Expand description
ModelEnsemble — Multi-model ensemble aggregator for distributed inference.
Supports voting, averaging, and stacking strategies with per-member tracking, weight normalization, and statistical disagreement measurement.
Structs§
- Ensemble
Config - Configuration for
ModelEnsemble. - Ensemble
Result - The aggregated output of the ensemble.
- Ensemble
Stats - Aggregate statistics over all ensemble members.
- Model
Ensemble - Multi-model ensemble aggregator supporting voting, averaging, and stacking.
- Model
Member - Metadata for a single member of the ensemble.
- Model
Prediction - A single model’s prediction package.
Enums§
- Ensemble
Error - Errors that can arise during ensemble aggregation or configuration.
- Ensemble
Strategy - Aggregation strategy used by the ensemble.