climb.tool.impl.data_suite.third_party.uq360.algorithms package

Subpackages

• climb.tool.impl.data_suite.third_party.uq360.algorithms.actively_learned_model package
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.actively_learned_model.actively_learned_model module
    • ActivelyLearnedModel
      • ActivelyLearnedModel.fit()
      • ActivelyLearnedModel.predict()
  • Module contents
• climb.tool.impl.data_suite.third_party.uq360.algorithms.auxiliary_interval_predictor package
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.auxiliary_interval_predictor.auxiliary_interval_predictor module
    • AuxiliaryIntervalPredictor
      • AuxiliaryIntervalPredictor.fit()
      • AuxiliaryIntervalPredictor.get_params()
      • AuxiliaryIntervalPredictor.predict()
  • Module contents
• climb.tool.impl.data_suite.third_party.uq360.algorithms.blackbox_metamodel package
  • Subpackages
    • climb.tool.impl.data_suite.third_party.uq360.algorithms.blackbox_metamodel.predictors package
      • Subpackages
      • Submodules
      • climb.tool.impl.data_suite.third_party.uq360.algorithms.blackbox_metamodel.predictors.predictor_driver module
      • Module contents
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.blackbox_metamodel.confidence_classification module
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.blackbox_metamodel.metamodel_classification module
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.blackbox_metamodel.metamodel_regression module
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.blackbox_metamodel.passthrough_classification module
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.blackbox_metamodel.short_text_classification module
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.blackbox_metamodel.structured_data_classification module
  • Module contents
• climb.tool.impl.data_suite.third_party.uq360.algorithms.classification_calibration package
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.classification_calibration.classification_calibration module
    • ClassificationCalibration
      • ClassificationCalibration.fit()
      • ClassificationCalibration.get_params()
      • ClassificationCalibration.predict()
  • Module contents
• climb.tool.impl.data_suite.third_party.uq360.algorithms.ensemble_heteroscedastic_regression package
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.ensemble_heteroscedastic_regression.ensemble_heteroscedastic_regression module
    • EnsembleHeteroscedasticRegression
      • EnsembleHeteroscedasticRegression.fit()
      • EnsembleHeteroscedasticRegression.predict()
  • Module contents
• climb.tool.impl.data_suite.third_party.uq360.algorithms.heteroscedastic_regression package
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.heteroscedastic_regression.heteroscedastic_regression module
    • HeteroscedasticRegression
      • HeteroscedasticRegression.fit()
      • HeteroscedasticRegression.get_params()
      • HeteroscedasticRegression.predict()
  • Module contents
• climb.tool.impl.data_suite.third_party.uq360.algorithms.homoscedastic_gaussian_process_regression package
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.homoscedastic_gaussian_process_regression.homoscedastic_gaussian_process_regression module
  • Module contents
• climb.tool.impl.data_suite.third_party.uq360.algorithms.infinitesimal_jackknife package
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.infinitesimal_jackknife.infinitesimal_jackknife module
    • InfinitesimalJackknife
      • InfinitesimalJackknife.approx_ij()
      • InfinitesimalJackknife.get_parameter_uncertainty()
      • InfinitesimalJackknife.get_params()
      • InfinitesimalJackknife.ij()
      • InfinitesimalJackknife.predict()
  • Module contents
• climb.tool.impl.data_suite.third_party.uq360.algorithms.layer_scoring package
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.layer_scoring.aklpe module
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.layer_scoring.knn module
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.layer_scoring.latent_scorer module
    • LatentScorer
      • LatentScorer.fit()
      • LatentScorer.get_latents()
      • LatentScorer.predict()
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.layer_scoring.mahalanobis module
  • Module contents
• climb.tool.impl.data_suite.third_party.uq360.algorithms.quantile_regression package
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.quantile_regression.quantile_regression module
    • QuantileRegression
      • QuantileRegression.fit()
      • QuantileRegression.get_params()
      • QuantileRegression.predict()
  • Module contents
• climb.tool.impl.data_suite.third_party.uq360.algorithms.ucc_recalibration package
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.ucc_recalibration.ucc_recalibration module
  • Module contents
• climb.tool.impl.data_suite.third_party.uq360.algorithms.variational_bayesian_neural_networks package
  • Submodules
  • climb.tool.impl.data_suite.third_party.uq360.algorithms.variational_bayesian_neural_networks.bnn module
    • BnnClassification
      • BnnClassification.fit()
      • BnnClassification.get_params()
      • BnnClassification.predict()
    • BnnRegression
      • BnnRegression.fit()
      • BnnRegression.get_params()
      • BnnRegression.predict()
  • Module contents

Submodules

climb.tool.impl.data_suite.third_party.uq360.algorithms.builtinuq module

class climb.tool.impl.data_suite.third_party.uq360.algorithms.builtinuq.BuiltinUQ(*argv, **kwargs)

Bases: ABC

BuiltinUQ is the base class for any algorithm that has UQ built into it.

abstract fit(*argv, **kwargs)

Learn the UQ related parameters..

abstract predict(*argv, **kwargs)

Method to obtain the predicitve uncertainty, this can return the total, epistemic and/or aleatoric uncertainty in the predictions.

set_params(**parameters)

climb.tool.impl.data_suite.third_party.uq360.algorithms.posthocuq module

class climb.tool.impl.data_suite.third_party.uq360.algorithms.posthocuq.PostHocUQ(*argv, **kwargs)

Bases: ABC

PostHocUQ is the base class for any algorithm that quantifies uncertainty of a pre-trained model.

get_params()

This method should not take any arguments and returns a dict of the __init__ parameters.

abstract predict(*argv, **kwargs)

Method to obtain the predicitve uncertainty, this can return the total, epistemic and/or aleatoric uncertainty in the predictions.

set_params(**parameters)

Module contents