Skip to content

Base

jimgw.core.base ¤

Data ¤

Bases: ABC

Abstract base class for data containers.

__init__() abstractmethod ¤
fetch() -> None abstractmethod ¤

Fetch or load the data into the container.

Model ¤

Bases: Module

Abstract base class for probabilistic models.

params: dict instance-attribute ¤
__init__() abstractmethod ¤
__call__(x: Array) -> Float ¤

Evaluate the model at x.

Parameters:

Name Type Description Default
x Array

Input array.

 required

Returns:

Name Type Description
Float Float

Model output (e.g. log-probability).

LikelihoodBase ¤

Bases: ABC

Abstract base class for likelihoods.

Handles two main components: the data and the model. Subclasses must implement :meth:evaluate.

_model: object instance-attribute ¤
_data: object instance-attribute ¤
model: object property ¤

The model used by the likelihood.

data: object property ¤

The data used by the likelihood.

evaluate(params: dict[str, Float], data: dict) -> Float abstractmethod ¤

Evaluate the log-likelihood for a given set of parameters.

Parameters:

Name Type Description Default
params dict[str, Float]

Dictionary mapping parameter names to values.

 required
data dict

Auxiliary data (typically empty for pre-loaded data).

 required

Returns:

Name Type Description
Float Float

Log-likelihood value.