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

Timestamp when this was created

createdByUserId: OwnerId

ID of owner/user for this entity

dataSetId: DataSetId

The Qlik catalog dataset ID

datasetOrigin?: DatasetOrigin

Whether this is a new or other dataset

dateIndexes?: DateIndexes

A optional date column name to index

errorMessage?: ErrorMessage

JSON string of error object

errors?: Errors

JSON string with list of error objects

experimentId: ExperimentId

ID of the experiment

experimentMode?: ExperimentMode

The model training mode for the experiment version

experimentType: ExperimentType

Experiment type

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

ID of this entity

lastBatchNum?: number

Number of the last batch

name?: EntityName

Name of this entity

pipeline?: Pipeline

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

preprocessedInsights?: PreprocessedInsightColumn[]

Preprocessed insights. Like feature insights but with fewer details.

profileId?: string

ID of the dataset profile with metadata about source data

status:
    | "ready"
    | "error"
    | "cancelled"
    | "pending"
    | "dataprep_requested"
    | "datasync_requested"
    | "datasync_done"

Current status of this entity

target: string

The target field in the dataset

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

Timestamp when this was updated

versionNumber?: number

1-based sequential version number within the experiment

Settings

Member Visibility

On This Page

Properties