# map(_:) Creates a new column by applying a given thread-safe transform to every row in the data table. ``` func map(_ lazyTransform: @escaping (MLDataTable.Row) -> T) -> MLDataColumn where T : MLDataValueConvertible ``` ## Parameters `lazyTransform` A thread-safe row transformation function. The implementation of your transform must accept a row from the data table and return a type that conforms to `MLDataValueConvertible` . ## Return Value A new `MLDataColumn`. ## Discussion Use this method to create a new column derived from the existing data in the table. The closure you pass evaluates lazily only when the transformed values are needed for a subsequent operation. Your implementation should accept a data table row and must be thread-safe because the framework may invoke the closure concurrently on unspecified threads. -92wrj-1@2x.png) For example, to perform the column derivation operation shown above, begin by creating a table of data. ```swift let data: [String: MLDataValueConvertible] = [ "Day": ["Monday", "Tuesday", "Wednesday", "Thursday", "Friday"], "Temperature": [91.3, 85.8, 79.5, 83.4, 72.2] ] var table = try MLDataTable(dictionary: data) ``` After creating the table, use `MLDataTable/map(_:)-92wrj` to create a new column of data from the original data. The example closure implementation below is stateless and safe to invoke concurrently on any thread, so no synchronization is necessary. ```swift let derivedColumn = table.map { row -> String in guard let day = row["Day"]?.stringValue, let temperature = row["Temperature"]?.doubleValue else { fatalError("Missing or invalid columns in row.") } return String(format: "%@ %.1fº", day, temperature) } ``` Then use `MLDataTable/addColumn(_:named:)-kkbw` to add the derived column to a table. ```swift table.addColumn(derivedColumn, named: "Description") ``` ### Mapping elements - - `map(_:)-3yamp` --- 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)