Enum TaskRegularization 

pub enum TaskRegularization {
    None,
    L2 {
        beta: Float,
    },
    L1 {
        beta: Float,
    },
    NuclearNorm {
        beta: Float,
    },
    GroupSparsity {
        beta: Float,
    },
    Custom {
        beta: Float,
        regularizer: fn(&Array2<Float>) -> Float,
    },
}

Cross-task regularization strategies for multi-task learning

Variants

None

No cross-task regularization (independent tasks)

L2

L2 regularization on task weight differences

Fields

beta: Float

L1

L1 regularization promoting sparsity across tasks

Fields

beta: Float

NuclearNorm

Nuclear norm regularization on weight matrix (low-rank)

Fields

beta: Float

GroupSparsity

Group sparsity regularization

Fields

beta: Float

Custom

Custom regularization function

Fields

beta: Float

regularizer: fn(&Array2<Float>) -> Float

Trait Implementations

impl Clone for TaskRegularization

fn clone(&self) -> TaskRegularization

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TaskRegularization

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

Formats the value using the given formatter.

impl Default for TaskRegularization

fn default() -> Self

Returns the “default value” for a type.