Struct RandomEffectTermSpec 

pub struct RandomEffectTermSpec {
    pub name: String,
    pub feature_col: usize,
    pub drop_first_level: bool,
    pub penalized: bool,
    pub frozen_levels: Option<Vec<u64>>,
}

Random-effects term specification.

The selected feature column is interpreted as a categorical grouping variable. The term contributes a one-hot dummy block with an identity penalty on group coefficients, equivalent to i.i.d. Gaussian random effects.

Fields

name: String

feature_col: usize

drop_first_level: bool

If true, drop the lexicographically first group level to use treatment coding. If false, keep all levels (full one-hot block, still identifiable under ridge).

penalized: bool

If true, add a ridge penalty and estimate this block as a random effect. If false, leave the one-hot/treatment-coded block unpenalized so it is a fixed categorical main effect. The default preserves older saved models.

frozen_levels: Option<Vec<u64>>

Optional fixed kept-level set (sorted by f64 bit pattern) captured at fit time. When present, prediction uses exactly these columns to avoid design drift.

Trait Implementations

impl Clone for RandomEffectTermSpec

fn clone(&self) -> RandomEffectTermSpec

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RandomEffectTermSpec

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for RandomEffectTermSpec

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for RandomEffectTermSpec

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

where __S: Serializer,

Serialize this value into the given Serde serializer.