orthos

High-performance linear algebra library written in Rust.

Installation

[dependencies]
orthos = "0.1"

Quick Start

use orthos::{Matrix, Vector};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Create matrices
    let a = Matrix::from_rows(vec![
        vec![1.0, 2.0],
        vec![3.0, 4.0],
    ])?;
    let b = Matrix::identity(2);

    // Matrix operations
    let c = a.matmul(&b)?;       // Matrix multiplication
    let d = a.add(&b)?;          // Addition
    let e = a.transpose();       // Transpose
    let f = a.inverse()?;        // Inverse

    // Properties
    println!("Determinant: {}", a.determinant()?);  // -2.0
    println!("Trace: {}", a.trace());               // 5.0
    println!("Shape: {:?}", a.shape());             // (2, 2)

    // Element access
    println!("a[0,0] = {}", a.get(0, 0)?);

    // Vectors
    let v = Vector::new(vec![1.0, 2.0, 3.0]);
    let w = Vector::new(vec![4.0, 5.0, 6.0]);

    println!("Dot product: {}", v.dot(&w)?);   // 32.0
    println!("Norm: {}", v.norm());             // 3.7416...
    let normalized = v.normalize()?;            // Unit vector

    // Matrix-vector multiplication
    let result = orthos::ops::matvec(&a, &Vector::new(vec![1.0, 1.0]))?;

    Ok(())
}

Features

Matrix Operations

• Creation: zeros, ones, identity, from_rows, new
• Arithmetic: add, sub, mul_elementwise, matmul, scale
• Methods: transpose, inverse, determinant, trace
• Access: get, set, shape, rows, cols

Vector Operations

• Creation: new, zeros, ones
• Arithmetic: add, sub, scale
• Methods: dot, norm, norm_squared, normalize
• Access: get, set, len, is_empty

Matrix-Vector Operations

• orthos::ops::matvec(&matrix, &vector) - Matrix-vector multiplication
• orthos::ops::matmul(&a, &b) - Matrix multiplication

Error Handling

All fallible operations return Result types:

use orthos::{Matrix, MatrixError};

let result: Result<Matrix, MatrixError> = Matrix::from_rows(vec![
    vec![1.0, 2.0],
    vec![3.0],  // Mismatched length - will error
]);

match result {
    Ok(m) => println!("Created matrix"),
    Err(e) => println!("Error: {}", e),
}

Error types:

• MatrixError::DimensionMismatch - Incompatible dimensions
• MatrixError::IndexOutOfBounds - Invalid index
• MatrixError::SingularMatrix - Cannot invert singular matrix
• MatrixError::InvalidInput - Invalid input data
• VectorError::DimensionMismatch - Incompatible vector lengths
• VectorError::IndexOutOfBounds - Invalid index

Feature Flags

# With Python bindings
[dependencies]
orthos = { version = "0.1", features = ["python"] }

Performance

Built on faer, optimized for:

• Cache-friendly memory layouts
• SIMD vectorization
• Multi-threaded operations for large matrices

Links

• Documentation
• Source Code
• Issue Tracker
• Python package

License

Apache-2.0