Ensemble Strategy

Combine multiple operators with aggregation for ensemble processing.

See Also

• Operators Overview - All operator types
• Composite Operator - Operator composition
• Parallel Strategy - Parallel execution
• Merging Strategy - Merge results

---

datarax.operators.strategies.ensemble

Ensemble composition strategies.

logger module-attribute

logger = getLogger(__name__)

EnsembleStrategy

EnsembleStrategy(mode: str)

Bases: CompositionStrategyImpl

Applies operators in parallel and reduces outputs (mean, sum, etc).

Parameters:

Name	Type	Description	Default
mode	str	Reduction mode ("mean", "sum", "max", "min")	required

mode instance-attribute

mode = mode

apply

apply(operators: list[OperatorModule], context: StrategyContext) -> tuple[PyTree, PyTree, dict[str, Any]]

Apply operators in parallel and reduce outputs element-wise.

Parameters:

Name	Type	Description	Default
operators	list[OperatorModule]	Operators to execute on identical input.	required
context	StrategyContext	Execution context with input data, state, and RNG params.	required

Returns:

Type	Description
tuple[PyTree, PyTree, dict[str, Any]]	Tuple of (reduced_data, last_state, last_metadata).

Raises:

Type	Description
ValueError	If self.mode is not one of mean/sum/max/min.

describe

describe() -> dict[str, Any]

Return a serializable description of this strategy.