Random Variables

aGrUM/pyAgrum is currently dedicated for discrete probability distributions.

There are 5 types of discrete random variables in aGrUM/pyAgrum: pyagrum.LabelizedVariable, pyagrum.DiscretizedVariable, pyagrum.IntegerVariable, pyagrum.RangeVariable and pyagrum.NumericalDiscreteVariable. The 5 types are mainly provided in order to ease modelization. Derived from DiscreteVariable, they share a common API. They essentially differ by the means to create, name and access to their modalities.

The function pyagrum.fastVariable() allows to easily create variables of any types with the fast syntax.

Common API for Random Discrete Variables

class pyagrum.DiscreteVariable(*args, **kwargs)

DiscreteVariable is the (abstract) base class for discrete random variables.

asDiscretizedVar()

Tries to cast the variable as a pyagrum.DiscretizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.DiscretizedVar
• Return type: DiscretizedVariable

asIntegerVar()

Tries to cast the variable as a pyagrum.IntegerVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.IntegerVar
• Return type: IntegerVariable

asLabelizedVar()

Tries to cast the variable as a pyagrum.LabelizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.LabelizedVar
• Return type: LabelizedVariable

asNumericalDiscreteVar()

Tries to cast the variable as a pyagrum.NumericalDiscreteVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.NumericalDiscreteVar
• Return type: NumericalDiscreteVariable

asRangeVar()

Tries to cast the variable as a pyagrum.RangeVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.RangeVar
• Return type: RangeVariable

belongs(x)

returns whether x belongs to the variable domain (if the variable is a pyagrum.RangeVariable)

• Parameters: x (int) – the value for which we want to know if it belongs to the variable domain
• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: bool
• Returns: bool : True if x belongs to the variable domain, False otherwise

closestIndex(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the index of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the index of the closest value to value in the variable’s domain
• Return type: int
• Raises: pyagrum.NotImplementedYet – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

closestLabel(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the label of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the label of the closest value to value in the variable’s domain
• Return type: str
• Raises: pyagrum.InvalidArgument – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

description()

• Returns: the description of the variable
• Return type: str

domain()

• Returns: the domain of the variable
• Return type: str

domainSize()

• Returns: the number of modalities in the variable domain
• Return type: int

draw(x)

Draw a random float value from the x-th interval of the variable (only for pyagrum.DiscretizedVariable).

• Parameters: x (int) – the index of the interval from which to draw
• Returns: a random value drawn uniformly from the x-th interval
• Return type: float
• Raises:
  • NotImplementedError – if the variable is not a pyagrum.DiscretizedVariable
  • pyagrum.OutOfBounds – if x is not a valid interval index

empty()

• Returns: True if the domain size < 2
• Return type: bool

index(label)

• Parameters: label (str) – a label
• Returns: the index of the label
• Return type: int

integerDomain()

Returns the integer domain of the variable (if the variable is a pyagrum.integerVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.IntegerVariable
• Return type: tuple[int, ...]
• Returns: tuple[int,…] : the integer domain of the variable

isEmpirical()

Returns whether the variable is empirical (if the variable is a pyagrum.DiscretizedVariable). I.e. whether its limits (max and min) are defined by data and accept any values below or above these limits, contrary to non-empirical discretized variables for which the limits are fixed and values outside of these limits are not accepted)

• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Return type: bool
• Returns: bool : whether the variable is empirical

isLabel(s)

returns whether s is a label of the variable

• Raises: NotImplementedError – If the variable is not a pyagrum.LabelizedVariable
• Return type: bool
• Returns: bool : True if s is a label of the variable, False otherwise
• Parameters: s (str)

isNumerical()

Returns True if the variable has a meaningful numerical representation, i.e. it is not a LabelizedVariable.

• Returns: True for DiscretizedVariable, IntegerVariable, NumericalDiscreteVariable and RangeVariable; False for LabelizedVariable.
• Return type: bool

isTick(x)

Returns whether x is a tick for the variable

• Parameters: x (float) – the value for which we want to know if it is a tick for the variable
• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Returns: bool : whether x is a tick for the variable, i.e. whether x belongs to the variable’s domain

isValue(x)

Returns whether x is a value for the variable

• Parameters: x (float) – the value for which we want to know if it is a value for the variable
• Raises: NotImplementedError – If the variable is not a pyagrum.NumericalDiscreteVariable or a pyagrum.IntegerVariable
• Return type: bool
• Returns: bool : whether x is a value for the variable, i.e. whether x belongs to the variable domain

label(i)

• Parameters: i (int) – the index of the label we wish to return
• Returns: the label at the given index
• Return type: str
• Raises: pyagrum.OutOfBounds – If the variable does not contain the label

labels()

• Returns: a tuple containing the labels
• Return type: tuple[str, ...]

maxVal()

returns the maximum value of the variable domain (if the variable is a pyagrum.RangeVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: int
• Returns: int : the maximum value of the variable domain

minVal()

returns the minimum value of the variable domain (if the variable is a pyagrum.RangeVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: int
• Returns: int : the minimum value of the variable domain

name()

• Returns: the name of the variable
• Return type: str

numerical(indice)

• Parameters:
  • index (int) – an index
  • indice (int)
• Returns: the numerical representation of the value at the given index
• Return type: float

numericalDomain()

Returns the numerical domain of the variable (if the variable is a pyagrum.NumericalDiscreteVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.NumericalDiscreteVariable
• Return type: tuple[float, ...]
• Returns: tuple[float,…] : the numerical domain of the variable

posLabel(s)

Returns the index of the label s in the variable’s domain (if the variable is a pyagrum.LabelizedVariable)

• Raises:
  • NotImplementedError – If the variable is not a pyagrum.LabelizedVariable
  • NotFoundError – If the variable does not contain the label s
• Return type: int
• Returns: int : the index of the label s in the variable’s domain
• Parameters: s (str)

setDescription(theValue)

set the description of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

setName(theValue)

sets the name of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

stype()

• Returns: a description of its type
• Return type: str

tick(x)

Returns the x-th tick of the variable (if the variable is a pyagrum.DiscretizedVariable)

• Parameters: x (int) – the index of the tick we want to get
• Raises:
  • NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
  • pyagrum.OutOfBounds – If x is not a valid index for the ticks of the variable
• Returns: float : the x-th tick of the variable

ticks()

Returns the ticks of the variable

• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Return type: tuple[float, ...]
• Returns: tuple[float,…] : the ticks of the variable, i.e. the values that belong to the variable’s domain

toFast()

Return the variable description in fast syntax (e.g. name[val1|val2|...]).

