Struct SvmModel 

pub struct SvmModel {
    pub param: SvmParameter,
    pub nr_class: usize,
    pub sv: Vec<Vec<SvmNode>>,
    pub sv_coef: Vec<Vec<f64>>,
    pub rho: Vec<f64>,
    pub prob_a: Vec<f64>,
    pub prob_b: Vec<f64>,
    pub prob_density_marks: Vec<f64>,
    pub sv_indices: Vec<usize>,
    pub label: Vec<i32>,
    pub n_sv: Vec<usize>,
}

A trained SVM model.

Produced by training, or loaded from a LIBSVM model file.

load_model validates model-file shape contracts before returning this type: class counts, support-vector counts, decision-function arrays, optional probability metadata, and sparse support-vector rows must be internally consistent. Manually constructed values bypass those checks, so callers that accept external model text should prefer crate::io::load_model or crate::io::load_model_from_reader.

Fields

param: SvmParameter

Parameters used during training.

nr_class: usize

Number of classes (2 for binary, >2 for multiclass, 2 for regression).

sv: Vec<Vec<SvmNode>>

Support vectors (sparse feature vectors).

sv_coef: Vec<Vec<f64>>

Support vector coefficients. For k classes, this is a (k-1) × num_sv matrix stored as Vec<Vec<f64>>.

rho: Vec<f64>

Bias terms (rho). One per class pair: k*(k-1)/2 values.

prob_a: Vec<f64>

Pairwise probability parameter A (Platt scaling). Empty if not trained with probability estimates.

prob_b: Vec<f64>

Pairwise probability parameter B (Platt scaling). Empty if not trained with probability estimates.

prob_density_marks: Vec<f64>

Probability density marks (for one-class SVM).

sv_indices: Vec<usize>

Original indices of support vectors in the training set (1-based).

label: Vec<i32>

Class labels (in the order used internally).

n_sv: Vec<usize>

Number of support vectors per class.

Implementations

impl SvmModel

pub fn svm_type(&self) -> SvmType

Return the SVM type used by the model.

pub fn class_count(&self) -> usize

Return number of classes.

pub fn labels(&self) -> &[i32]

Return class labels in internal one-vs-one order.

pub fn support_vector_indices(&self) -> &[usize]

Return original 1-based support-vector indices.

pub fn support_vector_count(&self) -> usize

Return total number of support vectors.

pub fn svr_probability(&self) -> Option<f64>

Return SVR sigma when a probability-capable SVR model is available.

pub fn has_probability_model(&self) -> bool

Check whether the model contains probability metadata.

Trait Implementations

impl Clone for SvmModel

fn clone(&self) -> SvmModel

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for SvmModel

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for SvmModel

Available on crate feature serde only.

fn deserialize<D>(deserializer: D) -> Result<Self, D::Error>

where D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for SvmModel

fn eq(&self, other: &SvmModel) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for SvmModel

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for SvmModel