This product is not supported for your selected Datadog site. ().
Cette page n'est pas encore disponible en français, sa traduction est en cours.
Si vous avez des questions ou des retours sur notre projet de traduction actuel, n'hésitez pas à nous contacter.

aws_rekognition_project_version

account_id

Type: STRING

base_model_version

Type: STRING
Provider name: BaseModelVersion
Description: The base detection model version used to create the project version.

billable_training_time_in_seconds

Type: INT64
Provider name: BillableTrainingTimeInSeconds
Description: The duration, in seconds, that you were billed for a successful training of the model version. This value is only returned if the model version has been successfully trained.

creation_timestamp

Type: TIMESTAMP
Provider name: CreationTimestamp
Description: The Unix datetime for the date and time that training started.

evaluation_result

Type: STRUCT
Provider name: EvaluationResult
Description: The training results. EvaluationResult is only returned if training is successful.

  • f1_score
    Type: FLOAT
    Provider name: F1Score
    Description: The F1 score for the evaluation of all labels. The F1 score metric evaluates the overall precision and recall performance of the model as a single value. A higher value indicates better precision and recall performance. A lower score indicates that precision, recall, or both are performing poorly.
  • summary
    Type: STRUCT
    Provider name: Summary
    Description: The S3 bucket that contains the training summary.
    • s3_object
      Type: STRUCT
      Provider name: S3Object
      • bucket
        Type: STRING
        Provider name: Bucket
        Description: Name of the S3 bucket.
      • name
        Type: STRING
        Provider name: Name
        Description: S3 object key name.
      • version
        Type: STRING
        Provider name: Version
        Description: If the bucket is versioning enabled, you can specify the object version.

feature

Type: STRING
Provider name: Feature
Description: The feature that was customized.

feature_config

Type: STRUCT
Provider name: FeatureConfig
Description: Feature specific configuration that was applied during training.

  • content_moderation
    Type: STRUCT
    Provider name: ContentModeration
    Description: Configuration options for Custom Moderation training.
    • confidence_threshold
      Type: FLOAT
      Provider name: ConfidenceThreshold
      Description: The confidence level you plan to use to identify if unsafe content is present during inference.

kms_key_id

Type: STRING
Provider name: KmsKeyId
Description: The identifer for the AWS Key Management Service key (AWS KMS key) that was used to encrypt the model during training.

manifest_summary

Type: STRUCT
Provider name: ManifestSummary
Description: The location of the summary manifest. The summary manifest provides aggregate data validation results for the training and test datasets.

  • s3_object
    Type: STRUCT
    Provider name: S3Object
    • bucket
      Type: STRING
      Provider name: Bucket
      Description: Name of the S3 bucket.
    • name
      Type: STRING
      Provider name: Name
      Description: S3 object key name.
    • version
      Type: STRING
      Provider name: Version
      Description: If the bucket is versioning enabled, you can specify the object version.

max_inference_units

Type: INT32
Provider name: MaxInferenceUnits
Description: The maximum number of inference units Amazon Rekognition uses to auto-scale the model. Applies only to Custom Labels projects. For more information, see StartProjectVersion.

min_inference_units

Type: INT32
Provider name: MinInferenceUnits
Description: The minimum number of inference units used by the model. Applies only to Custom Labels projects. For more information, see StartProjectVersion.

output_config

Type: STRUCT
Provider name: OutputConfig
Description: The location where training results are saved.

  • s3_bucket
    Type: STRING
    Provider name: S3Bucket
    Description: The S3 bucket where training output is placed.
  • s3_key_prefix
    Type: STRING
    Provider name: S3KeyPrefix
    Description: The prefix applied to the training output files.

project_version_arn

Type: STRING
Provider name: ProjectVersionArn
Description: The Amazon Resource Name (ARN) of the project version.

source_project_version_arn

Type: STRING
Provider name: SourceProjectVersionArn
Description: If the model version was copied from a different project, SourceProjectVersionArn contains the ARN of the source model version.

status

Type: STRING
Provider name: Status
Description: The current status of the model version.

status_message

Type: STRING
Provider name: StatusMessage
Description: A descriptive message for an error or warning that occurred.

tags

Type: UNORDERED_LIST_STRING

testing_data_result

Type: STRUCT
Provider name: TestingDataResult
Description: Contains information about the testing results.

