---
title: "init(column:)"
framework: createml
role: symbol
role_heading: Initializer
path: "createml/mldatacolumn/init(column:)"
---

# init(column:)

Creates a new column of machine learning sequences from a given column whose elements can be converted to sequences.

## Declaration

```swift
init<T>(column: MLDataColumn<T>) where T : MLDataValueConvertible
```

## Parameters

- `column`: An doc://com.apple.createml/documentation/CreateML/MLDataColumn of elements convertible to doc://com.apple.createml/documentation/CreateML/MLDataValue/SequenceType.

## Discussion

Discussion Use this initializer to create a column of sequences from another column. Start by creating a column that is convertible to a column of sequences. let intSequenceString = "[1, 2, 3]" let intSequenceString2 = "[4, 5, 6]" let stringsColumn = MLDataColumn([intSequenceString, intSequenceString2])

print(stringsColumn) // Prints "[1, 2, 3]", "[4, 5, 6]"] Then use init(column:) to convert the column to a column of sequences. let sequenceColumn = MLDataColumn<MLDataValue.SequenceType>(column: stringsColumn) print(sequenceColumn) // Prints [[DataValue(1), DataValue(2), DataValue(3)],                       // [DataValue(4), DataValue(5), DataValue(6)]]

## See Also

### Creating a data column by converting another column

- [map(to:)](createml/mldatacolumn/map(to:).md)
- [init(column:)](createml/mldatacolumn/init(column:)-5rg9u.md)
- [init(column:)](createml/mldatacolumn/init(column:)-2rxtu.md)
- [init(column:)](createml/mldatacolumn/init(column:)-86ge9.md)
- [init(column:)](createml/mldatacolumn/init(column:)-23pmx.md)
- [init(column:)](createml/mldatacolumn/init(column:)-ztkv.md)
- [init(column:)](createml/mldatacolumn/init(column:)-8uzuq.md)
- [init(column:)](createml/mldatacolumn/init(column:)-855l9.md)
- [init(column:)](createml/mldatacolumn/init(column:)-s8g5.md)