IParameterChunkSource<T>
Interfaces · AiDotNet.Models.Parameters
Exposes the complete persistent state as role-aware chunks without conflating it with the trainable-only optimizer view.
Known implementations: CausalModelBase<T>, ClassifierBase<T>, ClusteringBase<T>, DiffusionModelBase<T>, LayerBase<T>, LinearProjector<T>, MLPProjector<T>, MaskHead<T>, ModelBase<T, TInput, TOutput>, MultiLabelClassifierBase<T>, NeuralNetworkBase<T>, OnlineLearningModelBase<T>, PrototypeMaskHead<T>, RPN<T>, ReinforcementLearningAgentBase<T>, ResizableTensorFieldParameterSource<T>, SurvivalModelBase<T>, SymmetricProjector<T>, TensorFieldParameterSource<T>, TensorListParameterSource<T>, TimeSeriesModelBase<T>, TimestepDependentLora<T>, VAEModelBase<T>
Methods
| Method | Summary |
|---|---|
IEnumerable<ParameterChunk<T>> GetParameterStateChunks() |
Yields state chunks in the exact order used by count, flat read, and restore. |