  • input
    Type: STRUCT
    Provider name: Input
    Description: The testing dataset that was supplied for training.
    • assets
      Type: UNORDERED_LIST_STRUCT
      Provider name: Assets
      Description: The assets used for testing.
      • ground_truth_manifest
        Type: STRUCT
        Provider name: GroundTruthManifest
        • s3_object
          Type: STRUCT
          Provider name: S3Object
          • bucket
            Type: STRING
            Provider name: Bucket
            Description: Name of the S3 bucket.
          • name
            Type: STRING
            Provider name: Name
            Description: S3 object key name.
          • version
            Type: STRING
            Provider name: Version
            Description: If the bucket is versioning enabled, you can specify the object version.
    • auto_create
      Type: BOOLEAN
      Provider name: AutoCreate
      Description: If specified, Rekognition splits training dataset to create a test dataset for the training job.
  • output
    Type: STRUCT
    Provider name: Output
    Description: The subset of the dataset that was actually tested. Some images (assets) might not be tested due to file formatting and other issues.
    • assets
      Type: UNORDERED_LIST_STRUCT
      Provider name: Assets
      Description: The assets used for testing.
      • ground_truth_manifest
        Type: STRUCT
        Provider name: GroundTruthManifest
        • s3_object
          Type: STRUCT
          Provider name: S3Object
          • bucket
            Type: STRING
            Provider name: Bucket
            Description: Name of the S3 bucket.
          • name
            Type: STRING
            Provider name: Name
            Description: S3 object key name.
          • version
            Type: STRING
            Provider name: Version
            Description: If the bucket is versioning enabled, you can specify the object version.
    • auto_create
      Type: BOOLEAN
      Provider name: AutoCreate
      Description: If specified, Rekognition splits training dataset to create a test dataset for the training job.
  • validation
    Type: STRUCT
    Provider name: Validation
    Description: The location of the data validation manifest. The data validation manifest is created for the test dataset during model training.
    • assets
      Type: UNORDERED_LIST_STRUCT
      Provider name: Assets
      Description: The assets that comprise the validation data.
      • ground_truth_manifest
        Type: STRUCT
        Provider name: GroundTruthManifest
        • s3_object
          Type: STRUCT
          Provider name: S3Object
          • bucket
            Type: STRING
            Provider name: Bucket
            Description: Name of the S3 bucket.
          • name
            Type: STRING
            Provider name: Name
            Description: S3 object key name.
          • version
            Type: STRING
            Provider name: Version
            Description: If the bucket is versioning enabled, you can specify the object version.

training_data_result

Type: STRUCT
Provider name: TrainingDataResult
Description: Contains information about the training results.

  • input
    Type: STRUCT
    Provider name: Input
    Description: The training data that you supplied.
    • assets
      Type: UNORDERED_LIST_STRUCT
      Provider name: Assets
      Description: A manifest file that contains references to the training images and ground-truth annotations.
      • ground_truth_manifest
        Type: STRUCT
        Provider name: GroundTruthManifest
        • s3_object
          Type: STRUCT
          Provider name: S3Object
          • bucket
            Type: STRING
            Provider name: Bucket
            Description: Name of the S3 bucket.
          • name
            Type: STRING
            Provider name: Name
            Description: S3 object key name.
          • version
            Type: STRING
            Provider name: Version
            Description: If the bucket is versioning enabled, you can specify the object version.
  • output
    Type: STRUCT
    Provider name: Output
    Description: Reference to images (assets) that were actually used during training with trained model predictions.
    • assets
      Type: UNORDERED_LIST_STRUCT
      Provider name: Assets
      Description: A manifest file that contains references to the training images and ground-truth annotations.
      • ground_truth_manifest
        Type: STRUCT
        Provider name: GroundTruthManifest
        • s3_object
          Type: STRUCT
          Provider name: S3Object
          • bucket
            Type: STRING
            Provider name: Bucket
            Description: Name of the S3 bucket.
          • name
            Type: STRING
            Provider name: Name
            Description: S3 object key name.
          • version
            Type: STRING
            Provider name: Version
            Description: If the bucket is versioning enabled, you can specify the object version.
  • validation
    Type: STRUCT
    Provider name: Validation
    Description: A manifest that you supplied for training, with validation results for each line.
    • assets
      Type: UNORDERED_LIST_STRUCT
      Provider name: Assets
      Description: The assets that comprise the validation data.
      • ground_truth_manifest
        Type: STRUCT
        Provider name: GroundTruthManifest
        • s3_object
          Type: STRUCT
          Provider name: S3Object
          • bucket
            Type: STRING
            Provider name: Bucket
            Description: Name of the S3 bucket.
          • name
            Type: STRING
            Provider name: Name
            Description: S3 object key name.
          • version
            Type: STRING
            Provider name: Version
            Description: If the bucket is versioning enabled, you can specify the object version.

training_end_timestamp

Type: TIMESTAMP
Provider name: TrainingEndTimestamp
Description: The Unix date and time that training of the model ended.

version_description

Type: STRING
Provider name: VersionDescription
Description: A user-provided description of the project version.