Crate sklears_multiclass 

Multiclass classification strategies

This module provides meta-estimators for multiclass classification problems. It implements strategies like One-vs-Rest and One-vs-One for transforming binary classifiers into multiclass ones.

Known Limitations

The following modules are disabled due to ndarray HRTB (Higher-Ranked Trait Bound) lifetime constraints introduced in ndarray 0.17. Planned for re-enabling in v0.2.0:

• advanced, calibration, core, ensemble - Core multiclass strategies
• boosting, dynamic_ensemble, ecoc - Ensemble multiclass methods
• one_vs_one, one_vs_rest, rotation_forest - Classification strategies

Re-exports

pub use utils::*;

Modules

export

Model Export and Serialization

gpu

GPU Acceleration Framework for Multiclass Classification

incremental

Incremental Learning Framework for Multiclass Classification

memory

Memory Optimization Features for Multiclass Classification

simd

SIMD Optimizations for Multiclass Classification

uncertainty

Uncertainty Quantification for Multiclass Classification

utils

Utility functions and structures for multiclass classification