• Returns: fast-syntax string representation
• Return type: str

toStringWithDescription()

• Returns: a description of the variable
• Return type: str

varType()

returns the type of variable

• Returns: the type of the variable.

  0: DiscretizedVariable, 1: LabelizedVariable, 2: IntegerVariable, 3: NumericalDiscreteVariable, 4: RangeVariable

• Return type: int

Concrete classes for Random Discrete Variables

LabelizedVariable

class pyagrum.LabelizedVariable(*args)

LabelizedVariable is a discrete random variable with a customizable sequence of labels.

LabelizedVariable(aName, aDesc=’’, nbrLabel=2) -> LabelizedVariable : Parameters: : - aName (str) – the name of the variable - aDesc (str) – the (optional) description of the variable - nbrLabel (int) – the number of labels to create (2 by default)

LabelizedVariable(aName, aDesc=’’, labels) -> LabelizedVariable : Parameters: : - aName (str) – the name of the variable - aDesc (str) – the (optional) description of the variable - labels (list of str) – the labels to create

LabelizedVariable(aLDRV) -> LabelizedVariable : Parameters: : - aLDRV (pyagrum.LabelizedVariable) – The pyagrum.LabelizedVariable that will be copied

Examples

>>> import pyagrum as gum
>>> # creating a variable with 3 labels : '0', '1' and '2'
>>> va=pyagrum.LabelizedVariable('a','a labelized variable',3)
>>> print(va)
a:Labelized(<0,1,2>)
>>> va.addLabel('foo')
(pyagrum.LabelizedVariable@0x7fc4c840dd90) a:Labelized(<0,1,2,foo>)
>>> va.changeLabel(1,'bar')
>>> print(va)
a:Labelized(<0,bar,2,foo>)
>>> vb=pyagrum.LabelizedVariable('b','b',0).addLabel('A').addLabel('B').addLabel('C')
>>> print(vb)
b:Labelized(<A,B,C>)
>>> vb.labels()
('A', 'B', 'C')
>>> vb.isLabel('E')
False
>>> vb.label(2)
'C'
>>> vc=pyagrum.LabelizedVariable('b','b',['one','two','three'])
>>> vc
(pyagrum.LabelizedVariable@0x7fc4c840c130) b:Labelized(<one,two,three>)

addLabel(*args)

Add a label with a new index (we assume that we will NEVER remove a label).

• Parameters: aLabel (str) – the label to be added to the labelized variable
• Returns: the labelized variable
• Return type: pyagrum.LabelizedVariable
• Raises: pyagrum.DuplicateElement – If the variable already contains the label

asDiscretizedVar()

Tries to cast the variable as a pyagrum.DiscretizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.DiscretizedVar
• Return type: DiscretizedVariable

asIntegerVar()

Tries to cast the variable as a pyagrum.IntegerVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.IntegerVar
• Return type: IntegerVariable

asLabelizedVar()

Tries to cast the variable as a pyagrum.LabelizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.LabelizedVar
• Return type: LabelizedVariable

asNumericalDiscreteVar()

Tries to cast the variable as a pyagrum.NumericalDiscreteVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.NumericalDiscreteVar
• Return type: NumericalDiscreteVariable

asRangeVar()

Tries to cast the variable as a pyagrum.RangeVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.RangeVar
• Return type: RangeVariable

belongs(x)

returns whether x belongs to the variable domain (if the variable is a pyagrum.RangeVariable)

• Parameters: x (int) – the value for which we want to know if it belongs to the variable domain
• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: bool
• Returns: bool : True if x belongs to the variable domain, False otherwise

changeLabel(pos, aLabel)

Change the label at the specified index

• Parameters:
  • pos (int) – the index of the label to be changed
  • aLabel (str) – the label to be added to the labelized variable
• Raises:
  • pyagrum.DuplicateElement – If the variable already contains the new label
  • pyagrum.OutOfBounds – If the index is greater than the size of the variable
• Return type: None

closestIndex(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the index of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the index of the closest value to value in the variable’s domain
• Return type: int
• Raises: pyagrum.NotImplementedYet – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

closestLabel(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the label of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the label of the closest value to value in the variable’s domain
• Return type: str
• Raises: pyagrum.InvalidArgument – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

description()

• Returns: the description of the variable
• Return type: str

domain()

• Returns: the domain of the variable as a string
• Return type: str

domainSize()

• Returns: the number of modalities in the variable domain
• Return type: int

draw(x)

Draw a random float value from the x-th interval of the variable (only for pyagrum.DiscretizedVariable).

• Parameters: x (int) – the index of the interval from which to draw
• Returns: a random value drawn uniformly from the x-th interval
• Return type: float
• Raises:
  • NotImplementedError – if the variable is not a pyagrum.DiscretizedVariable
  • pyagrum.OutOfBounds – if x is not a valid interval index

empty()

• Returns: True if the domain size < 2
• Return type: bool

eraseLabels()

Erase all the labels from the variable.

