Enum ClassWeight 

pub enum ClassWeight<F> {
    Balanced,
    Dict(Vec<(usize, F)>),
}

Per-class weighting strategy, mirroring scikit-learn’s class_weight parameter (sklearn/linear_model/_logistic.py:1111, "class_weight": [dict, StrOptions({"balanced"}), None]).

None (the absence of this enum) means uniform weights. The two non-uniform modes both produce a per-class multiplier that is folded into the effective per-sample weight sample_weight[i] * class_weight[y[i]] (sklearn/linear_model/_logistic.py:312-313).

Variants

Balanced

'balanced': per-class weight n_samples / (n_classes * bincount[class]) (sklearn/utils/class_weight.py:73, recip_freq = len(y) / (len(le.classes_) * np.bincount(y_ind))).

Dict(Vec<(usize, F)>)

A user-supplied {class_label: weight} map. Classes absent from the map default to weight 1.0 (sklearn/utils/class_weight.py:77-83). Stored as (class_label, weight) pairs to preserve the F generic.

Trait Implementations

impl<F: Clone> Clone for ClassWeight<F>

fn clone(&self) -> ClassWeight<F>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Debug> Debug for ClassWeight<F>

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

Formats the value using the given formatter.

impl<F: PartialEq> PartialEq for ClassWeight<F>

fn eq(&self, other: &ClassWeight<F>) -> 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<F> StructuralPartialEq for ClassWeight<F>