Crate libsvm_rs 

libsvm-rs

A pure Rust reimplementation of LIBSVM, targeting numerical equivalence and model-file compatibility with the original C++ library.

Status

Phases 0–4 complete: types, I/O, kernels, cache, prediction, full SMO solver, probability estimates (Platt scaling), and cross-validation. Training works for all 5 SVM types (C-SVC, ν-SVC, one-class, ε-SVR, ν-SVR). See train::svm_train for training, predict::predict for inference, and predict::predict_probability for probabilistic outputs.

Feature Flags

• rayon — Enable parallel cross-validation (off by default).

Re-exports

pub use error::SvmError;

pub use types::*;

Modules

cache

LRU kernel cache matching the original LIBSVM.

cross_validation

Cross-validation for SVM models.

error

io

I/O routines for LIBSVM problem and model files.

kernel

Kernel functions matching the original LIBSVM.

predict

Prediction functions matching the original LIBSVM.

probability

Probability estimation functions for SVM models.

qmatrix

Q matrix implementations for the SMO solver.

solver

SMO solver for the SVM dual problem.

train

SVM training pipeline.

types

Functions

set_quiet

Enable or disable quiet mode. When quiet, solver diagnostic messages are suppressed (equivalent to LIBSVM’s -q flag).