Member Junction
    Preparing search index...

    POST /train response. Carries the serialized model plus the fitted preprocessing parameters (the anti-skew payload) and the deterministic metrics used to grade and rank the run.

    interface TrainResponse {
        artifact_b64: string;
        duration_sec: number;
        feature_importance: FeatureImportance;
        fitted_preprocessing: FittedPreprocessing;
        holdout_metrics?: ModelMetrics;
        metrics: ModelMetrics;
        training_row_count: number;
    }
    Index

    Properties

    artifact_b64: string

    Base64-encoded serialized model artifact.

    duration_sec: number

    Wall-clock training time in seconds.

    feature_importance: FeatureImportance

    Per-feature importance/contribution.

    fitted_preprocessing: FittedPreprocessing

    Fitted preprocessing parameters (means/stds, vocabularies, bin edges, fill values) learned during /train. Travels with the model and is replayed at inference — never re-fit (plan §6.2).

    holdout_metrics?: ModelMetrics

    Honest metrics on the locked holdout, scored exactly once (plan §8.2).

    metrics: ModelMetrics

    Train + validation metrics.

    training_row_count: number

    Number of training rows the model was fit on.