summit.multiview_platform.multiview_classifiers package

Subpackages

• summit.multiview_platform.multiview_classifiers.additions package
  • Submodules
  • summit.multiview_platform.multiview_classifiers.additions.diversity_utils module
    • CoupleDiversityFusionClassifier
      • CoupleDiversityFusionClassifier.choose_combination()
      • CoupleDiversityFusionClassifier.set_fit_request()
      • CoupleDiversityFusionClassifier.set_predict_request()
    • DiversityFusionClassifier
      • DiversityFusionClassifier.fit()
      • DiversityFusionClassifier.get_classifiers_decisions()
      • DiversityFusionClassifier.init_combinations()
      • DiversityFusionClassifier.predict()
      • DiversityFusionClassifier.set_fit_request()
      • DiversityFusionClassifier.set_predict_request()
    • GlobalDiversityFusionClassifier
      • GlobalDiversityFusionClassifier.choose_combination()
      • GlobalDiversityFusionClassifier.set_fit_request()
      • GlobalDiversityFusionClassifier.set_predict_request()
  • summit.multiview_platform.multiview_classifiers.additions.early_fusion_from_monoview module
    • BaseEarlyFusion
      • BaseEarlyFusion.fit()
      • BaseEarlyFusion.get_feature_importances()
      • BaseEarlyFusion.get_params()
      • BaseEarlyFusion.hdf5_to_monoview()
      • BaseEarlyFusion.predict()
      • BaseEarlyFusion.set_fit_request()
      • BaseEarlyFusion.set_params()
      • BaseEarlyFusion.set_predict_request()
      • BaseEarlyFusion.transform_data_to_monoview()
  • summit.multiview_platform.multiview_classifiers.additions.fusion_utils module
    • BaseFusionClassifier
      • BaseFusionClassifier.init_monoview_estimator()
  • summit.multiview_platform.multiview_classifiers.additions.jumbo_fusion_utils module
    • BaseJumboFusion
      • BaseJumboFusion.fit()
      • BaseJumboFusion.fit_monoview_estimators()
      • BaseJumboFusion.predict()
      • BaseJumboFusion.predict_monoview()
      • BaseJumboFusion.set_fit_request()
      • BaseJumboFusion.set_params()
      • BaseJumboFusion.set_predict_request()
  • summit.multiview_platform.multiview_classifiers.additions.kernel_learning module
    • KernelClassifier
      • KernelClassifier.extract_labels()
      • KernelClassifier.format_X()
      • KernelClassifier.init_kernels()
      • KernelClassifier.set_fit_request()
      • KernelClassifier.set_predict_request()
    • KernelConfigDistribution
      • KernelConfigDistribution.draw()
    • KernelConfigGenerator
      • KernelConfigGenerator.rvs()
    • KernelDistribution
      • KernelDistribution.draw()
    • KernelGenerator
      • KernelGenerator.rvs()
  • summit.multiview_platform.multiview_classifiers.additions.late_fusion_utils module
    • ClassifierCombinator
      • ClassifierCombinator.rvs()
    • ClassifierDistribution
      • ClassifierDistribution.draw()
    • ConfigDistribution
      • ConfigDistribution.draw()
    • LateFusionClassifier
      • LateFusionClassifier.fit()
      • LateFusionClassifier.get_classifiers()
      • LateFusionClassifier.init_classifiers()
      • LateFusionClassifier.init_params()
      • LateFusionClassifier.set_fit_request()
      • LateFusionClassifier.set_predict_request()
    • MultipleConfigGenerator
      • MultipleConfigGenerator.rvs()
    • WeightDistribution
      • WeightDistribution.draw()
    • WeightsGenerator
      • WeightsGenerator.rvs()
  • summit.multiview_platform.multiview_classifiers.additions.utils module
    • get_names()
  • Module contents

Submodules

summit.multiview_platform.multiview_classifiers.bayesian_inference_fusion module

class BayesianInferenceClassifier(random_state, classifiers_names=None, classifier_configs=None, nb_cores=1, weights=None, rs=None)

