForge.population_fisher.base

Base class for population models, and how their scores get computed.

Attributes

SCORE_METHODS
FD_RELATIVE_STEP
FD_ABSOLUTE_STEP

Classes

PopulationModel	Base class: a normalised density over event parameters, given hyper-parameters.
JointPopulationModel	Product of independent sub-models over disjoint event parameters.

Module Contents

GWForge.population_fisher.base.SCORE_METHODS = ('auto', 'analytic', 'finite-difference')

GWForge.population_fisher.base.FD_RELATIVE_STEP = 0.0001

GWForge.population_fisher.base.FD_ABSOLUTE_STEP = 1e-08

class GWForge.population_fisher.base.PopulationModel

Base class: a normalised density over event parameters, given hyper-parameters.

Variables:

• parameter_names (list of str) – Hyper-parameter names, in the order the score columns come out.

• fiducial (dict) – Default hyper-parameter values.

• event_keys (list of str) – Keys the model requires in the events dict.

parameter_names = []

fiducial

event_keys = []

abstractmethod log_prob(events, parameters)

Log density of each event.

Parameters:

• events (dict) – Arrays of event parameters, all the same length.

• parameters (dict) – Hyper-parameter values; must cover parameter_names.

Returns:

Shape (n_events,). -inf outside the support.

Return type:

numpy.ndarray

analytic_score(events, parameters)

Closed-form score, or None if this model has none.

Returns:

Shape (n_events, len(parameter_names)), columns in parameter_names order.

Return type:

numpy.ndarray or None

score(events, parameters=None, method='auto')

Score of each event with respect to every hyper-parameter.

Parameters:

• events (dict)

• parameters (dict or None) – Defaults to fiducial.

• method (str) – One of SCORE_METHODS. "auto" tries analytic, then falls back to finite differences.

Returns:

Shape (n_events, len(parameter_names)).

Return type:

numpy.ndarray

to_vector(parameters)

Hyper-parameter dict to array, in parameter_names order.

from_vector(vector, parameters=None)

Array back to a hyper-parameter dict, merged onto parameters.

check_events(events)

Raise if events is missing a key this model needs.

class GWForge.population_fisher.base.JointPopulationModel(models)

Bases: PopulationModel

Product of independent sub-models over disjoint event parameters.

models

parameter_names = []

event_keys = []

property fiducial

Sub-model fiducials, merged live.

log_prob(events, parameters)

Sum of the sub-model log densities.

analytic_score(events, parameters)

Horizontally stacked sub-model scores, None if any sub-model lacks one.

score(events, parameters=None, method='auto')

Per-sub-model dispatch, so one analytic block is not lost to another’s fallback.