Trait Distance

Source

pub trait Distance {
    // Required method
    fn as_components(&self) -> impl Iterator<Item = &f64>;

    // Provided methods
    fn square_err(&self, rhs: &impl Distance) -> f64 { ... }
    fn l0_err(&self, rhs: &impl Distance) -> f64 { ... }
    fn l1_err(&self, rhs: &impl Distance) -> f64 { ... }
    fn l2_err(&self, rhs: &impl Distance) -> f64 { ... }
    fn lp_err(&self, rhs: &impl Distance, p: f64) -> f64 { ... }
    fn linf_err(&self, rhs: &impl Distance) -> f64 { ... }
    fn l0_norm(&self) -> f64 { ... }
    fn l1_norm(&self) -> f64 { ... }
    fn l2_norm(&self) -> f64 { ... }
    fn linf_norm(&self) -> f64 { ... }
    fn lp_norm(&self, p: f64) -> f64 { ... }
}

Expand description

Be able to calculate the distance between two instances.

Each data number has the same weight by default, but you can change it by implementing Distance::as_components().

Required Methods§

Source

fn as_components(&self) -> impl Iterator<Item = &f64>

Get the components of the data.

The Iterator::size_hint() method is used to determine the size of the data.

Provided Methods§

Source

fn square_err(&self, rhs: &impl Distance) -> f64

Calculate the square error.

Source

fn l0_err(&self, rhs: &impl Distance) -> f64

Calculate the L0 norm of the error.

This method is also called Hamming distance, which counts the number of errors that are less than the epsilon.

Source

fn l1_err(&self, rhs: &impl Distance) -> f64

Calculate the L1 norm of the error.

This method is also called Manhattan distance.

Source

fn l2_err(&self, rhs: &impl Distance) -> f64

Calculate the L2 norm of the error.

This method is also called Euclidean distance.

Source

fn lp_err(&self, rhs: &impl Distance, p: f64) -> f64

Calculate the Lp norm of the error.

This method is also called Minkowski distance.

Source

fn linf_err(&self, rhs: &impl Distance) -> f64

Calculate the Linf norm of the error.

This method is also called Chebyshev distance.

Source