Expand description
Isotonic regression
This module is part of sklears, providing scikit-learn compatible machine learning algorithms in Rust.
This crate provides comprehensive isotonic regression functionality including:
- Basic isotonic regression with Pool Adjacent Violators Algorithm
- Robust loss functions (L1, L2, Huber, Quantile)
- Weighted isotonic regression
- Constraint handling and validation
Additional advanced features are being progressively enabled as they pass compilation and testing.
Re-exports§
pub use robust::huber_weighted_mean;pub use robust::loss_functions;pub use robust::robust_statistics;pub use optimization::create_partial_order;pub use optimization::interpolate_multidimensional;pub use optimization::isotonic_regression_active_set;pub use optimization::isotonic_regression_dual_decomposition;pub use optimization::isotonic_regression_interior_point;pub use optimization::isotonic_regression_projected_gradient;pub use optimization::isotonic_regression_qp;pub use optimization::non_separable_isotonic_regression;pub use optimization::parallel_dual_decomposition;pub use optimization::separable_isotonic_regression;pub use optimization::simd_armijo_line_search;pub use optimization::simd_constraint_violations;pub use optimization::simd_dot_product;pub use optimization::simd_gradient_computation;pub use optimization::simd_hessian_approximation;pub use optimization::simd_isotonic_projection;pub use optimization::simd_newton_step;pub use optimization::simd_qp_matrix_vector_multiply;pub use optimization::simd_vector_norm;pub use optimization::sparse_isotonic_regression;pub use optimization::ActiveSetIsotonicRegressor;pub use optimization::BenchmarkResults;pub use optimization::DualDecompositionIsotonicRegressor;pub use optimization::InteriorPointIsotonicRegressor;pub use optimization::NonSeparableMultiDimensionalIsotonicRegression;pub use optimization::OptimizationAlgorithm;pub use optimization::OptimizationConfig;pub use optimization::ProjectedGradientIsotonicRegressor;pub use optimization::QuadraticProgrammingIsotonicRegressor;pub use optimization::SeparableMultiDimensionalIsotonicRegression;pub use optimization::SparseIsotonicRegression;pub use constraints::*;pub use core::*;pub use pav::*;pub use utils::*;pub use algorithms::*;pub use fluent_api::*;pub use serialization::*;pub use regularized::*;pub use differential_equations::*;pub use engineering_applications::*;pub use environmental_science::*;pub use ml_integration::*;pub use advanced_bayesian::*;pub use graph_methods::*;pub use middleware::*;pub use case_studies::*;pub use unsafe_optimizations::*;pub use advanced_benchmarks::*;
Modules§
- advanced_
bayesian - advanced_
benchmarks - Advanced benchmarking suite for isotonic regression
- algorithms
- Core isotonic regression algorithms
- case_
studies - Real-world case studies and practical examples for isotonic regression
- constraints
- Constraint handling and validation for isotonic regression
- convex_
optimization - Convex Optimization Module for Isotonic Regression
- core
- Core isotonic regression types and implementations
- differential_
equations - engineering_
applications - environmental_
science - fluent_
api - Fluent API for isotonic regression configuration
- graph_
methods - Advanced graph-based methods for isotonic regression
- middleware
- Middleware for constraint pipelines
- ml_
integration - optimization
- Advanced optimization algorithms for isotonic regression
- pav
- Pool Adjacent Violators Algorithm (PAVA) implementations
- regularized
- Regularized isotonic regression algorithms
- robust
- Robust loss functions and utilities for isotonic regression
- serialization
- Serialization and deserialization for isotonic regression models
- unsafe_
optimizations - Unsafe optimizations for performance-critical paths
- utils
- Utility functions for isotonic regression