cast function
The cast function casts the type of the selected attribute to the specified data type.
This function can also be used when the attribute is empty, to ensure the correct data type is assigned to the column, as an empty column will be otherwise be assigned the nominal data type.
Parameters
cast(column, newtype, forced)
Parameter | Description |
|---|---|
column | The attribute whose values we want to cast. The column parameter is mandatory. |
newtype | The datatype the attribute will be cast to. The newtype parameter is mandatory. Possible values for this parameter are: “integer”, "nominal", “continuous”, "long", "double", "binary", "ordinal", "percentage", "currency", "date", "week", "month", "quarter", "datetime", "time", "index", "boolean", "int", "float", "short", "string”. |
forced | If set to False, any cast operations will would result in the loss of data precision (e.g. converting a continuous value to an integer) will not be performed, and an error message will be displayed. This parameter is set to True by default. |
Example
The following example uses the car_sales dataset.
Description | Screenshot |
|---|---|
In the car_sales dataset, the Sales_in_thousands attribute contains continuous values with detailed numbers |  |
If we cast this continuous attribute to an integer we will round off the number into the nearest thousand. The required formula is |  |
If we entered the value False for the forced parameter, the cast operation would not be performed, as it results in a loss of data precision of the continuous attribute values. The formula used in this case is An error message will be displayed at the bottom of the screen. |  |