so_magic.data.variables package

Submodules

so_magic.data.variables.types module

class so_magic.data.variables.types.IntervalVariableType

Bases: so_magic.data.variables.types.NumericalVariableType

Interval numerical variable type

Variables of type interval have interpretable differences; supported operations: [+, -]. There is no true zero.

Example: temperature in Celsius can be measured with an interval variable interval variable

Interpretable difference:

10 degrees drop from 30 degrees Celsius actually means 30 - 10 = 20 degrees Celsius

5 degrees rise 20 degrees Celsius actually means 20 + 5 = 25 degrees Celsius degrees Celsius - 10 degrees Celsius = 20 degrees Celsius

There is no true zero:

Theoretically we can go plus infinite degrees Celsius and minus infinite

There is no number that can “eliminate” (even zero has valid Celsius degrees smaller than 0) a temperature measurement in Celsius degrees

class so_magic.data.variables.types.NominalVariableType

Bases: so_magic.data.variables.types.CategoricalVariableType

Nominal variable; discrete variables with undefined ordering; eg country-names

class so_magic.data.variables.types.OrdinalVariableType

Bases: so_magic.data.variables.types.CategoricalVariableType

Ordinal variable; discrete variables with a defined ordering; eg days-of-the-week

class so_magic.data.variables.types.RatioVariableType

Bases: so_magic.data.variables.types.NumericalVariableType

Ratio numerical variable where all operations are supported (+, -, *, /) and true zero is defined; eg weight

class so_magic.data.variables.types.VariableTypeFactory

Bases: object

static create(variable_type: str, *args, **kwargs)

static infer(datapoints, attribute, sortable=True, ratio=None)

Semi-automatic identification; requires some input to assist;

Module contents