Struct SparseMode 

pub struct SparseMode;

Sparse linear algebra mode.

Uses SparseColMat<usize, f64> for Jacobians and Hessians. Optimal for large-scale problems with sparse structure (e.g., pose graphs).

Trait Implementations

impl AssemblyBackend for SparseMode

fn assemble( problem: &Problem, variables: &HashMap<String, VariableEnum>, variable_index_map: &HashMap<String, usize>, symbolic_structure: Option<&SymbolicStructure>, _total_dof: usize, ) -> LinearizerResult<(Mat<f64>, SparseColMat<usize, f64>)>

Assemble residuals and Jacobian from the problem.

fn compute_column_norms(jacobian: &SparseColMat<usize, f64>) -> Vec<f64>

Compute column norms of the Jacobian (for Jacobi scaling).

fn apply_column_scaling( jacobian: &SparseColMat<usize, f64>, scaling: &[f64], ) -> SparseColMat<usize, f64>

Apply diagonal column scaling to the Jacobian. Returns a new Jacobian with columns scaled by 1 / (1 + norm).

fn apply_inverse_scaling(step: &Mat<f64>, scaling: &[f64]) -> Mat<f64>

Apply inverse scaling to a step vector: step_i *= scaling_i

fn hessian_vec_product( hessian: &SparseColMat<usize, f64>, vec: &Mat<f64>, ) -> Mat<f64>

Hessian-vector product: H * v (needed by DogLeg for Cauchy point)

impl LinearSolver<SparseMode> for IterativeSchurSolver

fn solve_normal_equation( &mut self, residuals: &Mat<f64>, jacobian: &SparseColMat<usize, f64>, ) -> LinAlgResult<Mat<f64>>

Solve the normal equations: (J^T · J) · dx = −J^T · r

fn solve_augmented_equation( &mut self, residuals: &Mat<f64>, jacobian: &SparseColMat<usize, f64>, lambda: f64, ) -> LinAlgResult<Mat<f64>>

Solve the augmented equations: (J^T · J + λI) · dx = −J^T · r

fn get_hessian(&self) -> Option<&SparseColMat<usize, f64>>

Get the cached Hessian matrix (J^T · J) from the last solve

fn get_gradient(&self) -> Option<&Mat<f64>>

Get the cached gradient vector (J^T · r) from the last solve

fn compute_covariance_matrix(&mut self) -> Option<&Mat<f64>>

Compute the covariance matrix (H^{-1}) by inverting the cached Hessian.

fn get_covariance_matrix(&self) -> Option<&Mat<f64>>

Get the cached covariance matrix (H^{-1}) computed from the Hessian.

impl LinearSolver<SparseMode> for SparseCholeskySolver

fn solve_normal_equation( &mut self, residuals: &Mat<f64>, jacobians: &SparseColMat<usize, f64>, ) -> LinAlgResult<Mat<f64>>

Solve the normal equations: (J^T · J) · dx = −J^T · r

fn solve_augmented_equation( &mut self, residuals: &Mat<f64>, jacobians: &SparseColMat<usize, f64>, lambda: f64, ) -> LinAlgResult<Mat<f64>>

Solve the augmented equations: (J^T · J + λI) · dx = −J^T · r

fn get_hessian(&self) -> Option<&SparseColMat<usize, f64>>

Get the cached Hessian matrix (J^T · J) from the last solve

fn get_gradient(&self) -> Option<&Mat<f64>>

Get the cached gradient vector (J^T · r) from the last solve

fn compute_covariance_matrix(&mut self) -> Option<&Mat<f64>>

Compute the covariance matrix (H^{-1}) by inverting the cached Hessian.

fn get_covariance_matrix(&self) -> Option<&Mat<f64>>

Get the cached covariance matrix (H^{-1}) computed from the Hessian.

impl LinearSolver<SparseMode> for SparseQRSolver

fn solve_normal_equation( &mut self, residuals: &Mat<f64>, jacobians: &SparseColMat<usize, f64>, ) -> LinAlgResult<Mat<f64>>

Solve the normal equations: (J^T · J) · dx = −J^T · r

fn solve_augmented_equation( &mut self, residuals: &Mat<f64>, jacobians: &SparseColMat<usize, f64>, lambda: f64, ) -> LinAlgResult<Mat<f64>>

Solve the augmented equations: (J^T · J + λI) · dx = −J^T · r

fn get_hessian(&self) -> Option<&SparseColMat<usize, f64>>

Get the cached Hessian matrix (J^T · J) from the last solve

fn get_gradient(&self) -> Option<&Mat<f64>>

Get the cached gradient vector (J^T · r) from the last solve

fn compute_covariance_matrix(&mut self) -> Option<&Mat<f64>>

Compute the covariance matrix (H^{-1}) by inverting the cached Hessian.

fn get_covariance_matrix(&self) -> Option<&Mat<f64>>

Get the cached covariance matrix (H^{-1}) computed from the Hessian.

impl LinearSolver<SparseMode> for SparseSchurComplementSolver

fn solve_normal_equation( &mut self, residuals: &Mat<f64>, jacobian: &SparseColMat<usize, f64>, ) -> LinAlgResult<Mat<f64>>

Solve the normal equations: (J^T · J) · dx = −J^T · r

fn solve_augmented_equation( &mut self, residuals: &Mat<f64>, jacobian: &SparseColMat<usize, f64>, lambda: f64, ) -> LinAlgResult<Mat<f64>>

Solve the augmented equations: (J^T · J + λI) · dx = −J^T · r

fn get_hessian(&self) -> Option<&SparseColMat<usize, f64>>

Get the cached Hessian matrix (J^T · J) from the last solve

fn get_gradient(&self) -> Option<&Mat<f64>>

Get the cached gradient vector (J^T · r) from the last solve

fn compute_covariance_matrix(&mut self) -> Option<&Mat<f64>>

Compute the covariance matrix (H^{-1}) by inverting the cached Hessian.

fn get_covariance_matrix(&self) -> Option<&Mat<f64>>

Get the cached covariance matrix (H^{-1}) computed from the Hessian.

impl LinearizationMode for SparseMode

type Jacobian = SparseColMat<Own<usize, f64>>

The Jacobian matrix type (SparseColMat or Mat)

type Hessian = SparseColMat<Own<usize, f64>>

The Hessian matrix type (SparseColMat or Mat)