Expand description
Model Evaluator for standardized evaluation and comparison
Provides ModelEvaluator for running comprehensive evaluations, comparing multiple models, and generating leaderboards.
Structs§
- Eval
Config - Configuration for model evaluation
- Eval
Result - Model evaluation results
- KFold
- K-Fold cross-validation splitter
- Leaderboard
- Leaderboard for comparing multiple models
- Model
Evaluator - Model Evaluator for running evaluations
Enums§
- Metric
- Available evaluation metrics
- Rouge
Variant - ROUGE variant for text generation evaluation