Bases: LateFusionClassifier

predict(X, sample_indices=None, view_indices=None)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → BayesianInferenceClassifier

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → BayesianInferenceClassifier

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.difficulty_fusion module

class DifficultyFusion(random_state=None, classifier_names=None, monoview_estimators=None, classifier_configs=None)

Bases: GlobalDiversityFusionClassifier

This classifier is inspired by Kuncheva, Ludmila & Whitaker, Chris. (2000). Measures of Diversity in Classifier Ensembles. It find the subset of monoview classifiers with the best difficulty

diversity_measure(classifiers_decisions, combination, y)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → DifficultyFusion

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → DifficultyFusion

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.disagree_fusion module

class DisagreeFusion(random_state=None, classifier_names=None, monoview_estimators=None, classifier_configs=None)

Bases: CoupleDiversityFusionClassifier

This classifier is inspired by Kuncheva, Ludmila & Whitaker, Chris. (2000). Measures of Diversity in Classifier Ensembles. It find the subset of monoview classifiers with the best disagreement

diversity_measure(first_classifier_decision, second_classifier_decision, _)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → DisagreeFusion

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → DisagreeFusion

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.double_fault_fusion module

class DoubleFaultFusion(random_state=None, classifier_names=None, monoview_estimators=None, classifier_configs=None)

Bases: CoupleDiversityFusionClassifier

This classifier is inspired by Kuncheva, Ludmila & Whitaker, Chris. (2000). Measures of Diversity in Classifier Ensembles. It find the subset of monoview classifiers with the best double fault

diversity_measure(first_classifier_decision, second_classifier_decision, y)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → DoubleFaultFusion

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → DoubleFaultFusion

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.early_fusion_adaboost module

class EarlyFusionAdaboost(random_state=None, n_estimators=50, estimator=None, base_estimator_config=None, **kwargs)

Bases: BaseEarlyFusion

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionAdaboost

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionAdaboost

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.early_fusion_decision_tree module

class EarlyFusionDT(random_state=None, max_depth=None, criterion='gini', splitter='best', **kwargs)

Bases: BaseEarlyFusion

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionDT

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionDT

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.early_fusion_gradient_boosting module

class EarlyFusionGB(random_state=None, loss='exponential', max_depth=1.0, n_estimators=100, init=CustomDecisionTreeGB(max_depth=1), **kwargs)

Bases: BaseEarlyFusion

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionGB

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionGB

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.early_fusion_lasso module

class EarlyFusionLasso(random_state=None, alpha=1.0, max_iter=10, warm_start=False, **kwargs)

Bases: BaseEarlyFusion

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionLasso

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionLasso

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.early_fusion_random_forest module

class EarlyFusionRF(random_state=None, n_estimators=10, max_depth=None, criterion='gini', **kwargs)

Bases: BaseEarlyFusion

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionRF

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionRF

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.early_fusion_sgd module

class EarlyFusionSGD(random_state=None, loss='hinge', penalty='l2', alpha=0.0001, max_iter=5, tol=None, **kwargs)

Bases: BaseEarlyFusion

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionSGD

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionSGD

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.early_fusion_svm_rbf module

class EarlyFusionSVMRBF(random_state=None, C=1.0, **kwargs)

Bases: BaseEarlyFusion

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionSVMRBF

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EarlyFusionSVMRBF

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.entropy_fusion module

class EntropyFusion(random_state=None, classifier_names=None, monoview_estimators=None, classifier_configs=None)

Bases: GlobalDiversityFusionClassifier

This classifier is inspired by Kuncheva, Ludmila & Whitaker, Chris. (2000). Measures of Diversity in Classifier Ensembles. It find the subset of monoview classifiers with the best entropy

diversity_measure(classifiers_decisions, combination, y)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EntropyFusion

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → EntropyFusion

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.lp_norm_mkl module

class LPNormMKL(random_state=None, lmbda=0.1, nystrom_param=1, n_loops=50, precision=0.0001, use_approx=True, kernel='rbf', kernel_params=None)

Bases: KernelClassifier, MKL

fit(X, y, train_indices=None, view_indices=None)

Parameters:

• X (different formats are supported) –

  • Metriclearn_array {array-like, sparse matrix}, shape = (n_samples, n_features) Training multi-view input samples. can be also Kernel where attibute ‘kernel’ is set to precompute “precomputed”

  • Dictionary of {array like} with shape = (n_samples, n_features) for multi-view for each view.

