ClassificationObservation

confidence

get confidence(): number

let confidence: Float

The level of confidence normalized to `[0, 1]` where `1` is most confident. The only exception is results coming from `CoreMLRequest`, where confidence values are forwarded as is from relevant CoreML models

identifier

get identifier(): string

let identifier: String

The is the identifier of a classification request. An example classification could be a string like 'cat' or 'hotdog'. The string is defined in the model that was used for the classification. Usually these are technical labels that are not localized and not meant to be used directly to be presented to an end user in the UI.

description

get description(): string

var description: String { get }

A textual representation of this instance. Calling this property directly is discouraged. Instead, convert an instance of any type to a string by using the `String(describing:)` initializer. This initializer works with any type, and uses the custom `description` property for types that conform to `CustomStringConvertible`: struct Point: CustomStringConvertible { let x: Int, y: Int var description: String { return "(\(x), \(y))" } } let p = Point(x: 21, y: 30) let s = String(describing: p) print(s) // Prints "(21, 30)" The conversion of `p` to a string in the assignment to `s` uses the `Point` type's `description` property.

hasMinimumRecall

hasMinimumRecall(minimumRecall: number, precision: number): boolean

func hasMinimumRecall(_ minimumRecall: Float, forPrecision precision: Float) -> Bool

Determines whether the observation has a minimum recall value for a specific precision. - Parameters: - minimumRecall: The minimum desired percentage of relevant results that the algorithm correctly classified. - precision: The percentage of classification results that are relevant. - Returns: A Boolean indicating whether or not this classification observation provides a minimum percentage of relevant results that meet the desired precision criterion.

hasMinimumPrecision

hasMinimumPrecision(minimumPrecision: number, recall: number): boolean

func hasMinimumPrecision(_ minimumPrecision: Float, forRecall recall: Float) -> Bool

Determines whether the observation has a minimum precision value for a specific recall. - Parameters: - minimumPrecision: The minimum percentage of classification results that are relevant. - recall: The percentage of relevant results that the algorithm correctly classified. - Returns: A Boolean indicating whether or not this classification observation provides a minimum percentage of relevant results that meet the desired recall criterion.

hasPrecisionRecallCurve

get hasPrecisionRecallCurve(): boolean

var hasPrecisionRecallCurve: Bool { get }

A Boolean variable indicating whether the observation contains precision and recall curves. Precision refers to the percentage of your classification results that are relevant, while recall refers to the percentage of total relevant results correctly classified. If this property is true, then you can call precision and recall-related methods in this observation. If this property is false, then the precision and recall-related methods won't return meaningful data.

originatingRequestDescriptor

get originatingRequestDescriptor(): (RequestDescriptor) | null

let originatingRequestDescriptor: RequestDescriptor?

The descriptor of the request that produced the observation.

uuid

get uuid(): string

let uuid: UUID

The unique identifier for the observation.

timeRange

get timeRange(): (_CoreMedia_CMTimeRange) | null

let timeRange: CMTimeRange?

The time range of the reported observation. When evaluating a sequence of image buffers, use this property to determine each observation’s start time and duration.

init

static init(observation: VNClassificationObservation): ClassificationObservation

init(_ observation: VNClassificationObservation)