• Return type: None

index(label)

• Parameters: label (str) – a label
• Returns: the index of the label
• Return type: int

integerDomain()

Returns the integer domain of the variable (if the variable is a pyagrum.integerVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.IntegerVariable
• Return type: tuple[int, ...]
• Returns: tuple[int,…] : the integer domain of the variable

isEmpirical()

Returns whether the variable is empirical (if the variable is a pyagrum.DiscretizedVariable). I.e. whether its limits (max and min) are defined by data and accept any values below or above these limits, contrary to non-empirical discretized variables for which the limits are fixed and values outside of these limits are not accepted)

• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Return type: bool
• Returns: bool : whether the variable is empirical

isLabel(aLabel)

Indicates whether the variable already has the label passed in argument

• Parameters: aLabel (str) – the label to be tested
• Returns: True if the label already exists
• Return type: bool

isNumerical()

Returns True if the variable has a meaningful numerical representation, i.e. it is not a LabelizedVariable.

• Returns: True for DiscretizedVariable, IntegerVariable, NumericalDiscreteVariable and RangeVariable; False for LabelizedVariable.
• Return type: bool

isTick(x)

Returns whether x is a tick for the variable

• Parameters: x (float) – the value for which we want to know if it is a tick for the variable
• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Returns: bool : whether x is a tick for the variable, i.e. whether x belongs to the variable’s domain

isValue(x)

Returns whether x is a value for the variable

• Parameters: x (float) – the value for which we want to know if it is a value for the variable
• Raises: NotImplementedError – If the variable is not a pyagrum.NumericalDiscreteVariable or a pyagrum.IntegerVariable
• Return type: bool
• Returns: bool : whether x is a value for the variable, i.e. whether x belongs to the variable domain

label(i)

• Parameters: i (int) – the index of the label we wish to return
• Returns: the label at the given index
• Return type: str
• Raises: pyagrum.OutOfBounds – If the variable does not contain the label

labels()

• Returns: a tuple containing the labels
• Return type: tuple[str, ...]

maxVal()

returns the maximum value of the variable domain (if the variable is a pyagrum.RangeVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: int
• Returns: int : the maximum value of the variable domain

minVal()

returns the minimum value of the variable domain (if the variable is a pyagrum.RangeVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: int
• Returns: int : the minimum value of the variable domain

name()

• Returns: the name of the variable
• Return type: str

numerical(index)

• Parameters: index (int) – an index
• Returns: the numerical representation of the value at the given index
• Return type: float

numericalDomain()

Returns the numerical domain of the variable (if the variable is a pyagrum.NumericalDiscreteVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.NumericalDiscreteVariable
• Return type: tuple[float, ...]
• Returns: tuple[float,…] : the numerical domain of the variable

posLabel(label)

Return the position (index) of a label in the variable’s domain.

• Parameters: label (str) – the label to look up
• Returns: the index of the label in the domain
• Return type: int
• Raises: pyagrum.NotFound – if the label does not exist in the domain

setDescription(theValue)

set the description of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

setName(theValue)

sets the name of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

stype()

• Returns: a description of its type
• Return type: str

tick(x)

Returns the x-th tick of the variable (if the variable is a pyagrum.DiscretizedVariable)

• Parameters: x (int) – the index of the tick we want to get
• Raises:
  • NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
  • pyagrum.OutOfBounds – If x is not a valid index for the ticks of the variable
• Returns: float : the x-th tick of the variable

ticks()

Returns the ticks of the variable

• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Return type: tuple[float, ...]
• Returns: tuple[float,…] : the ticks of the variable, i.e. the values that belong to the variable’s domain

toFast()

Return the variable description in fast syntax (e.g. name[val1|val2|...]).

• Returns: fast-syntax string representation
• Return type: str

toStringWithDescription()

• Returns: a description of the variable
• Return type: str

varType()

returns the type of variable

• Returns: the type of the variable.

  0: DiscretizedVariable, 1: LabelizedVariable, 2: IntegerVariable, 3: NumericalDiscreteVariable, 4: RangeVariable

• Return type: int

DiscretizedVariable

class pyagrum.DiscretizedVariable(*args)

DiscretizedVariable is a discrete random variable with a set of ticks defining intervals.

DiscretizedVariable(aName, aDesc ,ticks=None,is_empirical=False) -> pyagrum.DiscretizedVariable Parameters:

• aName (str) – the name of the variable
• aDesc (str) – the description of the variable
• ticks (list[float]) – the list of ticks to add
• is_empirical (bool) – if False, raise an error if a value is out of bound.

DiscretizedVariable(aDDRV) -> DiscretizedVariable : Parameters:

• aDDRV (pyagrum.DiscretizedVariable) – the pyagrum.DiscretizedVariable that will be copied

Examples

>>> import pyagrum as gum
>>> vX=pyagrum.DiscretizedVariable('X','X has been discretized').addTick(1).addTick(2).addTick(3).addTick(3.1415)
>>> print(vX)
X:Discretized(<[1;2[,[2;3[,[3;3.1415]>)
>>> vX.isTick(4)
False
>>> vX.labels()
('[1;2[', '[2;3[', '[3;3.1415]')
>>> # where is the real value 2.5 ?
>>> vX.index('2.5')
1

addTick(*args)

• Parameters: aTick (double) – the Tick to be added
• Returns: the discretized variable
• Return type: pyagrum.DiscretizedVariable
• Raises: pyagrum.DefaultInLabel – If the tick is already defined

asDiscretizedVar()

Tries to cast the variable as a pyagrum.DiscretizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.DiscretizedVar
• Return type: DiscretizedVariable

asIntegerVar()

Tries to cast the variable as a pyagrum.IntegerVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.IntegerVar
• Return type: IntegerVariable

asLabelizedVar()

Tries to cast the variable as a pyagrum.LabelizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.LabelizedVar
• Return type: LabelizedVariable

asNumericalDiscreteVar()

Tries to cast the variable as a pyagrum.NumericalDiscreteVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.NumericalDiscreteVar
• Return type: NumericalDiscreteVariable

asRangeVar()

