probabilistic_model.learning.jpt.variables

Classes

Integer	Class for ordered, discrete random variables.
Continuous	Base class for continuous variables in JPTs. This class does not standardize the data,
ScaledContinuous	A continuous variable that is standardized.

Functions

infer_variables_from_dataframe(...)

Infer the variables from a dataframe.

Module Contents

probabilistic_model.learning.jpt.variables.infer_variables_from_dataframe(data: pandas.DataFrame, scale_continuous_types: bool = False, min_likelihood_improvement: float = 0.1, min_samples_per_quantile: int = 10) → typing_extensions.List[random_events.variable.Variable]

Infer the variables from a dataframe. The variables are inferred by the column names and types of the dataframe.

Parameters:

• data – The dataframe to infer the variables from.

• scale_continuous_types – Whether to scale numeric types.

• min_likelihood_improvement – The minimum likelihood improvement passed to the Continuous Variables.

• min_samples_per_quantile – The minimum number of samples per quantile passed to the Continuous Variables.

Returns:

The inferred variables.

class probabilistic_model.learning.jpt.variables.Integer(name: str, mean, std)

Bases: random_events.variable.Integer

Class for ordered, discrete random variables.

The domain of an integer variable is the number line.

mean: float

Mean of the random variable.

std: float

Standard Deviation of the random variable.

to_json() → typing_extensions.Dict[str, typing_extensions.Any]

classmethod _from_json(data: typing_extensions.Dict[str, typing_extensions.Any]) → typing_extensions.Self

Create a variable from a json dict. This method is called from the from_json method after the correct subclass is determined and should be overwritten by the respective subclass.

Parameters:

data – The json dict

Returns:

The deserialized object

__eq__(other)

__hash__()

class probabilistic_model.learning.jpt.variables.Continuous(name: str, mean: float, std: float, minimal_distance: float = 1.0, min_likelihood_improvement: float = 0.1, min_samples_per_quantile: int = 10)

Bases: random_events.variable.Continuous

Base class for continuous variables in JPTs. This class does not standardize the data, but needs to know mean and std anyway.

mean: float

Mean of the random variable.

std: float

Standard Deviation of the random variable.

minimal_distance: float

The minimal distance between two values of the variable.

min_likelihood_improvement: float

The minimum likelihood improvement passed to the Nyga Distributions.

min_samples_per_quantile: int

The minimum number of samples per quantile passed to the Nyga Distributions.

to_json() → typing_extensions.Dict[str, typing_extensions.Any]

classmethod _from_json(data: typing_extensions.Dict[str, typing_extensions.Any]) → typing_extensions.Self

Create a variable from a json dict. This method is called from the from_json method after the correct subclass is determined and should be overwritten by the respective subclass.

Parameters:

data – The json dict

Returns:

The deserialized object

class probabilistic_model.learning.jpt.variables.ScaledContinuous(name: str, mean: float, std: float, minimal_distance: float = 1.0, min_likelihood_improvement: float = 0.1, min_samples_per_quantile: int = 10)

Bases: Continuous

A continuous variable that is standardized.

encode(value: typing_extensions.Any)

decode(value: float) → float

__str__()