bach.DataFrame.create_variable

create_variable​

(name, value, *, dtype=None)

​[source]

Create a Series object that can be used as a variable, within the returned DataFrame. The DataFrame will have the variable with the given values set in DataFrame.variables().

The variable value can later be changed using DataFrame.set_variable()

Multiple variables with the same name

For variables the combination (name, dtype) uniquely identifies a variable. That means that there can be multiple variables with the same name, if they are of different types. This is counter to how a lot of programming languages handle variables. But it prevents a lot of error conditions and edge cases around merging DataFrames and Series with the same variables, and building on top of SqlModels that already have variables.

Parameters​

  • name (str) – name of variable to update
  • value (Any) – value of variable
  • dtype (Optional[str]) – optional. If not set it will be derived from the value, if set we check that it matches the value. If dtype doesn’t match the value’s dtype, then an Exception is raised.

Returns​

Tuple with DataFrame and the Series object that can be used as a variable.

Return type​

Tuple[DataFrame, Series]