Member Junction
    Preparing search index...

    The frozen inference contract resolved off a loaded MJ: ML Models row — the artifact bytes plus everything the sidecar needs to apply (never re-fit) the model. Cached warm per processor run and reused across every record in a batch.

    interface LoadedModel {
        artifactB64: string;
        asOf: AsOfStrategy;
        featureSchema: FeatureSchemaEntry[];
        featureSteps: FeatureStepGraph;
        fittedPreprocessing: FittedPreprocessing;
        leakageGuard: LeakageGuard;
        modelId: string;
        problemType: ProblemType;
        sourceBindings: SourceBinding[];
        targetEntityName: string;
        targetVariable: string;
    }
    Index

    Properties

    artifactB64: string

    Base64 of the serialized artifact, ready for the sidecar PredictRequest.

    Point-in-time strategy frozen on the model's pipeline (assembly input).

    featureSchema: FeatureSchemaEntry[]

    The model's frozen, ordered feature schema (the inference input contract).

    featureSteps: FeatureStepGraph

    The frozen FeatureStep DAG (assembly input).

    fittedPreprocessing: FittedPreprocessing

    Frozen fitted-preprocessing params — passed through to /predict unchanged (§6.2).

    leakageGuard: LeakageGuard

    Leakage guard frozen on the model's pipeline (assembly input).

    modelId: string

    The model primary-key id (cache key + lineage).

    problemType: ProblemType

    What the model predicts (classification / regression).

    sourceBindings: SourceBinding[]

    Ordered source bindings the model's features draw from (assembly input).

    targetEntityName: string

    Target entity whose records this model scores (training-unit entity).

    targetVariable: string

    The label/target column name (diagnostics + lineage; never a feature).