Type Alias ExperimentVersion

An AutoML experiment version. This is a configuration for training models within an experiment.

type ExperimentVersion = {
    algorithms?: ModelAlgorithm[];
    createdAt: CreatedAt;
    createdByUserId: OwnerId;
    dataSetId: DataSetId;
    datasetOrigin?: DatasetOrigin;
    dateIndexes?: DateIndexes;
    errorMessage?: ErrorMessage;
    errors?: Errors;
    experimentId: ExperimentId;
    experimentMode?: ExperimentMode;
    experimentType: ExperimentType;
    featuresList?: FeaturesList;
    id: EntityId;
    lastBatchNum?: number;
    name?: EntityName;
    pipeline?: Pipeline;
    preprocessedInsights?: PreprocessedInsightColumn[];
    profileId?: string;
    status:
        | "ready"
        | "error"
        | "cancelled"
        | "pending"
        | "dataprep_requested"
        | "datasync_requested"
        | "datasync_done";
    target: string;
    topModelId?: string;
    trainingDuration?: TrainingDuration;
    updatedAt: UpdatedAt;
    versionNumber?: number;
}
Index

Properties

algorithms? createdAt createdByUserId dataSetId datasetOrigin? dateIndexes? errorMessage? errors? experimentId experimentMode? experimentType featuresList? id lastBatchNum? name? pipeline? preprocessedInsights? profileId? status target topModelId? trainingDuration? updatedAt versionNumber?

Properties

algorithms?: ModelAlgorithm[]

List of algorithms selected for model training in this version

createdAt: CreatedAt
createdByUserId: OwnerId
dataSetId: DataSetId
datasetOrigin?: DatasetOrigin
dateIndexes?: DateIndexes
errorMessage?: ErrorMessage
errors?: Errors
experimentId: ExperimentId
experimentMode?: ExperimentMode
experimentType: ExperimentType
featuresList?: FeaturesList

List of features from your dataset for creating Experiment Versions. This appears in from ProfileInsights response (in the defaultVersionConfig). You can adjust the default settings before using it as input to create or update Experiment Versions.

id: EntityId
lastBatchNum?: number
name?: EntityName
pipeline?: Pipeline

Pipeline metadata including transformations to apply to columns and specific schema configuration data

preprocessedInsights?: PreprocessedInsightColumn[]
profileId?: string
status:
    | "ready"
    | "error"
    | "cancelled"
    | "pending"
    | "dataprep_requested"
    | "datasync_requested"
    | "datasync_done"
target: string
topModelId?: string

ID of the top model (based on training scores) in this experiment version

trainingDuration?: TrainingDuration

Optional training duration in seconds. If not provided, max value used. If provided, min 900 (15m) and max 21600 (6h).

updatedAt: UpdatedAt
versionNumber?: number

Settings

Member Visibility

On This Page

Properties