transformation¶
Transformations for models.
Submodules¶
|
Create a BatchEnsemble predictor from a base predictor based on [WTB20]. |
|
Create a Bayesian predictor from a base predictor based on [BCKW15]. |
|
Create an ensemble of Bayesian-layer transformed predictors based on [BCKW15]. |
Calibration transformations for logit predictors. |
|
|
Return a predictor unchanged while optionally registering its predictor type. |
|
Create an ensemble with class-specific final-layer bias initialization based on [LohrHMHullermeier25]. |
Conformal transformations for regression and classification. |
|
Conformalized Credal Set Prediction implementation. |
|
Append clipped exp plus one so logits become Dirichlet concentrations. |
|
|
Append an exp activation so logits become Dirichlet concentrations. |
Append Softplus plus one so logits become Dirichlet concentrations. |
|
|
Create a DropConnect predictor from a base predictor based on [MYM+21]. |
|
Create a Dropout predictor from a base predictor based on [GG16]. |
|
Create an ensemble predictor from a base predictor based on [LPB17]. |
|
Create an interval classifier from a base classifier based on [WCM+24]. |
Module for natural posterior network implementations. |
|
Replace the final linear layer with a Normal-Inverse-Gamma head. |
|
|
Create a Posterior Network predictor from an encoder based on [CZugnerGunnemann20]. |
|
Create a subensemble predictor from a base model or a base model and head model. |
Utilities for the definition of methods. |