  • Array of {array like} with shape = (n_samples, n_features) for multi-view for each view.

  • {array like} with (n_samples, nviews * n_features) with ‘views_ind’ diferent to ‘None’

• y (array-like, shape = (n_samples,)) – Target values (class labels). array of length n_samples containing the classification/regression labels for training data

• views_ind (array-like (default=[0, n_features//2, n_features])) –

  Paramater specifying how to extract the data views from X:

  • views_ind is a 1-D array of sorted integers, the entries indicate the limits of the slices used to extract the views, where view n is given by X[:, views_ind[n]:views_ind[n+1]].

    With this convention each view is therefore a view (in the NumPy sense) of X and no copy of the data is done.

Returns:

self – Returns self.

Return type:

object

predict(X, sample_indices=None, view_indices=None)

Parameters:

• X (dict dictionary with all views {array like} with shape = (n_samples, n_features) for multi-view) – for each view. or MultiModalData , MultiModalArray or {array-like,}, shape = (n_samples, n_features) Training multi-view input samples. can be also Kernel where attibute ‘kernel’ is set to precompute “precomputed”

• views_ind (array-like (default=[0, n_features//2, n_features])) –

  Paramater specifying how to extract the data views from X:

  • views_ind is a 1-D array of sorted integers, the entries indicate the limits of the slices used to extract the views, where view n is given by X[:, views_ind[n]:views_ind[n+1]].

    With this convention each view is therefore a view (in the NumPy sense) of X and no copy of the data is done.

Returns:

y – Predicted classes.

Return type:

numpy.ndarray, shape = (n_samples,)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → LPNormMKL

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → LPNormMKL

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.majority_voting_fusion module

class MajorityVoting(random_state, classifiers_names=None, classifier_configs=None, weights=None, nb_cores=1, rs=None)

Bases: LateFusionClassifier

This classifier is a late fusion that builds a majority vote between the views

predict(X, sample_indices=None, view_indices=None)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → MajorityVoting

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → MajorityVoting

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

exception VotingIndecision

Bases: Exception

summit.multiview_platform.multiview_classifiers.mucombo module

class MuCombo(estimator=None, n_estimators=50, random_state=None, **kwargs)

Bases: BaseMultiviewClassifier, MuComboClassifier

fit(X, y, train_indices=None, view_indices=None)

Build a multimodal boosted classifier from the training set (X, y).

Parameters:

• X (dict dictionary with all views) – or MultiModalData , MultiModalArray, MultiModalSparseArray or {array-like, sparse matrix}, shape = (n_samples, n_features) Training multi-view input samples. Sparse matrix can be CSC, CSR, COO, DOK, or LIL. COO, DOK and LIL are converted to CSR.

• y (array-like, shape = (n_samples,)) – Target values (class labels).

• views_ind (array-like (default=[0, n_features//2, n_features])) –

  Paramater specifying how to extract the data views from X:

  • If views_ind is a 1-D array of sorted integers, the entries indicate the limits of the slices used to extract the views, where view n is given by X[:, views_ind[n]:views_ind[n+1]].

    With this convention each view is therefore a view (in the NumPy sense) of X and no copy of the data is done.

  • If views_ind is an array of arrays of integers, then each array of integers views_ind[n] specifies the indices of the view n, which is then given by X[:, views_ind[n]].

    With this convention each view creates therefore a partial copy of the data in X. This convention is thus more flexible but less efficient than the previous one.

Returns:

self – Returns self.

