Enum Precision

pub enum Precision<T> {
    Exact(T),
    Inexact(T),
}

A statistic has a precision Exact or Inexact. This represents uncertainty in that value. Exact values are computed, where can inexact values are likely inferred from compute functions.

Inexact statistics form a range of possible values that the statistic could be. This is statistic specific, for max this will be an upper bound. Meaning that the actual max in an array is guaranteed to be less than or equal to the inexact value, but equal to the exact value.

Variants

Exact(T)

Inexact(T)

Implementations

impl<T> Precision<Option<T>>

pub fn transpose(self) -> Option<Precision<T>>

Transpose the Option<Precision<T>> into Option<Precision<T>>.

impl<T> Precision<T>

where T: Copy,

pub fn to_inexact(&self) -> Self

impl<T> Precision<T>

pub fn exact<S: Into<T>>(s: S) -> Precision<T>

Creates an exact value

pub fn inexact<S: Into<T>>(s: S) -> Precision<T>

Creates an inexact value

pub fn as_ref(&self) -> Precision<&T>

Pushed the ref into the Precision enum

pub fn into_inexact(self) -> Self

Converts self into an inexact bound

pub fn as_exact(self) -> Option<T>

Returns the exact value from the bound, if that value is inexact, otherwise None.

pub fn as_inexact(self) -> Option<T>

Returns the exact value from the bound, if that value is inexact, otherwise None.

pub fn is_exact(&self) -> bool

True iff self == Exact(_)

pub fn map<U, F: FnOnce(T) -> U>(self, f: F) -> Precision<U>

Map the value of either precision value

pub fn zip<U>(self, other: Precision<U>) -> Precision<(T, U)>

Zip two Precision values into a tuple, keeping the inexactness if any.

pub fn try_map<U, F: FnOnce(T) -> VortexResult<U>>( self, f: F, ) -> VortexResult<Precision<U>>

Similar to map but handles fucntions that can fail.

impl Precision<ScalarValue>

pub fn into_scalar(self, dtype: DType) -> Precision<Scalar>

impl Precision<&ScalarValue>

pub fn into_scalar(self, dtype: DType) -> Precision<Scalar>

impl<T> Precision<T>

This allows a stat with a Precision to be interpreted as a bound.

pub fn bound<S: StatType<T>>(self) -> S::Bound

Applied the stat associated bound to the precision value

Trait Implementations

impl<T> Clone for Precision<T>

where T: Clone,

fn clone(&self) -> Self

Returns a copy of the value.

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

Performs copy-assignment from source.

impl<T: Debug> Debug for Precision<T>

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

Formats the value using the given formatter.

impl<T: Display> Display for Precision<T>

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

Formats the value using the given formatter.

impl<T: PartialEq> PartialEq<T> for Precision<T>

fn eq(&self, other: &T) -> 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<T: PartialEq> PartialEq for Precision<T>

fn eq(&self, other: &Precision<T>) -> 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<T: PartialOrd + Clone> StatBound<T> for Precision<T>

fn lift(value: Precision<T>) -> Self

Creates a new bound from a Precision statistic.

fn union(&self, other: &Self) -> Option<Self>

Finds the smallest bound that covers both bounds. A.k.a. the meet of the bound.

fn intersection(&self, other: &Self) -> Option<IntersectionResult<Self>>

Refines the bounds to the most precise estimate we can make for that bound. If the bounds are disjoint, then the result is None. e.g. Precision::Inexact(5) and Precision::Exact(6) would result in Precision::Inexact(5). A.k.a. the join of the bound.

fn to_exact(&self) -> Option<&T>

Returns the exact value from the bound if that value is exact, otherwise None.

fn into_value(self) -> Precision<T>

Converts Self back to Precision<T>, inverse of lift.

impl<T: Eq> Eq for Precision<T>

impl<T> StructuralPartialEq for Precision<T>