MLModelConfigOutput

Type: Object

ML model configuration for training pipelines

Fields

Name	Type	Args	Description	Deprecated
autoPublish	Boolean!		Whether new versions are automatically published after training
backend	MLBackend!		Engine that trains/serves this model
createdAt	String!		Creation timestamp (ISO8601)
csvFilename	String		Original CSV filename (if CSV data source)
csvPreviewRows	[[String!]!]		First few rows of CSV data for preview (header + up to 10 rows)
csvSizeBytes	Int		Size of uploaded CSV in bytes
dataQuerySQL	String		MicroSQL query for data retrieval
description	String		Config description
emittedSchema	[MLEmittedColumn!]		The pipeline's emitted columns + types (cached from the last preview of the saved config)
evaluationHistory	[MLEvaluationEntry!]		Per-round evaluation history
featureImportance	[MLFeatureImportance!]		Feature importance from trained model
hasTrainingCSV	Boolean!		Whether a CSV file has been uploaded as training data
id	MLModelConfigID!		Unique identifier
inferenceSchema	String		JSON Schema describing the model's input features (types, enum values). Generated after training.
isSystem	Boolean!		Whether this model is system-managed (auto-created and auto-retrained by the platform)
lastTrainedAt	String		When the model was last trained (ISO8601)
modelType	MLModelType!		Type of ML model
name	String!		Config name
publishedVersionId	MLModelConfigVersionID		ID of the version currently used for predictions
publishedVersionNumber	Int		Version number of the published version (e.g. 1, 2, 3)
spec	MLModelSpec!		Concept-specific configuration (tabular or time-series)
status	MLModelTrainingStatus!		Current training status
tenantId	TenantID!		Tenant this config belongs to
trainedModelID	String		Blob storage key for the trained model
trainingDurationMs	Int		Duration of last training run in milliseconds
trainingError	String		Error message if last training failed
trainingMetrics	TrainingMetrics		Metrics from last training run
updatedAt	String!		Last update timestamp (ISO8601)

Relationships

• TrainingMetrics