DataPrunerOptions
Options & Configuration · AiDotNet.Data.Quality
Configuration options for data pruning.
How It Works
Data pruning removes easy or redundant samples based on training signals (e.g., loss, confidence, forgetting events), keeping only the most informative examples.
Constructors
| Constructor | Summary |
|---|---|
DataPrunerOptions() |
Properties
| Property | Summary |
|---|---|
PruneStrategy Strategy { get; set; } |
Strategy for selecting samples to prune. |
double PruneRatio { get; set; } |
Fraction of data to prune (remove). |
int MinEpochsForScoring { get; set; } |
Minimum number of epochs before pruning scores are reliable. |
Methods
| Method | Summary |
|---|---|
void Validate() |
Validates that all option values are within acceptable ranges. |