Trait std_dev::regression::binary_search::NVariableStorage

pub trait NVariableStorage: IndexMut<usize, Output = f64> + AsRef<[f64]> + AsMut<[f64]> + Clone {
    type Data;
    type Given<'a>: AsRef<[f64]>
       where Self: 'a;

    // Required methods
    fn new_filled(data: &Self::Data, num: f64) -> Self;
    fn borrow(&self) -> Self::Given<'_>;
}

Available on crate feature regression only.

A trait which allows storage of n-variable optimization, either on the stack through arrays ([f64; VARIABLE_COUNT]) or allocated on the heap through Vec.

Required Associated Types

type Data

Associated data for use in construction of this type. The number of arguments in case of using a Vec, nothing when using arrays, as we know their length in compile time.

type Given<'a>: AsRef<[f64]> where Self: 'a

The structure that’s given to the fitness function. Needs to have the same length as the .as_ref() implementation of this struct.

Required Methods

fn new_filled(data: &Self::Data, num: f64) -> Self

Creates a new storage filled with num.

fn borrow(&self) -> Self::Given<'_>

Borrow the current variables.

Implementations on Foreign Types

impl NVariableStorage for Vec<f64>

type Data = VariableLengthStorage

type Given<'a> = &'a [f64]

fn new_filled(data: &Self::Data, num: f64) -> Self

fn borrow(&self) -> Self::Given<'_>

impl<const LENGTH: usize> NVariableStorage for [f64; LENGTH]

type Data = ()

type Given<'a> = [f64; LENGTH]

fn new_filled(_data: &(), num: f64) -> Self

fn borrow(&self) -> Self::Given<'_>

Implementors