Struct NelderMeadBuilder

pub struct NelderMeadBuilder { /* private fields */ }

Builder for NelderMead.

Implementations

impl NelderMeadBuilder

pub fn ulps(&mut self, value: i64) -> &mut Self

The required number of floating point representations that separate two numbers to consider them equal. See crate float_cmp for more information.

pub fn maxiter<VALUE: Into<Option<usize>>>(&mut self, value: VALUE) -> &mut Self

The maximum number of iterations to optimize. If neither maxiter nor maxfun are given, both default to n*200 where n is the number of parameters to optimize.

Examples found in repository

examples/simple.rs (line 14)

pub fn main() {
    let function =
        |x: ArrayView1<f64>| (1.0 - x[0]).powi(2) + 100.0 * (x[1] - x[0].powi(2)).powi(2);
    let minimizer = NelderMeadBuilder::default()
        .xtol(1e-6f64)
        .ftol(1e-6f64)
        .maxiter(50000) 
        .build()
        .unwrap();
    let args = Array::from_vec(vec![3.0, -8.3]);
    let ans = minimizer.minimize(&function, args.view());
    println!("Final optimized arguments: {}", ans);
}

pub fn maxfun<VALUE: Into<Option<usize>>>(&mut self, value: VALUE) -> &mut Self

The maximum number of function calls used to optimize. If neither maxiter nor maxfun are given, both default to n*200 where n is the number of parameters to optimize.

pub fn adaptive(&mut self, value: bool) -> &mut Self

Adapt algorithm parameters to dimensionality of the problem. Useful for high-dimensional minimization.

pub fn xtol(&mut self, value: f64) -> &mut Self

Absolute error in function parameters between iterations that is acceptable for convergence.

Examples found in repository

examples/simple.rs (line 12)

pub fn main() {
    let function =
        |x: ArrayView1<f64>| (1.0 - x[0]).powi(2) + 100.0 * (x[1] - x[0].powi(2)).powi(2);
    let minimizer = NelderMeadBuilder::default()
        .xtol(1e-6f64)
        .ftol(1e-6f64)
        .maxiter(50000) 
        .build()
        .unwrap();
    let args = Array::from_vec(vec![3.0, -8.3]);
    let ans = minimizer.minimize(&function, args.view());
    println!("Final optimized arguments: {}", ans);
}

pub fn ftol(&mut self, value: f64) -> &mut Self

Absolute error in function values between iterations that is acceptable for convergence.

Examples found in repository

examples/simple.rs (line 13)

pub fn main() {
    let function =
        |x: ArrayView1<f64>| (1.0 - x[0]).powi(2) + 100.0 * (x[1] - x[0].powi(2)).powi(2);
    let minimizer = NelderMeadBuilder::default()
        .xtol(1e-6f64)
        .ftol(1e-6f64)
        .maxiter(50000) 
        .build()
        .unwrap();
    let args = Array::from_vec(vec![3.0, -8.3]);
    let ans = minimizer.minimize(&function, args.view());
    println!("Final optimized arguments: {}", ans);
}

pub fn build(&self) -> Result<NelderMead, String>

Builds a new NelderMead.

Errors

If a required field has not been initialized.

Examples found in repository

examples/simple.rs (line 15)

pub fn main() {
    let function =
        |x: ArrayView1<f64>| (1.0 - x[0]).powi(2) + 100.0 * (x[1] - x[0].powi(2)).powi(2);
    let minimizer = NelderMeadBuilder::default()
        .xtol(1e-6f64)
        .ftol(1e-6f64)
        .maxiter(50000) 
        .build()
        .unwrap();
    let args = Array::from_vec(vec![3.0, -8.3]);
    let ans = minimizer.minimize(&function, args.view());
    println!("Final optimized arguments: {}", ans);
}

Trait Implementations

impl Clone for NelderMeadBuilder

fn clone(&self) -> NelderMeadBuilder

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Default for NelderMeadBuilder

fn default() -> NelderMeadBuilder

Returns the “default value” for a type.