Return type:

object

Raises:

• ValueError estimator must support sample_weight –

• ValueError where X and view_ind are not compatibles –

get_interpretation(directory, base_file_name, y_test, feature_ids, multi_class=False)

Base method that returns an empty string if there is not interpretation method in the classifier’s module

predict(X, sample_indices=None, view_indices=None)

Predict classes for X.

The predicted class of an input sample is computed as the weighted mean prediction of the classifiers in the ensemble.

Parameters:

X ({array-like, sparse matrix}, shape = (n_samples, n_features)) – Multi-view input samples. Sparse matrix can be CSC, CSR, COO, DOK, or LIL. COO, DOK and LIL are converted to CSR.

Returns:

y – Predicted classes.

Return type:

numpy.ndarray, shape = (n_samples,)

Raises:

ValueError 'X' input matrix must be have the same total number of features – of ‘X’ fit data

set_base_estim_from_dict(dict)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → MuCombo

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → MuCombo

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.mumbo module

class Mumbo(estimator=None, n_estimators=50, random_state=None, best_view_mode='edge', **kwargs)

Bases: BaseMultiviewClassifier, MumboClassifier

fit(X, y, train_indices=None, view_indices=None)

Build a multimodal boosted classifier from the training set (X, y).

Parameters:

• X (dict dictionary with all views) – or MultiModalData , MultiModalArray, MultiModalSparseArray or {array-like, sparse matrix}, shape = (n_samples, n_features) Training multi-view input samples. Sparse matrix can be CSC, CSR, COO, DOK, or LIL. COO, DOK and LIL are converted to CSR.

• y (array-like, shape = (n_samples,)) – Target values (class labels).

• views_ind (array-like (default=[0, n_features//2, n_features])) –

  Paramater specifying how to extract the data views from X:

  • If views_ind is a 1-D array of sorted integers, the entries indicate the limits of the slices used to extract the views, where view n is given by X[:, views_ind[n]:views_ind[n+1]].

    With this convention each view is therefore a view (in the NumPy sense) of X and no copy of the data is done.

  • If views_ind is an array of arrays of integers, then each array of integers views_ind[n] specifies the indices of the view n, which is then given by X[:, views_ind[n]].

    With this convention each view creates therefore a partial copy of the data in X. This convention is thus more flexible but less efficient than the previous one.

Returns:

self – Returns self.

Return type:

object

get_interpretation(directory, base_file_name, y_test, feature_ids, multi_class=False)

Base method that returns an empty string if there is not interpretation method in the classifier’s module

predict(X, sample_indices=None, view_indices=None)

Predict classes for X.

The predicted class of an input sample is computed as the weighted mean prediction of the classifiers in the ensemble.

Parameters:

X ({array-like, sparse matrix}, shape = (n_samples, n_features)) – Multi-view input samples. Sparse matrix can be CSC, CSR, COO, DOK, or LIL. COO, DOK and LIL are converted to CSR.

Returns:

y – Predicted classes.

Return type:

numpy.ndarray, shape = (n_samples,)

set_base_estim_from_dict(dict)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → Mumbo

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_params(estimator=None, **params)

Sets the estimator from a dict. :param estimator: :param params: :return:

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → Mumbo

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.mvml module

class MVMLClassifier(random_state=None, lmbda=0.1, eta=0.1, nystrom_param=1, n_loops=50, precision=0.0001, learn_A=0, kernel='rbf', learn_w=0, kernel_params=None)

Bases: KernelClassifier, MVML

fit(X, y, train_indices=None, view_indices=None)

Fit the MVML classifier

Parameters:

X –

Training multi-view input samples. can be also Kernel where attibute ‘kernel’

is set to precompute “precomputed”

or - Dictionary of {array like} with shape = (n_samples, n_features) for multi-view

for each view.

• Array of {array like} with shape = (n_samples, n_features) for multi-view for each view.

• {array like} with (n_samples, nviews * n_features) with ‘views_ind’ diferent to ‘None’

yarray-like, shape = (n_samples,)

Target values (class labels). array of length n_samples containing the classification/regression labels for training data

views_indarray-like (default=[0, n_features//2, n_features])

Paramater specifying how to extract the data views from X:

• views_ind is a 1-D array of sorted integers, the entries indicate the limits of the slices used to extract the views, where view n is given by X[:, views_ind[n]:views_ind[n+1]] .

