Crate sklears_neighbors 

Neighbor-based algorithms for machine learning

This crate provides k-nearest neighbors (k-NN) and related algorithms for classification, regression, and outlier detection.

Examples

use sklears_neighbors::KNeighborsClassifier;
use sklears_core::traits::{Fit, Predict};
use scirs2_core::ndarray::{array, Array2};

// Create training data
let X = Array2::from_shape_vec((4, 2), vec![
    1.0, 2.0,
    2.0, 3.0,
    3.0, 1.0,
    4.0, 2.0,
]).unwrap();
let y = array![0, 0, 1, 1];

// Train classifier
let classifier = KNeighborsClassifier::new(3);
let fitted = classifier.fit(&X, &y).unwrap();

// Make predictions
let X_test = Array2::from_shape_vec((2, 2), vec![
    1.5, 2.5,
    3.5, 1.5,
]).unwrap();
let predictions = fitted.predict(&X_test).unwrap();

Re-exports

pub use abod::AngleBasedOutlierDetection;

pub use adaptive_distance::AdaptiveDensityDistance;

pub use adaptive_distance::CombinationMethod;

pub use adaptive_distance::ContextDependentDistance;

pub use adaptive_distance::EnsembleDistance;

pub use adaptive_distance::OnlineAdaptiveDistance;

pub use advanced_outliers::ConnectivityBasedOutlierFactor;

pub use advanced_outliers::IsolationForest;

pub use advanced_outliers::LocalCorrelationIntegral;

pub use approximate_distance::ApproximateDistance;

pub use batch_processing::BatchConfiguration;

pub use batch_processing::BatchNeighborSearch;

pub use batch_processing::BatchProcessable;

pub use batch_processing::BatchProcessor;

pub use batch_processing::BatchResult;

pub use batch_processing::BatchStatistics;

pub use batch_processing::MemoryMonitor;

pub use bayesian_neighbors::BayesianKNeighborsClassifier;

pub use bayesian_neighbors::BayesianKNeighborsRegressor;

pub use bayesian_neighbors::BayesianPrediction;

pub use bayesian_neighbors::BayesianRegressionPrediction;

pub use bayesian_neighbors::CredibleNeighborSet;

pub use bayesian_neighbors::UncertaintyMethod;

pub use bioinformatics::BioSearchConfig;

pub use bioinformatics::GeneExpressionNeighbors;

pub use bioinformatics::GeneExpressionResult;

pub use bioinformatics::GeneMetadata;

pub use bioinformatics::KmerIndex;

pub use bioinformatics::ProteinMetadata;

pub use bioinformatics::ProteinSearchResult;

pub use bioinformatics::ProteinStructure;

pub use bioinformatics::ProteinStructureSearch;

pub use bioinformatics::ScoringScheme;

pub use bioinformatics::SequenceAligner;

pub use bioinformatics::SequenceAlignment;

pub use bioinformatics::SequenceMetadata;

pub use bioinformatics::SequenceSearchResult;

pub use bioinformatics::SequenceSimilaritySearch;

pub use bioinformatics::SequenceType;

pub use compressed_distance::CompressedDistanceMatrix;

pub use compressed_distance::CompressionMethod;

pub use compressed_distance::CompressionStats;

pub use computer_vision::ColorHistogramExtractor;

pub use computer_vision::DescriptorMatch;

pub use computer_vision::FeatureDescriptorMatcher;

pub use computer_vision::FeatureExtractor;

pub use computer_vision::FeatureType;

pub use computer_vision::HistogramOfGradientsExtractor;

pub use computer_vision::ImageMetadata;

pub use computer_vision::ImageSearchConfig;

pub use computer_vision::ImageSearchResult;

pub use computer_vision::ImageSimilaritySearch;

pub use computer_vision::Keypoint;

pub use computer_vision::LocalBinaryPatternExtractor;

pub use computer_vision::PatchBasedMatching;

