Learners

Use these objects to configure nuisance models, cross-fitting folds, conditional densities, and thread limits. Learners follow the relevant scikit-learn fit and predict contracts.

cleverly.SuperLearner	Cross-validated convex ensemble of a library of learners.
cleverly.learners.default_library	Return the concrete default ensemble for one prediction task.
cleverly.learners.SuperLearnerDiagnostics	What the ensemble learned, for reporting and validation.
cleverly.learners.CorrelationScreener	scikit-learn selector wrapping screen_by_correlation().
cleverly.learners.ParametricOutcome	An unpenalized logistic GLM of the outcome on a declared feature map.
cleverly.learners.ParametricMechanism	An unpenalized logistic regression of an indicator on a polynomial in logit q.
cleverly.learners.ParametricFitDiagnostics	What one parametric fit reports.
cleverly.learners.CrossFitPlan	The fold policy a fit declared, as distinct from the split it got.
cleverly.learners.Folds	A cross-fitting partition.
cleverly.learners.FoldOrigin	The record of how random_partition() drew one split.
cleverly.learners.ConditionalDensity	An evaluated conditional density: one row of bin probabilities per observation.
cleverly.learners.DensityDiagnostics	What the density fit looked like, for the report and for the positivity check.
cleverly.learners.make_folds	Build a cross-fitting partition of n observations.
cleverly.learners.random_partition	Draw an unstratified split of rows, or of whole clusters, from a seed.
cleverly.learners.thread_limit	Limit native thread pools for the duration of the block.
cleverly.learners.set_thread_limit	Set the native thread limit applied around each nuisance-model fit.
cleverly.learners.get_thread_limit	The thread limit currently applied around each fit.