Struct Random 

pub struct Random {
    pub seed: u64,
}

Uniform in every knob, independently.

The baseline and not a straw man: over a space where few knobs matter, random search beats a grid, which spends its budget re-testing the ones that do not (Bergstra and Bengio, 2012). It is what Tpe falls back to before it has anything to learn from. Its point is a function of the seed and the index, so two machines drawing trial 7 draw the same point.

Fields

seed: u64

The seed. There is no “unseeded”: a search you cannot re-run is a result you cannot check.

Implementations

impl Random

pub fn ask( &self, space: &Space, trial: usize, _seen: &[(Point, Option<f64>)], ) -> Option<Point>

The trial-th point. It never runs out, and it never looks at what the finished trials did — which is the whole of what it is.

Trait Implementations

impl Clone for Random

fn clone(&self) -> Random

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Random

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl From<Random> for Sampler

fn from(how: Random) -> Self

Converts to this type from the input type.

impl Hash for Random

fn hash<__H: Hasher>(&self, state: &mut __H)

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)

where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl PartialEq for Random

fn eq(&self, other: &Random) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Eq for Random

impl StructuralPartialEq for Random