Crate optimize

This crate provides (non-linear) numerical optimization methods.

It is heavily based on scipy.optimize.

The crate is actively developed and expanded to include more methods.

A simple example follows:

// Define a function that we aim to minimize
let function = |x: ArrayView1<f64>| (1.0 - x[0]).powi(2) + 100.0 * (x[1] - x[0].powi(2)).powi(2);

// Create a minimizer using the builder pattern. If some of the parameters are not given, default values are used.
let minimizer = NelderMeadBuilder::default()
    .xtol(1e-6f64)
    .ftol(1e-6f64)
    .maxiter(50000)
    .build()
    .unwrap();

// Set the starting guess
let args = Array::from_vec(vec![3.0, -8.3]);

// Run the optimization
let ans = minimizer.minimize(&function, args.view());

// Print the optimized values
println!("Final optimized arguments: {}", ans);

Structs

NelderMead	A minimizer for a scalar function of one or more variables using the Nelder-Mead algorithm.
NelderMeadBuilder	Builder for NelderMead.

Traits

Minimizer

A general minimizer trait.