# ClassificationMetrics Classification metrics. ``` struct ClassificationMetrics where Label : Hashable ``` ## Topics ### Creating the distribution [`init(Predicted, Correct)`](/documentation/CreateMLComponents/ClassificationMetrics/init(_:_:)) Creates classification metrics for predicted and ground truth labels. [`init()`](/documentation/CreateMLComponents/ClassificationMetrics/init()) Creates empty classification metrics. [`init(_:)`](/documentation/CreateMLComponents/ClassificationMetrics/init(_:)) Creates classification metrics for a sequence of predicted and ground truth label pairs. [`init(some Sequence<(predicted: Label, label: Label)>, labels: Set)`](/documentation/CreateMLComponents/ClassificationMetrics/init(_:labels:)) Creates classification metrics for a sequence of predicted and ground truth label pairs. [`init(predicted: Predicted, groundTruth: Correct, labels: Set)`](/documentation/CreateMLComponents/ClassificationMetrics/init(predicted:groundTruth:labels:)) Creates classification metrics for predicted and ground truth labels. ### Getting the properties [`var accuracy: Double`](/documentation/CreateMLComponents/ClassificationMetrics/accuracy) The number of correctly classified examples out of the total number of examples. [`var exampleCount: Int`](/documentation/CreateMLComponents/ClassificationMetrics/exampleCount) The number of examples used to compute the metrics. [`var labels: Set`](/documentation/CreateMLComponents/ClassificationMetrics/labels) The set of labels. [`var restrictToKnownLabels: Bool`](/documentation/CreateMLComponents/ClassificationMetrics/restrictToKnownLabels) A Boolean value indicating whether to restrict metrics to labels in the labels set. ### Computing and scoring [`func makeConfusionMatrix() -> MLShapedArray`](/documentation/CreateMLComponents/ClassificationMetrics/makeConfusionMatrix()) Computes the confusion matrix. [`func precisionScore(label: Label) -> Double`](/documentation/CreateMLComponents/ClassificationMetrics/precisionScore(label:)) Computes the precision score for a class label. [`func recallScore(label: Label) -> Double`](/documentation/CreateMLComponents/ClassificationMetrics/recallScore(label:)) Computes the recall score for a class label. [`func count(label: Label) -> Int`](/documentation/CreateMLComponents/ClassificationMetrics/count(label:)) Returns the number of times a label appeared in the ground truth collection. [`func count(predicted: Label) -> Int`](/documentation/CreateMLComponents/ClassificationMetrics/count(predicted:)) Returns the number of times a label appeared in the predicted collection. [`func count(predicted: Label, label: Label) -> Int`](/documentation/CreateMLComponents/ClassificationMetrics/count(predicted:label:)) Returns the number of times a predicted, true label pair appeared in the label collections. [`func trueNegativeCount(of: Label) -> Int`](/documentation/CreateMLComponents/ClassificationMetrics/trueNegativeCount(of:)) Returns the number of times a label was not in the predicted or ground truth collections. [`func truePositiveCount(of: Label) -> Int`](/documentation/CreateMLComponents/ClassificationMetrics/truePositiveCount(of:)) Returns the number of times the predicted label matched the true label. [`func falseNegativeCount(of: Label) -> Int`](/documentation/CreateMLComponents/ClassificationMetrics/falseNegativeCount(of:)) Returns the number of times a true label was not predicted. [`func falsePositiveCount(of: Label) -> Int`](/documentation/CreateMLComponents/ClassificationMetrics/falsePositiveCount(of:)) Returns the number of times the predicted label did not match the true label. [`func f1Score(label: Label) -> Double`](/documentation/CreateMLComponents/ClassificationMetrics/f1Score(label:)) Computes the F1 score for a class label. [`func mapLabels((Label) throws -> T) rethrows -> ClassificationMetrics`](/documentation/CreateMLComponents/ClassificationMetrics/mapLabels(_:)) Returns new classification metrics where the labels are the result of applying a transformation. ### Updating the metrics [`func add(some Sequence<(predicted: Label, label: Label)>)`](/documentation/CreateMLComponents/ClassificationMetrics/add(_:)) Updates the metrics with more predicted and ground truth label pairs. [`func add(predicted: some Sequence, groundTruth: some Sequence)`](/documentation/CreateMLComponents/ClassificationMetrics/add(predicted:groundTruth:)) Updates the metrics with more predicted and ground truth labels. ## 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)