Parallelized using rayon with support for many common datatypes, sukker tries to make matrix operations easier for the user, while still giving you as the user the performance you deserve.

Regular matrices have many features already ready, while Sparse ones have most of them. Whenever you want to switch from one to the other, just call from_dense, or from_sparse to quickly and easily convert!

Need a feature? Please let me/us know!

Even have custom declarative macros to create hashmap for your sparse matrices!

use sukker::{LinAlgFloats, Matrix};

fn main() {
    let a = Matrix::<f32>::randomize((8, 56));
    let b = Matrix::<f32>::randomize((56, 8));

    let c = a.matmul(&b).unwrap();

    let res = c.sin().exp(3).unwrap().pow(2).add_val(4.0).abs();

    // To print this beautiful matrix:
    res.print(5);
}

use std::collections::HashMap;
use sukker::{SparseMatrix, SparseMatrixData};

fn main() {
    let indexes: SparseMatrixData<f64> = smd![
        ((0, 1), 2.0), 
        ((1, 0), 4.0), 
        ((2, 3), 6.0), 
        ((3, 3), 8.0)
    ];

    let sparse = SparseMatrix::<f64>::new(indexes, (4, 4));

    sparse.print(3);
}

More examples can be found here

Full API documentation can be found here.

• Easy to use!
• Blazingly fast
• Linear Algebra module fully functional on f32 and f64
• Optimized matrix multiplication for both sparse and dense matrices
• Easily able to convert between sparse and dense matrices
• Serde support
• Support for all signed numeric datatypes
• Can be sent over threads
• Sparse matrices