Tries to cast the variable as a pyagrum.RangeVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.RangeVar
• Return type: RangeVariable

belongs(x)

returns whether x belongs to the variable domain (if the variable is a pyagrum.RangeVariable)

• Parameters: x (int) – the value for which we want to know if it belongs to the variable domain
• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: bool
• Returns: bool : True if x belongs to the variable domain, False otherwise

closestIndex(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the index of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the index of the closest value to value in the variable’s domain
• Return type: int
• Raises: pyagrum.NotImplementedYet – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

closestLabel(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the label of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the label of the closest value to value in the variable’s domain
• Return type: str
• Raises: pyagrum.InvalidArgument – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

description()

• Returns: the description of the variable
• Return type: str

domain()

• Returns: the domain of the variable as a string
• Return type: str

domainSize()

• Returns: the number of modalities in the variable domain
• Return type: int

draw(indice)

Allow to draw a value in the i-th interval of the discretized variable.1

• Parameters:
  • i (int) – the index of the interval to draw
  • indice (int)
• Returns: the value randomly drawn in the i-th interval
• Return type: float

empty()

• Returns: True if the domain size < 2
• Return type: bool

eraseTicks()

erase all the Ticks

• Return type: None

index(*args)

• Parameters: label (str) – a label
• Returns: the index of the label
• Return type: int

integerDomain()

Returns the integer domain of the variable (if the variable is a pyagrum.integerVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.IntegerVariable
• Return type: tuple[int, ...]
• Returns: tuple[int,…] : the integer domain of the variable

isEmpirical()

Check whether the variable uses relaxed (empirical) tick bounds.

• Returns: True if the variable is in empirical mode
• Return type: bool

isLabel(s)

returns whether s is a label of the variable

• Raises: NotImplementedError – If the variable is not a pyagrum.LabelizedVariable
• Return type: bool
• Returns: bool : True if s is a label of the variable, False otherwise
• Parameters: s (str)

isNumerical()

Returns True if the variable has a meaningful numerical representation, i.e. it is not a LabelizedVariable.

• Returns: True for DiscretizedVariable, IntegerVariable, NumericalDiscreteVariable and RangeVariable; False for LabelizedVariable.
• Return type: bool

isTick(aTick)

• Parameters: aTick (float) – the Tick to be tested
• Returns: True if the Tick already exists
• Return type: bool

isValue(x)

Returns whether x is a value for the variable

• Parameters: x (float) – the value for which we want to know if it is a value for the variable
• Raises: NotImplementedError – If the variable is not a pyagrum.NumericalDiscreteVariable or a pyagrum.IntegerVariable
• Return type: bool
• Returns: bool : whether x is a value for the variable, i.e. whether x belongs to the variable domain

label(i)

• Parameters: i (int) – the index of the label we wish to return
• Returns: the label at the given index
• Return type: str
• Raises: pyagrum.OutOfBounds – If the variable does not contain the label

labels()

• Returns: a tuple containing the labels
• Return type: tuple[str, ...]

maxVal()

returns the maximum value of the variable domain (if the variable is a pyagrum.RangeVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: int
• Returns: int : the maximum value of the variable domain

minVal()

returns the minimum value of the variable domain (if the variable is a pyagrum.RangeVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: int
• Returns: int : the minimum value of the variable domain

name()

• Returns: the name of the variable
• Return type: str

numerical(indice)

• Parameters:
  • index (int) – an index
  • indice (int)
• Returns: the numerical representation of the value at the given index
• Return type: float

numericalDomain()

Returns the numerical domain of the variable (if the variable is a pyagrum.NumericalDiscreteVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.NumericalDiscreteVariable
• Return type: tuple[float, ...]
• Returns: tuple[float,…] : the numerical domain of the variable

posLabel(s)

Returns the index of the label s in the variable’s domain (if the variable is a pyagrum.LabelizedVariable)

• Raises:
  • NotImplementedError – If the variable is not a pyagrum.LabelizedVariable
  • NotFoundError – If the variable does not contain the label s
• Return type: int
• Returns: int : the index of the label s in the variable’s domain
• Parameters: s (str)

setDescription(theValue)

set the description of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

setEmpirical(state)

Enable or disable empirical (relaxed bounds) mode for the variable.

• Parameters: state (bool) – True to enable empirical mode, False to disable
• Return type: None

setName(theValue)

sets the name of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

stype()

• Returns: a description of its type
• Return type: str

tick(i)

Indicate the index of the Tick

• Parameters: i (int) – the index of the Tick
• Returns: aTick – the i-th Tick
• Return type: float
• Raises: pyagrum.NotFound – If the index is greater than the number of Ticks

ticks()

• Returns: a tuple containing all the Ticks
• Return type: tuple[float, ...]

toFast()

Return the variable description in fast syntax (e.g. name[v1:v2:v3]).

• Returns: fast-syntax string representation
• Return type: str

toStringWithDescription()

• Returns: a description of the variable
• Return type: str

varType()

returns the type of variable

• Returns: the type of the variable.

  0: DiscretizedVariable, 1: LabelizedVariable, 2: IntegerVariable, 3: NumericalDiscreteVariable, 4: RangeVariable

• Return type: int

IntegerVariable

class pyagrum.IntegerVariable(*args)

IntegerVariable is a discrete random variable with a customizable sequence of int.

IntegerVariable(aName, aDesc=’’, values=None) -> IntegerVariable : Parameters: : - aName (str) – the name of the variable - aDesc (str) – the (optional) description of the variable - values (list of int) – the values to create

IntegerVariable(aIDRV) -> IntegerVariable : Parameters: : - aIDRV (pyagrum.IntegerVariable) – The pyagrum.IntegerVariable that will be copied

Examples

>>> import pyagrum as gum
>>> # creating a variable with 3 values : 1,34,142
>>> va=pyagrum.IntegerVariable('a','a integer variable',[1,34,142])
>>> print(va)
a:Integer(<1,34,142>)
>>> va.addValue(25)
(pyagrum.IntegerVariable@000001E4F5D07490) a:Integer(<1,25,34,142>)
>>> va.changeValue(34,43)
>>> print(va)
a:Integer(<1,25,43,142>)
>>> vb=pyagrum.IntegerVariable('b','b').addValue(34).addValue(142).addValue(1)
>>> print(vb)
b:Integer(<1,34,142>)
>>> vb.labels()
('1', '34', '142')

addValue(*args)

Add a value to the list of values for the variable.

• Parameters: value (int) – the new value
• Returns: the Integer variable
• Return type: pyagrum.IntegerVariable
• Raises: pyagrum.DuplicateElement – If the variable already contains the value

asDiscretizedVar()

Tries to cast the variable as a pyagrum.DiscretizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.DiscretizedVar
• Return type: DiscretizedVariable

asIntegerVar()

Tries to cast the variable as a pyagrum.IntegerVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.IntegerVar
• Return type: IntegerVariable

asLabelizedVar()

Tries to cast the variable as a pyagrum.LabelizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.LabelizedVar
• Return type: LabelizedVariable

asNumericalDiscreteVar()

Tries to cast the variable as a pyagrum.NumericalDiscreteVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.NumericalDiscreteVar
• Return type: NumericalDiscreteVariable

asRangeVar()

Tries to cast the variable as a pyagrum.RangeVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.RangeVar
• Return type: RangeVariable

belongs(x)

returns whether x belongs to the variable domain (if the variable is a pyagrum.RangeVariable)

• Parameters: x (int) – the value for which we want to know if it belongs to the variable domain
• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: bool
• Returns: bool : True if x belongs to the variable domain, False otherwise

changeValue(old_value, new_value)

• Parameters:
  • old_value (int) – the value to be changed
  • new_value (int) – the new value
• Return type: None

closestIndex(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the index of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the index of the closest value to value in the variable’s domain
• Return type: int
• Raises: pyagrum.NotImplementedYet – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

closestLabel(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the label of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the label of the closest value to value in the variable’s domain
• Return type: str
• Raises: pyagrum.InvalidArgument – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

description()

• Returns: the description of the variable
• Return type: str

domain()

• Returns: the domain of the variable
• Return type: str

domainSize()

• Returns: the number of modalities in the variable domain
• Return type: int

draw(x)

Draw a random float value from the x-th interval of the variable (only for pyagrum.DiscretizedVariable).

• Parameters: x (int) – the index of the interval from which to draw
• Returns: a random value drawn uniformly from the x-th interval
• Return type: float
• Raises:
  • NotImplementedError – if the variable is not a pyagrum.DiscretizedVariable
  • pyagrum.OutOfBounds – if x is not a valid interval index

empty()

• Returns: True if the domain size < 2
• Return type: bool

eraseValue(value)

• Parameters: value (int) – the value to erase. If the value is not in the domain, the function does nothing (no exception raised)
• Return type: None

eraseValues()

Remove all the domain.

• Return type: None

index(label)

• Parameters: label (str) – a label
• Returns: the index of the label
• Return type: int

integerDomain()

• Returns: the tuple of integer values that form the domain of this variable
• Return type: tuple[int, ...]

isEmpirical()

Returns whether the variable is empirical (if the variable is a pyagrum.DiscretizedVariable). I.e. whether its limits (max and min) are defined by data and accept any values below or above these limits, contrary to non-empirical discretized variables for which the limits are fixed and values outside of these limits are not accepted)

• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Return type: bool
• Returns: bool : whether the variable is empirical

isLabel(s)

returns whether s is a label of the variable

• Raises: NotImplementedError – If the variable is not a pyagrum.LabelizedVariable
• Return type: bool
• Returns: bool : True if s is a label of the variable, False otherwise
• Parameters: s (str)

isNumerical()

Returns True if the variable has a meaningful numerical representation, i.e. it is not a LabelizedVariable.

• Returns: True for DiscretizedVariable, IntegerVariable, NumericalDiscreteVariable and RangeVariable; False for LabelizedVariable.
• Return type: bool

isTick(x)

Returns whether x is a tick for the variable

• Parameters: x (float) – the value for which we want to know if it is a tick for the variable
• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Returns: bool : whether x is a tick for the variable, i.e. whether x belongs to the variable’s domain

isValue(value)

• Parameters: value (int) – the value to look at.
• Returns: True if the value is in the domain.
• Return type: bool

label(index)

• Parameters:
  • i (int) – the index of the label we wish to return
  • index (int)
• Returns: the label at the given index
• Return type: str
• Raises: pyagrum.OutOfBounds – If the variable does not contain the label

labels()

• Returns: a tuple containing the labels
• Return type: tuple[str, ...]

maxVal()

returns the maximum value of the variable domain (if the variable is a pyagrum.RangeVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: int
• Returns: int : the maximum value of the variable domain

minVal()

returns the minimum value of the variable domain (if the variable is a pyagrum.RangeVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: int
• Returns: int : the minimum value of the variable domain

name()

• Returns: the name of the variable
• Return type: str

numerical(index)

• Parameters: index (int) – an index
• Returns: the numerical representation of the value at the given index
• Return type: float

numericalDomain()

Returns the numerical domain of the variable (if the variable is a pyagrum.NumericalDiscreteVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.NumericalDiscreteVariable
• Return type: tuple[float, ...]
• Returns: tuple[float,…] : the numerical domain of the variable

posLabel(s)

Returns the index of the label s in the variable’s domain (if the variable is a pyagrum.LabelizedVariable)

• Raises:
  • NotImplementedError – If the variable is not a pyagrum.LabelizedVariable
  • NotFoundError – If the variable does not contain the label s
• Return type: int
• Returns: int : the index of the label s in the variable’s domain
• Parameters: s (str)

setDescription(theValue)

set the description of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

setName(theValue)

sets the name of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

stype()

• Returns: a description of its type
• Return type: str

tick(x)

Returns the x-th tick of the variable (if the variable is a pyagrum.DiscretizedVariable)

• Parameters: x (int) – the index of the tick we want to get
• Raises:
  • NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
  • pyagrum.OutOfBounds – If x is not a valid index for the ticks of the variable
• Returns: float : the x-th tick of the variable

ticks()

Returns the ticks of the variable

• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Return type: tuple[float, ...]
• Returns: tuple[float,…] : the ticks of the variable, i.e. the values that belong to the variable’s domain

toFast()

Return the variable description in fast syntax (e.g. name{v1|v2|...}).

• Returns: fast-syntax string representation
• Return type: str

toStringWithDescription()

• Returns: a description of the variable
• Return type: str

varType()

returns the type of variable

• Returns: the type of the variable.

  0: DiscretizedVariable, 1: LabelizedVariable, 2: IntegerVariable, 3: NumericalDiscreteVariable, 4: RangeVariable

• Return type: int

RangeVariable

class pyagrum.RangeVariable(*args)

RangeVariable represents a variable with a range of integers as domain.

RangeVariable(aName, aDesc,minVal, maxVal) -> RangeVariable : Parameters: : - aName (str) – the name of the variable - aDesc (str) – the description of the variable - minVal (int) – the minimal integer of the interval - maxVal (int) – the maximal integer of the interval

RangeVariable(aName, aDesc=’’) -> RangeVariable : Parameters: : - aName (str) – the name of the variable - aDesc (str) – the description of the variable
By default minVal=0 and maxVal=1

RangeVariable(aRV) -> RangeVariable : Parameters: : - aDV (RangeVariable) – the pyagrum.RangeVariable that will be copied

Examples

>>> import pyagrum as gum
>>> vI=pyagrum.RangeVariable('I','I in [4,10]',4,10)
>>> print(vI)
I:Range([4,10])
>>> vI.maxVal()
10
>>> vI.belongs(1)
False
>>> # where is the value 5 ?
>>> vI.index('5')
1
>>> vI.labels()
('4', '5', '6', '7', '8', '9', '10')

asDiscretizedVar()

Tries to cast the variable as a pyagrum.DiscretizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.DiscretizedVar
• Return type: DiscretizedVariable

asIntegerVar()

Tries to cast the variable as a pyagrum.IntegerVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.IntegerVar
• Return type: IntegerVariable

asLabelizedVar()

Tries to cast the variable as a pyagrum.LabelizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.LabelizedVar
• Return type: LabelizedVariable

asNumericalDiscreteVar()

Tries to cast the variable as a pyagrum.NumericalDiscreteVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.NumericalDiscreteVar
• Return type: NumericalDiscreteVariable

asRangeVar()

Tries to cast the variable as a pyagrum.RangeVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.RangeVar
• Return type: RangeVariable

belongs(val)

• Parameters: val (int) – the value to be tested
• Returns: True if the value in parameters belongs to the variable’s interval.
• Return type: bool

closestIndex(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the index of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the index of the closest value to value in the variable’s domain
• Return type: int
• Raises: pyagrum.NotImplementedYet – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

closestLabel(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the label of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the label of the closest value to value in the variable’s domain
• Return type: str
• Raises: pyagrum.InvalidArgument – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

description()

• Returns: the description of the variable
• Return type: str

domain()

• Returns: the domain of the variable
• Return type: str

domainSize()

• Returns: the number of modalities in the variable domain
• Return type: int

draw(x)

Draw a random float value from the x-th interval of the variable (only for pyagrum.DiscretizedVariable).

• Parameters: x (int) – the index of the interval from which to draw
• Returns: a random value drawn uniformly from the x-th interval
• Return type: float
• Raises:
  • NotImplementedError – if the variable is not a pyagrum.DiscretizedVariable
  • pyagrum.OutOfBounds – if x is not a valid interval index

empty()

• Returns: True if the domain size < 2
• Return type: bool

index(arg2)

• Parameters: arg2 (str) – a label
• Returns: the indice of the label
• Return type: int

integerDomain()

Returns the integer domain of the variable (if the variable is a pyagrum.integerVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.IntegerVariable
• Return type: tuple[int, ...]
• Returns: tuple[int,…] : the integer domain of the variable

isEmpirical()

Returns whether the variable is empirical (if the variable is a pyagrum.DiscretizedVariable). I.e. whether its limits (max and min) are defined by data and accept any values below or above these limits, contrary to non-empirical discretized variables for which the limits are fixed and values outside of these limits are not accepted)

• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Return type: bool
• Returns: bool : whether the variable is empirical

isLabel(s)

returns whether s is a label of the variable

• Raises: NotImplementedError – If the variable is not a pyagrum.LabelizedVariable
• Return type: bool
• Returns: bool : True if s is a label of the variable, False otherwise
• Parameters: s (str)

isNumerical()

Returns True if the variable has a meaningful numerical representation, i.e. it is not a LabelizedVariable.

• Returns: True for DiscretizedVariable, IntegerVariable, NumericalDiscreteVariable and RangeVariable; False for LabelizedVariable.
• Return type: bool

isTick(x)

Returns whether x is a tick for the variable

• Parameters: x (float) – the value for which we want to know if it is a tick for the variable
• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Returns: bool : whether x is a tick for the variable, i.e. whether x belongs to the variable’s domain

isValue(x)

Returns whether x is a value for the variable

• Parameters: x (float) – the value for which we want to know if it is a value for the variable
• Raises: NotImplementedError – If the variable is not a pyagrum.NumericalDiscreteVariable or a pyagrum.IntegerVariable
• Return type: bool
• Returns: bool : whether x is a value for the variable, i.e. whether x belongs to the variable domain

label(index)

• Parameters:
  • indice (int) – the index of the label we wish to return
  • index (int)
• Returns: the indice-th label
• Return type: str
• Raises: pyagrum.OutOfBounds – If the variable does not contain the label

labels()

• Returns: a tuple containing the labels
• Return type: tuple[str, ...]

maxVal()

• Returns: the upper bound of the variable.
• Return type: int

minVal()

• Returns: the lower bound of the variable
• Return type: int

name()

• Returns: the name of the variable
• Return type: str

numerical(index)

• Parameters: index (int) – an index
• Returns: the numerical representation of the value at the given index
• Return type: float

numericalDomain()

Returns the numerical domain of the variable (if the variable is a pyagrum.NumericalDiscreteVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.NumericalDiscreteVariable
• Return type: tuple[float, ...]
• Returns: tuple[float,…] : the numerical domain of the variable

posLabel(s)

Returns the index of the label s in the variable’s domain (if the variable is a pyagrum.LabelizedVariable)

• Raises:
  • NotImplementedError – If the variable is not a pyagrum.LabelizedVariable
  • NotFoundError – If the variable does not contain the label s
• Return type: int
• Returns: int : the index of the label s in the variable’s domain
• Parameters: s (str)

setDescription(theValue)

set the description of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

setMaxVal(maxVal)

Set a new value of the upper bound

• Parameters: maxVal (int) – The new value of the upper bound
• Return type: None

Warning

An error should be raised if the value is lower than the lower bound.

setMinVal(minVal)

Set a new value of the lower bound

• Parameters: minVal (int) – The new value of the lower bound
• Return type: None

Warning

An error should be raised if the value is higher than the upper bound.

setName(theValue)

sets the name of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

stype()

• Returns: a description of its type
• Return type: str

tick(x)

Returns the x-th tick of the variable (if the variable is a pyagrum.DiscretizedVariable)

• Parameters: x (int) – the index of the tick we want to get
• Raises:
  • NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
  • pyagrum.OutOfBounds – If x is not a valid index for the ticks of the variable
• Returns: float : the x-th tick of the variable

ticks()

Returns the ticks of the variable

• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Return type: tuple[float, ...]
• Returns: tuple[float,…] : the ticks of the variable, i.e. the values that belong to the variable’s domain

toFast()

Return the variable description in fast syntax (e.g. name[min,max]).

• Returns: fast-syntax string representation
• Return type: str

toStringWithDescription()

• Returns: a description of the variable
• Return type: str

varType()

returns the type of variable

• Returns: the type of the variable.

  0: DiscretizedVariable, 1: LabelizedVariable, 2: IntegerVariable, 3: NumericalDiscreteVariable, 4: RangeVariable

• Return type: int

NumericalDiscreteVariable

class pyagrum.NumericalDiscreteVariable(*args)

NumericalDiscreteVariable is a discrete random variable with a customizable sequence of float.

NumericalDiscreteVariable(aName, aDesc=’’, values=None) -> NumericalDiscreteVariable : Parameters: : - aName (str) – the name of the variable - aDesc (str) – the (optional) description of the variable - values (list of float) – the values to create equivalent to fast syntax {v1|v2|v3|…|vn}

NumericalDiscreteVariable(aName, aDesc=’’, first, last, nbr) -> NumericalDiscreteVariable : Parameters: : - aName (str) – the name of the variable - aDesc (str) – the (optional) description of the variable - first (float) – specify a list of floats from first to last in nbr steps. - last (float) – - nbr (int) – equivalent to fast syntax {first:last:nbr}

NumericalDiscreteVariable(aNDRV) -> NumericalDiscreteVariable : Parameters: : - aNDRV (pyagrum.NumericalDiscreteVariable) – The pyagrum.NumericalDiscreteVariable that will be copied

Examples

>>> import pyagrum as gum
>>> # creating a variable with 3 values : 1.5,3.14,1.42
>>> va=pyagrum.NumericalDiscreteVariable('a','a numerica variable',[1.5,3.14,1.42])
>>> print(va)
a:NumericalDiscrete({1.42|1.5|3.14})
>>> va.addValue(2.01)
(pyagrum.NumericalDiscreteVariable@0x55ea157b8d60) a:NumericalDiscrete({1.42|1.5|2.01|3.14})
>>> va.changeValue(3.14,3.1415)
>>> print(va)
a:NumericalDiscrete({1.42|1.5|2.01|3.1415})
>>> vb=pyagrum.NumericalDiscreteVariable('b','b').addValue(3.14).addValue(1.42).addValue(1.5)
>>> print(vb)
b:NumericalDiscrete({1.42|1.5|3.14})
>>> vb.labels()
('1.42', '1.5', '3.14')
>>>> vc=pyagrum.NumericalDiscreteVariable('c','c',1.2,3.8,5)
>>> print(vc)
c:NumericalDiscrete({1.2|1.85|2.5|3.15|3.8})

addValue(*args)

Add a value to the list of values for the variable.

• Parameters: value (double) – the new value
• Returns: the Integer variable
• Return type: pyagrum.IntegerVariable
• Raises: pyagrum.DuplicateElement – If the variable already contains the value

asDiscretizedVar()

Tries to cast the variable as a pyagrum.DiscretizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.DiscretizedVar
• Return type: DiscretizedVariable

asIntegerVar()

Tries to cast the variable as a pyagrum.IntegerVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.IntegerVar
• Return type: IntegerVariable

asLabelizedVar()

Tries to cast the variable as a pyagrum.LabelizedVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.LabelizedVar
• Return type: LabelizedVariable

asNumericalDiscreteVar()

Tries to cast the variable as a pyagrum.NumericalDiscreteVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.NumericalDiscreteVar
• Return type: NumericalDiscreteVariable

asRangeVar()

Tries to cast the variable as a pyagrum.RangeVar.

• Raises: pyagrum.OperationNotAllowed if this is not possible –
• Returns: the variable as a pyagrum.RangeVar
• Return type: RangeVariable

belongs(x)

returns whether x belongs to the variable domain (if the variable is a pyagrum.RangeVariable)

• Parameters: x (int) – the value for which we want to know if it belongs to the variable domain
• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: bool
• Returns: bool : True if x belongs to the variable domain, False otherwise

changeValue(old_value, new_value)

• Parameters:
  • old_value (float) – the value to be changed
  • new_value (float) – the new value
• Return type: None

closestIndex(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the index of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the index of the closest value to value in the variable’s domain
• Return type: int
• Raises: pyagrum.NotImplementedYet – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

closestLabel(val)

For numerical discrete variables (all except pyagrum.LabelizedVariable), this method returns the label of the closest value to a given float value in the variable’s domain.

• Parameters: val (float) – the value for which we want to find the closest index
• Returns: the label of the closest value to value in the variable’s domain
• Return type: str
• Raises: pyagrum.InvalidArgument – if the variable is not numerical discrete (i.e., if it is a pyagrum.LabelizedVariable).

description()

• Returns: the description of the variable
• Return type: str

domain()

• Returns: the domain of the variable
• Return type: str

domainSize()

• Returns: the number of modalities in the variable domain
• Return type: int

draw(x)

Draw a random float value from the x-th interval of the variable (only for pyagrum.DiscretizedVariable).

• Parameters: x (int) – the index of the interval from which to draw
• Returns: a random value drawn uniformly from the x-th interval
• Return type: float
• Raises:
  • NotImplementedError – if the variable is not a pyagrum.DiscretizedVariable
  • pyagrum.OutOfBounds – if x is not a valid interval index

empty()

• Returns: True if the domain size < 2
• Return type: bool

eraseValue(value)

• Parameters: value (float) – the value to erase. If the value is not in the domain, the function does nothing (no exception raised)
• Return type: None

eraseValues()

Remove all the domain.

• Return type: None

index(label)

• Parameters: label (str) – a label
• Returns: the index of the label
• Return type: int

integerDomain()

Returns the integer domain of the variable (if the variable is a pyagrum.integerVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.IntegerVariable
• Return type: tuple[int, ...]
• Returns: tuple[int,…] : the integer domain of the variable

isEmpirical()

Returns whether the variable is empirical (if the variable is a pyagrum.DiscretizedVariable). I.e. whether its limits (max and min) are defined by data and accept any values below or above these limits, contrary to non-empirical discretized variables for which the limits are fixed and values outside of these limits are not accepted)

• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Return type: bool
• Returns: bool : whether the variable is empirical

isLabel(s)

returns whether s is a label of the variable

• Raises: NotImplementedError – If the variable is not a pyagrum.LabelizedVariable
• Return type: bool
• Returns: bool : True if s is a label of the variable, False otherwise
• Parameters: s (str)

isNumerical()

Returns True if the variable has a meaningful numerical representation, i.e. it is not a LabelizedVariable.

• Returns: True for DiscretizedVariable, IntegerVariable, NumericalDiscreteVariable and RangeVariable; False for LabelizedVariable.
• Return type: bool

isTick(x)

Returns whether x is a tick for the variable

• Parameters: x (float) – the value for which we want to know if it is a tick for the variable
• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Returns: bool : whether x is a tick for the variable, i.e. whether x belongs to the variable’s domain

isValue(value)

• Parameters: value (float) – the value to look at.
• Returns: True if the value is in the domain.
• Return type: bool

label(index)

• Parameters:
  • i (int) – the index of the label we wish to return
  • index (int)
• Returns: the label at the given index
• Return type: str
• Raises: pyagrum.OutOfBounds – If the variable does not contain the label

labels()

• Returns: a tuple containing the labels
• Return type: tuple[str, ...]

maxVal()

returns the maximum value of the variable domain (if the variable is a pyagrum.RangeVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: int
• Returns: int : the maximum value of the variable domain

minVal()

returns the minimum value of the variable domain (if the variable is a pyagrum.RangeVariable)

• Raises: NotImplementedError – If the variable is not a pyagrum.RangeVariable
• Return type: int
• Returns: int : the minimum value of the variable domain

name()

• Returns: the name of the variable
• Return type: str

numerical(index)

• Parameters: index (int) – an index
• Returns: the numerical representation of the value at the given index
• Return type: float

numericalDomain()

• Returns: the tuple of float values that form the domain of this variable
• Return type: tuple[float, ...]

posLabel(s)

Returns the index of the label s in the variable’s domain (if the variable is a pyagrum.LabelizedVariable)

• Raises:
  • NotImplementedError – If the variable is not a pyagrum.LabelizedVariable
  • NotFoundError – If the variable does not contain the label s
• Return type: int
• Returns: int : the index of the label s in the variable’s domain
• Parameters: s (str)

setDescription(theValue)

set the description of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

setName(theValue)

sets the name of the variable.

• Parameters: theValue (str) – the new description of the variable
• Return type: None

stype()

• Returns: a description of its type
• Return type: str

tick(x)

Returns the x-th tick of the variable (if the variable is a pyagrum.DiscretizedVariable)

• Parameters: x (int) – the index of the tick we want to get
• Raises:
  • NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
  • pyagrum.OutOfBounds – If x is not a valid index for the ticks of the variable
• Returns: float : the x-th tick of the variable

ticks()

Returns the ticks of the variable

• Raises: NotImplementedError – If the variable is not a pyagrum.DiscretizedVariable
• Return type: tuple[float, ...]
• Returns: tuple[float,…] : the ticks of the variable, i.e. the values that belong to the variable’s domain

toFast()

Return the variable description in fast syntax (e.g. name{v1|v2|...}).

• Returns: fast-syntax string representation
• Return type: str

toStringWithDescription()

• Returns: a description of the variable
• Return type: str

varType()

returns the type of variable

• Returns: the type of the variable.

  0: DiscretizedVariable, 1: LabelizedVariable, 2: IntegerVariable, 3: NumericalDiscreteVariable, 4: RangeVariable

• Return type: int