pub use computer_vision::VisualWordRecognizer;

pub use cross_validation::CVFoldResult;

pub use cross_validation::CVResults;

pub use cross_validation::CVStrategy;

pub use cross_validation::NeighborCrossValidator;

pub use density_estimation::BandwidthMethod;

pub use density_estimation::DensityBasedClustering;

pub use density_estimation::KNeighborsDensityEstimator;

pub use density_estimation::KernelType;

pub use density_estimation::LocalDensityEstimator;

pub use density_estimation::VariableBandwidthKDE;

pub use distance::Distance;

pub use distributed_neighbors::DataPartitioner;

pub use distributed_neighbors::DistributedConfiguration;

pub use distributed_neighbors::DistributedNeighborSearch;

pub use distributed_neighbors::DistributedNeighborSearchResult;

pub use distributed_neighbors::DistributedWorker;

pub use distributed_neighbors::LoadBalanceStrategy;

pub use distributed_neighbors::PartitionInfo;

pub use distributed_neighbors::PartitionStats;

pub use federated_neighbors::FederatedConfig;

pub use federated_neighbors::FederatedNeighborCoordinator;

pub use federated_neighbors::FederatedParticipant;

pub use federated_neighbors::NoiseStrategy;

pub use federated_neighbors::PrivacyLevel;

pub use federated_neighbors::PrivacyPreservingProtocol;

pub use gpu_distance::GpuBackend;

pub use gpu_distance::GpuComputationStats;

pub use gpu_distance::GpuConfig;

pub use gpu_distance::GpuDeviceInfo;

pub use gpu_distance::GpuDistanceCalculator;

pub use gpu_distance::GpuDistanceResult;

pub use gpu_distance::GpuKNeighborsSearch;

pub use gpu_distance::GpuMemoryEstimator;

pub use gpu_distance::GpuMemoryStrategy;

pub use graph_methods::EpsilonGraph;

pub use graph_methods::GabrielGraph;

pub use graph_methods::GraphEdge;

pub use graph_methods::GraphNeighborSearch;

pub use graph_methods::GraphStatistics;

pub use graph_methods::KNearestNeighborGraph;

pub use graph_methods::MutualKNearestNeighbors;

pub use graph_methods::NeighborhoodGraph;

pub use graph_methods::RelativeNeighborhoodGraph;

pub use incremental_index::IncrementalIndexBuilder;

pub use incremental_index::IncrementalIndexType;

pub use incremental_index::IncrementalNeighborIndex;

pub use incremental_index::IndexPerformanceMetrics;

pub use incremental_index::UpdateStrategy;

pub use interpretability::InfluenceAnalysis;

pub use interpretability::LocalImportanceExplanation;

pub use interpretability::NeighborExplainer;

pub use interpretability::NeighborExplanation;

pub use interpretability::Prototype;

pub use knn::KNeighborsClassifier;

pub use knn::KNeighborsRegressor;

pub use local_outlier_factor::LocalOutlierFactor;

pub use lsh::HashFamily;

pub use lsh::LshIndex;

pub use lsh::LshKNeighborsClassifier;

pub use manifold_learning::Isomap;

pub use manifold_learning::LaplacianEigenmaps;

pub use manifold_learning::LocallyLinearEmbedding;

pub use manifold_learning::TSNENeighbors;

pub use mapreduce_neighbors::DistributedMapReduce;

pub use mapreduce_neighbors::MapReduceConfig;

pub use mapreduce_neighbors::MapReduceNeighborSearch;

pub use mapreduce_neighbors::PartitionStrategy;

pub use mapreduce_neighbors::ReduceStrategy;

pub use memory_constrained::CacheObliviousNeighbors;

pub use memory_constrained::ExternalMemoryKNN;

pub use memory_constrained::MemoryBoundedApproximateNeighbors;

pub use memory_mapped::MmapNeighborIndex;

pub use memory_mapped::MmapNeighborIndexBuilder;

