method

Uncertainty-aware methods.

Submodules

batchensemble(base[, num_members, ...])	Create a BatchEnsemble predictor from a base predictor based on [WTB20].
bayesian(base[, use_base_weights, ...])	Create a Bayesian predictor from a base predictor based on [BCKW15].
calibration	Calibration method compatibility exports.
cast(base)	Return a predictor unchanged while optionally registering its predictor type.
conformal	Conformal method compatibility exports.
conformal_credal_set	Conformalized Credal Set Prediction implementation.
credal_bnn(base[, use_base_weights, ...])	Create a Credal BNN predictor from a base predictor based on [CDJ+24].
credal_ensembling(base, num_members[, ...])	Create a credal ensembling predictor from a base predictor based on [NZD25].
credal_net	Credal net method compatibility layer.
credal_relative_likelihood(base, num_members)	Create a Credal Relative Likelihood predictor from a base predictor.
credal_wrapper(base, num_members[, reset_params])	Create a credal wrapper predictor from a base predictor based on [WCS+25].
dare(base, num_members[, reset_params])	Create a DARE predictor from a base predictor based on [dMDMV23].
ddu(base[, sn_coeff])	Apply spectral normalization and add a Gaussian-mixture density head.
deup	Direct Epistemic Uncertainty Prediction (DEUP) method.
dropconnect(base[, p, rng_collection, rngs])	Create a DropConnect predictor from a base predictor based on [MYM+21].
dropout(base[, p, rng_collection, rngs, ...])	Create a Dropout predictor from a base predictor based on [GG16].
duq(base[, centroid_size, length_scale, gamma])	Replace the final classifier head with an RBF centroid head.
efficient_credal_prediction(base)	Create an efficient credal predictor from a base predictor based on [HLohrM+26].
ensemble(base, num_members[, reset_params])	Create an ensemble predictor from a base predictor based on [LPB17].
evidential	Evidential methods for probly.
graph_posterior_network(input_encoder, ...)	Create a Graph Posterior Network predictor.
het_net(base[, num_factors, temperature, ...])	Create a HetNets predictor from a base predictor base on [CMK+21].
laplace	Laplace approximation method.
mahalanobis(base[, feature_nodes, ...])	Turn a classifier into a Mahalanobis-distance OOD detector.
natural_posterior_network	Natural posterior network method compatibility exports.
posterior_network	Posterior network method compatibility exports.
prior_network	Prior network method compatibility layer.
sngp(base[, name, n_power_iterations, ...])	Wrap a model with SNGP (Spectral-normalized Neural Gaussian Process).
subensemble(base, num_heads[, head, ...])	Create a subensemble predictor from a base model or a base model and head model.