SklearnModelAdapter#

class causalpy.experiments.model_adapter.SklearnModelAdapter[source]#

Adapter for sklearn RegressorMixin backends.

Parameters:: model (RegressorMixin) – CausalPy-compatible sklearn backend model.

Methods

`SklearnModelAdapter.build`(X, y, *[, coords])	Record the design matrices for `sample_posterior()`.
`SklearnModelAdapter.coefficients`(*[, group])	Return fitted sklearn coefficients as singleton posterior draws.
`SklearnModelAdapter.fit`(X, y, *[, coords])	Fit the sklearn model.
`SklearnModelAdapter.predict`(X, *[, coords, ...])	Return point predictions as singleton posterior draws.
`SklearnModelAdapter.print_coefficients`(labels)	Print model coefficients with labels.
`SklearnModelAdapter.require_idata`()	Return fitted inference-result DataTree or raise an explicit capability error.
`SklearnModelAdapter.sample_posterior`(**kwargs)	Fit the sklearn model on the recorded design matrices.
`SklearnModelAdapter.sample_prior_predictive`(...)	Raise: point-estimate backends have no prior predictive phase.
`SklearnModelAdapter.score`(X, y, *[, coords, ...])	Return per-output \(R^2\) scores from the sklearn model.

Attributes

`has_posterior`	Whether the sklearn model has been fitted.
`has_prior`	Whether prior draws are available on this backend.
`idata`	Return `None` because sklearn models have no inference-result DataTree.
`is_bayesian`	Whether the backend is Bayesian (PyMC or pymc-forecast).
`is_built`	Whether design matrices have been recorded via `build()`.
`is_ols`	Whether the backend is OLS/sklearn.
`kind`	Backend identifier.
`model`	The underlying sklearn model.
`supports_idata`	Whether the backend exposes an inference-result DataTree.
`supports_prior_predictive`	Whether this backend can sample a prior predictive phase.

__init__(model)[source]#