pub use metric_learning::EnhancedLMNN;

pub use metric_learning::InformationTheoreticMetricLearning;

pub use metric_learning::LargeMarginNearestNeighbor;

pub use metric_learning::NeighborhoodComponentsAnalysis;

pub use metric_learning::OnlineMetricLearning;

pub use multi_view_learning::ConsensusAnalysis;

pub use multi_view_learning::FusionStrategy;

pub use multi_view_learning::MultiViewKNeighborsClassifier;

pub use multi_view_learning::MultiViewKNeighborsRegressor;

pub use multi_view_learning::RegressionFusionStrategy;

pub use multi_view_learning::ViewConfig;

pub use nearest_centroid::CentroidType;

pub use nearest_centroid::ClassConfig;

pub use nearest_centroid::NearestCentroid;

pub use nearest_neighbors::kneighbors_graph;

pub use nearest_neighbors::radius_neighbors_graph;

pub use nearest_neighbors::NearestNeighbors;

pub use nlp::DocumentFeatureExtractor;

pub use nlp::DocumentMetadata;

pub use nlp::DocumentSearchResult;

pub use nlp::DocumentSimilaritySearch;

pub use nlp::NlpSearchConfig;

pub use nlp::SentenceSimilaritySearch;

pub use nlp::TextFeatureType;

pub use nlp::TextPreprocessor;

pub use nlp::TfIdfExtractor;

pub use nlp::WordEmbeddingSearch;

pub use online_learning::AdaptiveKNeighborsClassifier;

pub use online_learning::DriftDetectionMethod;

pub use online_learning::DriftDetector;

pub use online_learning::StreamingOutlierDetector;

pub use parallel_tree::ParallelBuildStrategy;

pub use parallel_tree::ParallelTreeBuilder;

pub use parallel_tree::ParallelTreeIndex;

pub use parallel_tree::WorkUnit;

pub use performance::BenchmarkConfig;

pub use performance::BenchmarkResult;

pub use performance::NeighborBenchmark;

pub use performance::PerformanceMetrics;

pub use performance::QuickProfiler;

pub use radius_neighbors::AdaptiveRadiusNeighborsClassifier;

pub use radius_neighbors::AdaptiveRadiusNeighborsRegressor;

pub use radius_neighbors::RadiusNeighborsClassifier;

pub use radius_neighbors::RadiusNeighborsRegressor;

pub use radius_neighbors::RadiusStrategy;

pub use simd_distance::batch_euclidean_distances;

pub use simd_distance::pairwise_distances_simd;

pub use simd_distance::SimdCapability;

pub use simd_distance::SimdDistanceCalculator;

pub use sparse_neighbors::SparseIndexType;

pub use sparse_neighbors::SparseNeighborBuilder;

pub use sparse_neighbors::SparseNeighborMatrix;

pub use spatial::OctPoint;

pub use spatial::OctTree;

pub use spatial::QuadPoint;

pub use spatial::QuadTree;

pub use spatial::RTree;

pub use spatial::Rectangle;

pub use spatial::SpatialHash;

pub use spatial::SpatialHashStats;

pub use specialized_distances::CategoricalDistance;

pub use specialized_distances::GraphDistance;

pub use specialized_distances::ProbabilisticDistance;

pub use specialized_distances::SetDistance;

pub use specialized_distances::SimpleGraph;

pub use specialized_distances::StringDistance;

pub use streaming::IncrementalKNeighborsClassifier;

pub use streaming::IncrementalKNeighborsRegressor;

pub use streaming::MemoryStrategy;

pub use time_series_neighbors::DtwDistance;

pub use time_series_neighbors::DtwStepPattern;

pub use time_series_neighbors::Shapelet;

pub use time_series_neighbors::ShapeletDiscovery;

pub use time_series_neighbors::StreamingTimeSeriesNeighbors;

pub use time_series_neighbors::SubsequenceSearch;

pub use time_series_neighbors::TemporalNeighborSearch;

