Model ensemble

Run independent predictors and combine them through a versioned aggregation rule.

prepare input→predict Apredict Bpredict C→combine predictions→evaluate ensemble→publish

Model ensemble

Activity responsibilities

Activity	Responsibility
prepare-input	Execute the prepare input stage and publish its declared outputs for downstream activities.
predict-a	Execute the predict A stage and publish its declared outputs for downstream activities.
predict-b	Execute the predict B stage and publish its declared outputs for downstream activities.
predict-c	Execute the predict C stage and publish its declared outputs for downstream activities.
combine-predictions	Execute the combine predictions stage and publish its declared outputs for downstream activities.
evaluate-ensemble	Execute the evaluate ensemble stage and publish its declared outputs for downstream activities.
publish	Execute the publish stage and publish its declared outputs for downstream activities.

Inputs

• Input dataset
• Component models
• Aggregation policy

Outputs

• Component predictions
• Ensemble output
• Evaluation

Provenance to preserve

• Component versions
• Aggregation weights
• Per-model logs

Runnable example

This is the complete checked-in bundle for this pattern. Download the environment, scope, topology, workflow, input, container recipe, runner, and validation contract from this page before executing it.

Read setup instructions ↓Environment YAML ↓Execution scope YAML ↓Network topology YAML ↓Workflow YAML ↓Input data ↓Container recipe ↓Run the workflow ↓Validate outputs ↓Expected output manifest ↓

Verified local execution

These captures and the output manifest were produced by the fixture's local Docker run and validator. They are published with the same bundle as the runnable files.

Workflow evidence

Execution evidence

Output evidence

Open verified output manifest ↓

Execution considerations

Components may run in different environments. Transfer predictions when co-locating incompatible frameworks is more expensive.

AkôFlow boundary: the engine schedules, deploys, executes, transfers data, and records evidence. The ML or agent framework remains an implementation choice inside each activity.