Struct CsrMatrix 

pub struct CsrMatrix<T>(/* private fields */);

Compressed Sparse Row (CSR) Matrix.

CSR format is optimized for row-wise operations and matrix-vector products. It stores non-zero elements row by row, making it efficient for:

• Row access and iteration
• Sparse matrix-vector multiplication
• Row-based computations

Format

The CSR format uses three arrays:

• values: Non-zero values stored row by row
• col_indices: Column indices for each non-zero value
• row_offsets: Starting index in values/col_indices for each row

Examples

use algebra_sparse::CsrMatrix;
use nalgebra::DMatrix;

// Create from dense matrix
let dense = DMatrix::from_row_slice(2, 3, &[
    1.0, 0.0, 2.0,
    0.0, 3.0, 0.0,
]);
let csr = CsrMatrix::from_dense(dense.as_view());

// Get row as sparse vector
let row = csr.as_view().get_row(0);
for (col, val) in row.iter() {
    println!("({}, {})", col, val);
}

Implementations

impl<T> CsrMatrix<T>

where T: Real,

pub fn from_dense(dense_mat: DMatrixView<'_, T>) -> Self

Creates a CSR matrix from a dense matrix.

Values below the zero threshold are automatically filtered out. The zero threshold is obtained from the Real trait implementation.

Arguments

• dense_mat - The dense matrix to convert

Examples

use algebra_sparse::CsrMatrix;
use nalgebra::DMatrix;

let dense = DMatrix::from_row_slice(2, 2, &[1.0, 0.0, 0.0, 2.0]);
let csr = CsrMatrix::from_dense(dense.as_view());

pub fn new(ncols: usize) -> Self

Creates a new empty CSR matrix with the given number of columns.

Arguments

• ncols - Number of columns in the matrix

Examples

use algebra_sparse::CsrMatrix;

let mut csr: CsrMatrix<f64> = CsrMatrix::new(3);
// Now you can add rows using new_row_builder()

pub fn reset(&mut self, ncols: usize)

Resets the CSR matrix to empty state with the given number of columns.

This clears all data and allows reuse of the matrix.

Arguments

• ncols - New number of columns

pub fn clear(&mut self)

Clears all rows in the CSR matrix while preserving the number of columns.

This removes all rows but keeps the matrix ready for reuse.

pub fn new_row_builder(&mut self, zero_threshold: T) -> CsVecBuilder<'_, T>

Creates a new row builder for adding elements to the next row.

The builder allows efficient construction of sparse rows. When the builder is dropped, the row is finalized and added to the matrix.

Arguments

• zero_threshold - Values below this threshold are filtered out

Examples

use algebra_sparse::CsrMatrix;

let mut csr = CsrMatrix::new(3);
let mut builder = csr.new_row_builder(1e-10);
builder.push(0, 1.0);  // Add element at column 0
builder.push(2, 2.0);  // Add element at column 2
// Row is automatically added when builder is dropped

pub fn get_row_mut(&mut self, row_index: usize) -> CsVecMut<'_, T>

Gets a mutable row as a sparse vector.

Arguments

• row_index - Index of the row to retrieve

Returns

A CsVecMut allowing modification of the row’s values

Note

This only allows modification of existing values, not structural changes.

pub fn as_view(&self) -> CsrMatrixView<'_, T>

Returns an immutable view of this CSR matrix.

The view allows efficient read-only access without allocation and can be used for matrix operations.

Trait Implementations

impl<T: Clone> Clone for CsrMatrix<T>

fn clone(&self) -> CsrMatrix<T>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<T: Default> Default for CsrMatrix<T>

fn default() -> CsrMatrix<T>

Returns the “default value” for a type.

impl<'a, T> IntoView for &'a CsrMatrix<T>

where T: Real,

type View = CsrMatrixView<'a, T>

The view type produced by this trait.

fn into_view(self) -> Self::View

Converts the type into an immutable view.