Enum ScoringFunction 

pub enum ScoringFunction {
    FClassif,
    FRegression,
    Variance,
}

Pluggable scoring function used by SelectKBest.

Each variant defines a different univariate statistical test for ranking features.

Variants

FClassif

ANOVA F-value for classification (one-way between-groups F-test).

For each feature, groups samples by class label and computes between-class variance / within-class variance. Higher F means the feature is more discriminative. Requires target labels y.

FRegression

Univariate linear-regression F-statistic for regression.

For each feature j, computes the Pearson correlation r with the target, then F = r^2 * (n-2) / (1 - r^2). Higher F means the feature has a stronger linear relationship with the target. Requires target values y.

Variance

Feature variance (unsupervised).

Simply uses the variance of each feature as its score. Target y is ignored when this variant is used.

Trait Implementations

impl Clone for ScoringFunction

fn clone(&self) -> ScoringFunction

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ScoringFunction

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for ScoringFunction

fn deserialize<__D>( __deserializer: __D, ) -> Result<ScoringFunction, <__D as Deserializer<'de>>::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for ScoringFunction

fn serialize<__S>( &self, __serializer: __S, ) -> Result<<__S as Serializer>::Ok, <__S as Serializer>::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.