.. index:: single: bayesian_ridge_regression .. _bayesian_ridge_regression/0:

.. rst-class:: right

object

bayesian_ridge_regression

Bayesian ridge regression regressor supporting continuous and mixed-feature datasets using evidence maximization and posterior uncertainty over coefficients. Learns from a dataset object implementing the regression_dataset_protocol protocol and returns a regressor term that can be used for prediction, predictive-distribution queries, and export as predicate clauses.

| Availability: | logtalk_load(bayesian_ridge_regression(loader))

| Author: Paulo Moura | Version: 1:0:0 | Date: 2026-05-07

| Compilation flags: | static, context_switching_calls

| Imports: | public :ref:`regressor_common <regressor_common/0>` | Uses: | :ref:`format <format/0>` | :ref:`linear_algebra <linear_algebra/0>` | :ref:`list <list/0>` | :ref:`numberlist <numberlist/0>` | :ref:`population <population/0>` | :ref:`type <type/0>`

| Remarks: | (none)

| Inherited public predicates: |  :ref:`options_protocol/0::check_option/1`  :ref:`options_protocol/0::check_options/1`  :ref:`regressor_protocol/0::check_regressor/1`  :ref:`options_protocol/0::default_option/1`  :ref:`options_protocol/0::default_options/1`  :ref:`regressor_protocol/0::diagnostic/2`  :ref:`regressor_protocol/0::diagnostics/2`  :ref:`regressor_protocol/0::export_to_clauses/4`  :ref:`regressor_protocol/0::export_to_file/4`  :ref:`regressor_protocol/0::learn/2`  :ref:`regressor_protocol/0::learn/3`  :ref:`options_protocol/0::option/2`  :ref:`options_protocol/0::option/3`  :ref:`regressor_protocol/0::predict/3`  :ref:`regressor_protocol/0::print_regressor/1`  :ref:`regressor_protocol/0::regressor_options/2`  :ref:`options_protocol/0::valid_option/1`  :ref:`options_protocol/0::valid_options/1`  :ref:`regressor_protocol/0::valid_regressor/1`  

.. contents:: :local: :backlinks: top

Public predicates

.. index:: predict_distribution/3 .. _bayesian_ridge_regression/0::predict_distribution/3:

predict_distribution/3 ^^^^^^^^^^^^^^^^^^^^^^^^^^

Predicts the posterior predictive Gaussian distribution for a new instance using the learned regressor. The returned term has the shape gaussian(Mean, Variance) where Variance includes the learned observation noise variance and coefficient posterior uncertainty only; the intercept is not treated as a probabilistic parameter.

| Compilation flags: | static

| Template: | predict_distribution(Regressor,Instance,Distribution) | Mode and number of proofs: | predict_distribution(+compound,+list,-compound) - one

---

.. index:: weight_variances/2 .. _bayesian_ridge_regression/0::weight_variances/2:

weight_variances/2 ^^^^^^^^^^^^^^^^^^^^^^

Returns the posterior marginal variances of the encoded feature coefficients in encoder order. Encoded features dropped from fitting because they had zero variance are reported with posterior variance zero.

| Compilation flags: | static

| Template: | weight_variances(Regressor,Variances) | Mode and number of proofs: | weight_variances(+compound,-list(float)) - one

---

Protected predicates

(no local declarations; see entity ancestors if any)

Private predicates

(no local declarations; see entity ancestors if any)

Operators

(none)

.. seealso::

:ref:`linear_regression <linear_regression/0>`, :ref:`ridge_regression <ridge_regression/0>`, :ref:`lasso_regression <lasso_regression/0>`, :ref:`elastic_net_regression <elastic_net_regression/0>`, :ref:`gaussian_process_regression <gaussian_process_regression/0>`, :ref:`knn_regression <knn_regression/0>`, :ref:`regression_tree <regression_tree/0>`, :ref:`random_forest_regression <random_forest_regression/0>`, :ref:`gradient_boosting_regression <gradient_boosting_regression/0>`