IS
- Type of submodels input.IA
- Type of submodels output (same as aggregator model input).O
- Type of aggregator model output.AM
- Type of aggregator model.public final class StackedModel<IS,IA,O,AM extends IgniteModel<IA,O>> extends Object implements IgniteModel<IS,O>
StackedDatasetTrainer
.
Model consisting of two layers:
1. Submodels layerSubmodels layer is a(IS -> IA)
. 2. Aggregator layer(IA -> O)
.
ModelsParallelComposition
of several models IS -> IA
each of them getting same input
IS
and produce own output, these outputs outputs [IA]
are combined into a single output with a given binary "merger" operator IA -> IA -> IA
. Result of merge
is then passed to the aggregator layer.
Aggregator layer consists of a model IA -> O
.Modifier and Type | Method and Description |
---|---|
void |
close() |
O |
predict(IS is)
Make a prediction for the specified input arguments.
|
GridGain In-Memory Computing Platform : ver. 8.9.15 Release Date : December 3 2024