Crate sklears_gaussian_process 

Gaussian Process models for regression and classification

This module is part of sklears, providing scikit-learn compatible machine learning algorithms in Rust.

Re-exports

pub use automatic_kernel::AutomaticKernelConstructor;

pub use automatic_kernel::DataCharacteristics;

pub use automatic_kernel::KernelConstructionResult;

pub use bayesian_optimization::AcquisitionFunction;

pub use bayesian_optimization::BayesianOptimizer;

pub use bayesian_optimization::BayesianOptimizerFitted;

pub use classification::sigmoid;

pub use classification::sigmoid_derivative;

pub use classification::EpGpcTrained;

pub use classification::ExpectationPropagationGaussianProcessClassifier;

pub use classification::GaussianProcessClassifier;

pub use classification::GpcConfig;

pub use classification::GpcTrained;

pub use classification::McGpcTrained;

pub use classification::MultiClassGaussianProcessClassifier;

pub use features::RandomFourierFeatures;

pub use features::RandomFourierFeaturesGPR;

pub use features::RffGprTrained;

pub use gpr::GaussianProcessRegressor;

pub use gpr::GprTrained;

pub use heteroscedastic::ConstantNoise;

pub use heteroscedastic::HeteroscedasticGaussianProcessRegressor;

pub use heteroscedastic::LearnableNoiseFunction;

pub use heteroscedastic::LinearNoise;

pub use heteroscedastic::NeuralNetworkNoise;

pub use heteroscedastic::NoiseFunction;

pub use heteroscedastic::PolynomialNoise;

pub use heteroscedastic::Trained as HeteroscedasticGprTrained;

pub use hierarchical::HierarchicalGPConfig;

pub use hierarchical::HierarchicalGaussianProcessRegressor;

pub use intrinsic_coregionalization::IcmTrained;

pub use intrinsic_coregionalization::IntrinsicCoregionalizationModel;

pub use kernel_optimization::OptimizationResult as KernelOptimizationResult;

pub use kernel_optimization::optimize_kernel_parameters;

pub use kernel_optimization::KernelOptimizer;

pub use linear_model_coregionalization::LinearModelCoregionalization;

pub use linear_model_coregionalization::LmcTrained;

pub use kernel_structure_learning::ConvergenceInfo;

pub use kernel_structure_learning::KernelGrammar;

pub use kernel_structure_learning::KernelStructureLearner;

pub use kernel_structure_learning::NonTerminalOperation;

pub use kernel_structure_learning::SearchStrategy;

pub use kernel_structure_learning::StructureLearningResult;

pub use kernel_structure_learning::TerminalKernel;

pub use marginal_likelihood::OptimizationResult as MarginalLikelihoodOptimizationResult;

pub use marginal_likelihood::cross_validate_hyperparameters;

pub use marginal_likelihood::log_marginal_likelihood;

pub use marginal_likelihood::log_marginal_likelihood_stable;

pub use marginal_likelihood::optimize_hyperparameters;

pub use marginal_likelihood::MarginalLikelihoodOptimizer;

pub use multi_task::MtgpTrained;

pub use multi_task::MultiTaskGaussianProcessRegressor;

pub use noise_function_learning::AdaptiveRegularization;

pub use noise_function_learning::AutomaticNoiseFunctionSelector;

pub use noise_function_learning::CombinationMethod;

pub use noise_function_learning::EnsembleNoiseFunction;

pub use noise_function_learning::InformationCriterion;

pub use noise_function_learning::NoiseFunctionEvaluation;

pub use nystrom::LandmarkSelection;

pub use nystrom::NystromGaussianProcessRegressor;

pub use nystrom::NystromGprTrained;

pub use regression::MogprTrained;

pub use regression::MultiOutputGaussianProcessRegressor;

pub use regression::VariationalOptimizer;

pub use regression::VariationalSparseGaussianProcessRegressor;