  With this convention each view is therefore a view (in the NumPy sense) of X and no copy of the data is done.

Returns:

self – Returns self.

Return type:

object

predict(X, sample_indices=None, view_indices=None)

Parameters:

X (different formats are supported) –

• Metriclearn_array {array-like, sparse matrix}, shape = (n_samples, n_features) Training multi-view input samples. can be also Kernel where attibute ‘kernel’ is set to precompute “precomputed”

• Dictionary of {array like} with shape = (n_samples, n_features) for multi-view for each view.

• Array of {array like} with shape = (n_samples, n_features) for multi-view for each view.

• {array like} with (n_samples, nviews * n_features) with ‘views_ind’ diferent to ‘None’

Returns:

y – Predicted classes.

Return type:

numpy.ndarray, shape = (n_samples,)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → MVMLClassifier

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → MVMLClassifier

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.svm_jumbo_fusion module

class SVMJumboFusion(random_state=None, classifiers_names=None, classifier_configs=None, nb_cores=1, weights=None, nb_monoview_per_view=1, C=1.0, kernel='rbf', degree=2, rs=None)

Bases: BaseJumboFusion

This classifier learns monoview classifiers on each view and then uses an SVM on their decisions to aggregate them.

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → SVMJumboFusion

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_params(C=1.0, kernel='rbf', degree=1, **params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters:

**params (dict) – Estimator parameters.

Returns:

self – Estimator instance.

Return type:

estimator instance

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → SVMJumboFusion

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

summit.multiview_platform.multiview_classifiers.weighted_linear_early_fusion module

class WeightedLinearEarlyFusion(random_state=None, view_weights=None, monoview_classifier_name='decision_tree', monoview_classifier_config={})

Bases: BaseMultiviewClassifier, BaseFusionClassifier

Builds a monoview dataset by concatenating the views (with a weight if needed) and learns a monoview classifier on the concatenation

fit(X, y, train_indices=None, view_indices=None)

get_params(deep=True)

Get parameters for this estimator.

Parameters:

deep (bool, default=True) – If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns:

params – Parameter names mapped to their values.

Return type:

dict

hdf5_to_monoview(dataset, samples)

Here, we concatenate the views for the asked samples

predict(X, sample_indices=None, view_indices=None)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → WeightedLinearEarlyFusion

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_params(monoview_classifier_name='decision_tree', monoview_classifier_config={}, **params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters:

**params (dict) – Estimator parameters.

Returns:

self – Estimator instance.

Return type:

estimator instance

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → WeightedLinearEarlyFusion

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

transform_data_to_monoview(dataset, sample_indices, view_indices)

Here, we extract the data from the HDF5 dataset file and store all the concatenated views in one variable

summit.multiview_platform.multiview_classifiers.weighted_linear_late_fusion module

class WeightedLinearLateFusion(random_state, classifiers_names=None, classifier_configs=None, weights=None, nb_cores=1, rs=None)

Bases: LateFusionClassifier

Similar to the majority voting fusion.

predict(X, sample_indices=None, view_indices=None)

set_fit_request(*, train_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → WeightedLinearLateFusion

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to fit.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• train_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for train_indices parameter in fit.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in fit.

Returns:

self – The updated object.

Return type:

object

set_predict_request(*, sample_indices: bool | None | str = '$UNCHANGED$', view_indices: bool | None | str = '$UNCHANGED$') → WeightedLinearLateFusion

Request metadata passed to the predict method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

• True: metadata is requested, and passed to predict if provided. The request is ignored if metadata is not provided.

• False: metadata is not requested and the meta-estimator will not pass it to predict.

• None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

• str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

Added in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters:

• sample_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for sample_indices parameter in predict.

• view_indices (str, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED) – Metadata routing for view_indices parameter in predict.

Returns:

self – The updated object.

Return type:

object

Module contents