# loadRangedAnnotations(parameters:continueOnFailure:) Loads training examples from a data frame containing annotations. ``` func loadRangedAnnotations(parameters: DataFrameTemporalAnnotationParameters = .init(), continueOnFailure: Bool = false) throws -> [AnnotatedFeature] where Annotation : Equatable, Annotation : Sendable ``` ## Parameters `parameters` Annotation parameters for using specific column names from the data frame. By default the data frame is expected to have columns `filePath` and `annotation`. `continueOnFailure` A Boolean value indicating whether to continue reading the dataframe after encountering a row that is invalid. The default value is `false`. ## Return Value A list of annotated features containing `URLRange` as feature and `Annotation` as annotation. ## Discussion The exact column ids within the data frame which contains the file path, annotation, startTime and endTime of each temporal annotation, can be customized using the `parameters` argument. The data frame must contain at least two columns: one for file path and one for annotation. The file path must be of String type and annotation can be of any equatable type. The `filePath` column in the `dataFrame` should contain file path with its extension. All the files can have absolute path or a path relative to a `baseURL`. This option can be specified using `filePathType` in `parameters`. The `startTime` and `endTime` columns contain the start time and end time of a temporal segment in seconds. Their type is Double. Both the columns are optional. If only a `startTime` is specified, it is interpreted as an annotation for the range starting from `startTime` till the end of file, i.e., `startTime ...`. If only an `endTime` is specified, it is interpreted as an annotation for the range starting at 0 till `endTime`, i.e., `0 ..< endTime`. If both `startTime` and `endTime` columns are missing or their values are missing, the annotation is assumed to be for the entire file length. > Note: Ensure that the same file has no overlapping range in the dataframe. Since overlapping > ranges would mean conflicting annotation for that range. Consider an example of a dataframe containing the following values: ``` ------------------------------------------------ | fileName | annotation | startTime | endTime | | ----------|------------|-----------|---------| | file1.wav | class1 | 0 | 4 | | file1.wav | class2 | 5 | 10 | | file2.wav | class1 | 2 | | | file3.wav | class2 | | | | file4.wav | class1 | | 5 | ------------------------------------------------ ``` This will create an annotated feature list which has the following structure: ```swift [ AnnotatedFeature(feature: TemporalFileSegment("file1.wav", 0 ..< 4), annotation: "class1"), AnnotatedFeature(feature: TemporalFileSegment("file1.wav", 5 ..< 10), annotation: "class2"), AnnotatedFeature(feature: TemporalFileSegment("file2.wav", 2 ...), annotation: "class1"), AnnotatedFeature(feature: TemporalFileSegment("file3.wav", 0 ...), annotation: "class2"), AnnotatedFeature(feature: TemporalFileSegment("file4.wav", 0 ..< 5), annotation: "class1") ] ``` --- 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)