unpack(columnNamed:valueTypes:indexSubset:keySubset:)
Creates a new data table with additional columns that contain the unpacked collections in the given column.
Declaration
func unpack(columnNamed: String, valueTypes: [MLDataValue.ValueType]? = nil, indexSubset: [Int]? = nil, keySubset: [String]? = nil) -> MLDataTableParameters
- columnNamed:
The name of the column to unpack. The underlying type of the column must be either Sequencetype or Dictionarytype.
- valueTypes:
An array of the underlying types for the new, unpacked columns. If
nil, the method infers the underlying types in the sequence or dictionary. - indexSubset:
The subset of indicies to unpack from a specified sequence-typed column. If
nil, the method unpacks all indicies. - keySubset:
The subset of keys to unpack from a specified dictionary-typed column. If
nil, the method unpacks all keys.
Return Value
A new data table.
Discussion
This function performs the inverse of pack(columnsNamed:to:type:filling:).