Provides configuration information for auto-labeling of your data objects. A
LabelingJobAlgorithmsConfig object must be supplied in order to use
auto-labeling.
Contents
- LabelingJobAlgorithmSpecificationArn
-
Specifies the Amazon Resource Name (ARN) of the algorithm used for auto-labeling. You must select one of the following ARNs:
-
Image classification
arn:aws:sagemaker:region:027400017018:labeling-job-algorithm-specification/image-classification -
Text classification
arn:aws:sagemaker:region:027400017018:labeling-job-algorithm-specification/text-classification -
Object detection
arn:aws:sagemaker:region:027400017018:labeling-job-algorithm-specification/object-detection -
Semantic Segmentation
arn:aws:sagemaker:region:027400017018:labeling-job-algorithm-specification/semantic-segmentation
Type: String
Length Constraints: Maximum length of 2048.
Pattern:
arn:.*Required: Yes
-
- InitialActiveLearningModelArn
-
At the end of an auto-label job Ground Truth sends the Amazon Resource Name (ARN) of the final model used for auto-labeling. You can use this model as the starting point for subsequent similar jobs by providing the ARN of the model here.
Type: String
Length Constraints: Minimum length of 20. Maximum length of 2048.
Pattern:
arn:aws[a-z\-]*:sagemaker:[a-z0-9\-]*:[0-9]{12}:model/.*Required: No
- LabelingJobResourceConfig
-
Provides configuration information for a labeling job.
Type: LabelingJobResourceConfig object
Required: No
See Also
For more information about using this API in one of the language-specific AWS SDKs, see the following:
