Variable

class floulib.Variable(*args)

Bases: Plot

This class contains methods to define and use variables to be used with rules.

Note

Variable is a subclass of Plot, therefore all methods in Plot may be used.

__init__(*args)

Constructor

Parameters:

  • *args (numpy.ndarray | Terms | numbers.Number | callable) – If only one argument is provided.

  • *args_ (Singleton | Term) – If several arguments of the same type are provided.

Raises:

  • Exception – Raised if no argument is provided.

  • TypeError – Raised if one argument is provided but is not an instance of numpy.ndarray | Terms | numbers.Number | callable

  • Exception – Raised if several arguments are provided but are not of the same type.

  • TypeError – Raised if several arguments are provided but are not instances of Singleton or Term.

Return type:

None.

Example

Numeric variables

>>> from floulib import Term, Triangle, Rule, Variable
>>> import numpy as np
>>> A1 = Term('A1', Triangle(0, 5, 10, label = '$A_1$'))
>>> B1 = Term('B1', Triangle(0, 2, 4, label = '$B_1$'))
>>> v1 = Variable(np.linspace(0,15, 1000))
>>> v2 = Variable(np.linspace(0, 10, 1000))
>>> R = Rule().If(v1.Is(A1)).Then(v2.Is(B1))
>>> A = Triangle(6, 8, 9).label('$A$')
>>> B = R.inference(v1.Is(A)).label('$B$')
>>> B.plot().add_plot(B1, alpha = 0.3)
_images/Variable.__init__1.png

Symbolic variables

>>> from floulib import Discrete, Term, Terms, Triangle, Rule, Rules, Variable
>>> import numpy as np
>>> A1 = Term('A1', Triangle(0, 5, 10, label = '$A_1$'))
>>> A2 = Term('A2', Triangle(5, 10, 15, label = '$A_2$'))
>>> B1 = Term('B1', Triangle(0, 2, 4, label = '$B_1$'))
>>> B2 = Term('B2', Triangle(2, 4, 6, label = '$B_2$'))
>>> v1 = Variable(A1, A2) # defined with instances of Term
>>> T2 = Terms(B1,B2)
>>> v2 = Variable(T2) # defined with one instance of Terms
>>> R = Rules(
    Rule().If(v1.Is(A1)).Then(v2.Is(B2)),
    Rule().If(v1.Is(A2)).Then(v2.Is(B1))
    )
>>> A = Discrete(('A1', 0.3), ('A2', 0.5))
>>> B = R.inference(v1.Is(A)).label('$B$')
>>> B.plot()
_images/Variable.__init__2.png
Is(_object=None)

This method is used to set the “value” of a variable used in the definition of rules and in the inference.

Parameters:

_object (Any | Discrete | Multilinear | Term | numbers.Number, optional) – The “value” to set. The default is None.

Raises:

TypeError – Raised if the parameter is not an instance of Any, Discrete, Multilinear, numbers.Number or Term.

Returns:

The variable object.

Return type:

Variable

Example

See Variable.__init__().

Certainty(level)

Adds a certainty level to a variable (see floulib.Discrete.Certainty(), floulib.Multilinear.Certainty(), floulib.Term.Certainty() depending of the type of the object associated with the variable).

Parameters:

level (float) – The certainty level.

Raises:

TypeError – Raised if the object associated with the variable is not an instance of Discrete, Multilinear or Term.

Returns:

Variable with a certainty level.

Return type:

Variable

Uncertainty(level)

Adds an uncertainty level to a variable (see floulib.Discrete.Uncertainty(), floulib.Multilinear.Uncertainty(), floulib.Term.Uncertainty() depending of the type of the object associated with the variable).

Parameters:

level (float) – The uncertainty level.

Raises:

TypeError – Raised if the object associated with the variable is not an instance of Discrete, Multilinear or Term.

Returns:

Variable with an uncertainty level.

Return type:

Variable