pub use regression::VsgprTrained;

pub use robust::OutlierDetectionMethod;

pub use robust::RobustGPConfig;

pub use robust::RobustGaussianProcessRegressor;

pub use robust::RobustLikelihood;

pub use robust::RobustnessMetrics;

pub use robust::Trained as RobustGprTrained;

pub use sparse_gpr::InducingPointInit;

pub use sparse_gpr::SgprTrained;

pub use sparse_gpr::SparseGaussianProcessRegressor;

pub use sparse_spectrum::SparseSpectrumGaussianProcessRegressor;

pub use sparse_spectrum::SparseSpectrumGprTrained;

pub use sparse_spectrum::SpectralApproximationInfo;

pub use sparse_spectrum::SpectralSelectionMethod;

pub use spatial::KrigingType;

pub use spatial::SpatialGPConfig;

pub use spatial::SpatialGaussianProcessRegressor;

pub use spatial::SpatialKernel;

pub use spatial::Trained as SpatialGprTrained;

pub use spatial::Variogram;

pub use structured_kernel_interpolation::GridBoundsMethod;

pub use structured_kernel_interpolation::InterpolationMethod;

pub use structured_kernel_interpolation::SkiApproximationInfo;

pub use structured_kernel_interpolation::SkiGprTrained;

pub use structured_kernel_interpolation::StructuredKernelInterpolationGPR;

pub use variational_deep_gp::VariationalDeepGPBuilder;

pub use variational_deep_gp::VariationalDeepGPConfig;

pub use variational_deep_gp::VariationalDeepGaussianProcess;

pub use variational_deep_gp::VariationalLayerConfig;

pub use variational_deep_gp::VariationalLayerParameters;

pub use variational_deep_gp::VariationalLikelihood;

pub use kernels::Kernel;

pub use kernels::ARDRBF;

pub use kernels::RBF;

Modules

automatic_kernel

Automatic kernel construction and selection

bayesian_optimization

Bayesian Optimization framework for hyperparameter tuning and global optimization

classification

Gaussian Process Classification Models

convolution_processes

Convolution Processes for multi-output Gaussian processes

deep_gp

Deep Gaussian Processes

features

Random Fourier Features for approximating RBF kernels

fitc

Fully Independent Training Conditionals (FITC) approximation for Gaussian Processes

gpr

Gaussian Process Regression implementations

heteroscedastic

Heteroscedastic Gaussian Process Regression

hierarchical

Hierarchical Gaussian Processes

intrinsic_coregionalization

Intrinsic Coregionalization Model (ICM) for multi-output Gaussian processes

kernel_optimization

Kernel parameter optimization

kernel_selection

Kernel selection methods for Gaussian Processes

kernel_structure_learning

Advanced kernel structure learning using grammar-based search

kernel_trait

Core kernel trait for Gaussian Process models

kernels

Kernel functions for Gaussian Process models

linear_model_coregionalization

Linear Model of Coregionalization (LMC) for multi-output Gaussian Process regression

marginal_likelihood

Marginal likelihood optimization for Gaussian Process hyperparameters

multi_task

Multi-Task Gaussian Process for learning multiple related tasks

noise_function_learning

Advanced Noise Function Learning for Heteroscedastic Gaussian Processes

nystrom

Nyström approximation for scalable Gaussian Process regression

regression

Gaussian Process Regression Models

robust

Robust Gaussian Processes and Outlier-Resistant Methods

sparse_gpr

Sparse Gaussian Process Regression implementations

sparse_spectrum

Sparse spectrum Gaussian processes for large-scale approximation

spatial

Spatial Gaussian Processes and Kriging Methods

structured_kernel_interpolation

Structured Kernel Interpolation (SKI) for scalable Gaussian processes

utils

Utility functions for Gaussian Process computations

variational

Variational Gaussian Process models

variational_deep_gp

Variational Deep Gaussian Processes