pub use transformers::KNeighborsTransformer;

pub use transformers::RadiusNeighborsTransformer;

pub use type_safe_distance::ChebyshevMetric;

pub use type_safe_distance::ComputeDistance;

pub use type_safe_distance::CosineMetric;

pub use type_safe_distance::EuclideanMetric;

pub use type_safe_distance::ManhattanMetric;

pub use type_safe_distance::MetricDistance;

pub use type_safe_distance::MinkowskiMetric;

pub use type_safe_distance::NonMetricDistance;

pub use type_safe_distance::NormalizedDistance;

pub use type_safe_distance::TypeSafeDistance;

pub use type_safe_distance::TypeSafeKnnConfig;

pub use validation::BootstrapResult;

pub use validation::BootstrapValidator;

pub use validation::ClassificationMetric;

pub use validation::CrossValidationResult;

pub use validation::GridSearchCV;

pub use validation::GridSearchResult;

pub use validation::KFoldValidator;

pub use validation::RegressionMetric;

Modules

abod

Angle-Based Outlier Detection (ABOD)

adaptive_distance

Adaptive distance metrics that adjust based on data characteristics

advanced_outliers

Advanced outlier detection algorithms

approximate_distance

Approximate distance computations for faster neighbor search

batch_processing

bayesian_neighbors

Bayesian Nearest Neighbors with Uncertainty Quantification

bioinformatics

Bioinformatics Neighbor-Based Methods

compressed_distance

Compressed distance matrices for memory-efficient neighbor storage

computer_vision

Computer Vision Neighbor-Based Methods

cross_validation

Advanced cross-validation techniques for neighbor-based algorithms

density_estimation

Density estimation algorithms using nearest neighbors

distance

Distance metrics for neighbor-based algorithms

distributed_neighbors

federated_neighbors

Federated neighbor computation for privacy-preserving machine learning

gpu_distance

GPU-accelerated distance computations for high-performance neighbor search

graph_methods

Graph-based neighbor methods for machine learning

incremental_index

Incremental index construction for dynamic neighbor search

interpretability

Interpretability and explainability tools for neighbor-based algorithms

knn

K-Nearest Neighbors algorithms for classification and regression

local_outlier_factor

Local Outlier Factor for anomaly detection

lsh

Locality-Sensitive Hashing (LSH) for approximate nearest neighbor search

manifold_learning

Manifold learning algorithms for dimensionality reduction using neighbor-based methods

mapreduce_neighbors

MapReduce-style neighbor search for distributed computing

memory_constrained

Memory-constrained algorithms for neighbor search with limited memory

memory_mapped

Memory-mapped neighbor indices for large datasets

metric_learning

Metric learning algorithms for adaptive distance computation

multi_view_learning

Multi-View Learning for Neighbor-Based Methods

nearest_centroid

Nearest Centroid Classifier

nearest_neighbors

Unsupervised Nearest Neighbors learner

nlp

Natural Language Processing Neighbor-Based Methods

online_learning

Online Learning with Concept Drift Detection

parallel_tree

Parallel tree construction for neighbor search indices

performance

Performance analysis and benchmarking tools for neighbor algorithms

radius_neighbors

Radius-based neighbor algorithms for classification and regression

simd_distance

sparse_neighbors

Sparse neighbor representations for memory-efficient storage

spatial

Spatial data structures for efficient nearest neighbor search

specialized_distances

Specialized distance metrics for various data types

streaming

Incremental and streaming nearest neighbor algorithms

time_series_neighbors

Time series neighbor search methods

transformers

Neighbor-based transformers for building graphs and embeddings

tree

Tree-based data structures for efficient neighbor search

type_safe_distance

Type-safe distance metrics using phantom types

validation

Validation Framework for Neighbor-Based Methods

Enums

NeighborsError

Common error type for neighbors algorithms

Type Aliases

NeighborsResult

Type alias for neighbors results