Enum SolverType

pub enum SolverType {
    L2R_LR = 0,
    L2R_L2LOSS_SVC_DUAL = 1,
    L2R_L2LOSS_SVC = 2,
    L2R_L1LOSS_SVC_DUAL = 3,
    MCSVM_CS = 4,
    L1R_L2LOSS_SVC = 5,
    L1R_LR = 6,
    L2R_LR_DUAL = 7,
    L2R_L2LOSS_SVR = 11,
    L2R_L2LOSS_SVR_DUAL = 12,
    L2R_L1LOSS_SVR_DUAL = 13,
}

Types of generalized linear models supported by liblinear.

These combine several types of regularization schemes:

• L1
• L2

…and loss functions:

• L1-loss for SVM
• Regular L2-loss for SVM (hinge-loss)
• Logistic loss for logistic regression

Variants

L2R_LR = 0

L2-regularized logistic regression (primal).

L2R_L2LOSS_SVC_DUAL = 1

L2-regularized L2-loss support vector classification (dual).

L2R_L2LOSS_SVC = 2

L2-regularized L2-loss support vector classification (primal).

L2R_L1LOSS_SVC_DUAL = 3

L2-regularized L1-loss support vector classification (dual).

MCSVM_CS = 4

Support vector classification by Crammer and Singer.

L1R_L2LOSS_SVC = 5

L1-regularized L2-loss support vector classification.

L1R_LR = 6

L1-regularized logistic regression.

L2R_LR_DUAL = 7

L2-regularized logistic regression (dual).

L2R_L2LOSS_SVR = 11

L2-regularized L2-loss support vector regression (primal).

L2R_L2LOSS_SVR_DUAL = 12

L2-regularized L2-loss support vector regression (dual).

L2R_L1LOSS_SVR_DUAL = 13

L2-regularized L1-loss support vector regression (dual).

Implementations

impl SolverType

pub fn is_logistic_regression(&self) -> bool

Returns true if the solver is a probabilistic/logistic regression solver.

Supported solvers: L2R_LR, L1R_LR, L2R_LR_DUAL.

pub fn is_support_vector_regression(&self) -> bool

Returns true if the solver is a support vector regression solver.

Supported solvers: L2R_L2LOSS_SVR, L2R_L2LOSS_SVR_DUAL, L2R_L1LOSS_SVR_DUAL.

pub fn is_multi_class_classification(&self) -> bool

Returns true if the solver supports multi-class classification.

Supported solvers: All non-SVR solvers.

Trait Implementations

impl Default for SolverType

fn default() -> Self

Default: L2R_LR