Interface IModelAutoTrainer

Seam that turns "this model needs training" into a published, scoreable model. Production: PipelineAutoTrainer (train via the real TrainingEngine, publish via the promotion gate, repoint the Record Process + bindings). Injected so the processor stays free of training dependencies and is unit-testable.

interface IModelAutoTrainer {
    EnsureTrainedModel(
        request: AutoTrainRequest,
        contextUser: UserInfo,
        provider: IMetadataProvider,
    ): Promise<string>;
}

Implemented by

PipelineAutoTrainer

Index

Methods

EnsureTrainedModel

Methods

EnsureTrainedModel

EnsureTrainedModel(
    request: AutoTrainRequest,
    contextUser: UserInfo,
    provider: IMetadataProvider,
): Promise<string>
Ensure a published, artifact-backed model exists for the pipeline and that the Record Process + its bindings point at it. Returns that model's id.
Parameters
- request: AutoTrainRequest
- contextUser: UserInfo
- provider: IMetadataProvider
Returns Promise<string>
Throws
when training or publishing fails (e.g. leakage-flagged model needing sign-off)
- Defined in packages/AI/PredictiveStudio/Engine/src/scoring/types.ts:215

Interface IModelAutoTrainer

Implemented by

Index

Methods

Methods

EnsureTrainedModel

Parameters

Returns Promise<string>

Throws

Settings

On This Page