# ObjectDetectionMetrics Metrics for object detection model. ``` struct ObjectDetectionMetrics where Label : Comparable, Label : Decodable, Label : Encodable, Label : Hashable ``` ## Topics ### Creating a metrics object [`init()`](/documentation/CreateMLComponents/ObjectDetectionMetrics/init()) ### Getting the properties Properties [`var defaultConfidenceThreshold: Float`](/documentation/CreateMLComponents/ObjectDetectionMetrics/defaultConfidenceThreshold) The default confidence threshold. It is used as the confidence threshold for any label which does not have an explicit confidence threshold, while calculating averagePrecision and meanAveragePrecision. [`var labels: Set`](/documentation/CreateMLComponents/ObjectDetectionMetrics/labels) A set of labels present in the dataset. ### Calculating the precision [`func averageOfAveragePrecisionAtVariedThresholds(predictions: [[DetectedObject]], annotations: [ObjectDetectionAnnotation], confidenceThresholds: [Label : Float]) -> [Label : Scalar]`](/documentation/CreateMLComponents/ObjectDetectionMetrics/averageOfAveragePrecisionAtVariedThresholds(predictions:annotations:confidenceThresholds:)) Calculates average of average precision for all the labels, computed at varied bounding box overlap thresholds. The overlap thresholds range is from `[0.05, 0.95]` with a stride of `0.05`. [`func averageOfMeanAveragePrecisionAtVariedThresholds(predictions: [[DetectedObject]], annotations: [ObjectDetectionAnnotation], confidenceThresholds: [Label : Float]) -> Scalar`](/documentation/CreateMLComponents/ObjectDetectionMetrics/averageOfMeanAveragePrecisionAtVariedThresholds(predictions:annotations:confidenceThresholds:)) Calculates the average of mean average precision, computed at varied bounding box overlap thresholds. The overlap thresholds range is from `[0.05, 0.95]` with a stride of `0.05`. [`func averagePrecision(predictions: [[DetectedObject]], annotations: [ObjectDetectionAnnotation], confidenceThresholds: [Label : Float], overlapThreshold: Double) -> [Label : Scalar]`](/documentation/CreateMLComponents/ObjectDetectionMetrics/averagePrecision(predictions:annotations:confidenceThresholds:overlapThreshold:)) Calculates average precision for all the labels at the bounding box overlap threshold. [`func meanAveragePrecision(predictions: [[DetectedObject]], annotations: [ObjectDetectionAnnotation], confidenceThresholds: [Label : Float], overlapThreshold: Double) -> Scalar`](/documentation/CreateMLComponents/ObjectDetectionMetrics/meanAveragePrecision(predictions:annotations:confidenceThresholds:overlapThreshold:)) Calculates the mean average precision at the bounding box overlap threshold. ### Extracting labels [`static func extractLabels(from: [ObjectDetectionAnnotation]) -> Set`](/documentation/CreateMLComponents/ObjectDetectionMetrics/extractLabels(from:)) Extracts all the labels from a list of annotations. ## Relationships ### Conforms To [`Sendable`](/documentation/Swift/Sendable) [`SendableMetatype`](/documentation/Swift/SendableMetatype) --- Copyright © 2026 Apple Inc. All rights reserved. | [Terms of Use](https://www.apple.com/legal/internet-services/terms/site.html) | [Privacy Policy](https://www.apple.com/privacy/privacy-policy)