Struct ActiveLearningConfig

pub struct ActiveLearningConfig<T> {
    pub max_samples_per_iteration: usize,
    pub total_budget: usize,
    pub exploration_weight: T,
    pub min_sample_distance: T,
    pub convergence_threshold: T,
    pub uncertainty_weight: T,
    pub error_weight: T,
    pub gradient_weight: T,
}

Parameters controlling active learning behavior

Fields

max_samples_per_iteration: usize

Maximum number of new samples to suggest in one iteration

total_budget: usize

Total sampling budget (maximum number of samples)

exploration_weight: T

Weight for exploration vs exploitation (0.0 = pure exploitation, 1.0 = pure exploration)

min_sample_distance: T

Minimum distance between new sampling points (to avoid clustering)

convergence_threshold: T

Threshold for stopping active learning (when improvement is below this)

uncertainty_weight: T

Weight for uncertainty in combined strategies

error_weight: T

Weight for error estimates in combined strategies

gradient_weight: T

Weight for gradient estimates in combined strategies

Trait Implementations

impl<T: Clone> Clone for ActiveLearningConfig<T>

fn clone(&self) -> ActiveLearningConfig<T>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<T: Debug> Debug for ActiveLearningConfig<T>

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

Formats the value using the given formatter.

impl<T: Float + FromPrimitive> Default for ActiveLearningConfig<T>

fn default() -> Self

Returns the “default value” for a type.