Trained model with derivative information
Type of trained models with derivatives
Type of trained models for general hyper parameters
val calc : Model.t -> targets:Lacaml.D.vec -> tcalc model ~targets
trained model with derivative information given the untrained model and targets.
val calc_eval : t -> Eval.Trained.tcalc_eval trained
trained model without derivative information given trained.
val calc_log_evidence_sigma2 : t -> floatcalc_log_evidence_sigma2 trained
the derivative of the log evidence for the trained model with respect to the noise level (sigma2). This includes the contribution to the derivative by model.
prepare_hyper trained
the trained model prepared for calculating derivatives for arbitrary hyper parameters.
val calc_log_evidence : hyper_t -> Spec.Hyper.t -> floatcalc_log_evidence hyper_t hyper
the derivative of the log evidence given prepared, trained model hyper_t with respect to the hyper variable.