*fast* syntax

Quick specification of discrete random variables

aGrUM/pyAgrum offers a so-called fast syntax that allows to quickly and compactly specify prototypes of graphical models.

The type of the (discrete) random variables can be specifiy with different syntaxes:

• by default, a variable is a pyagrum.RangeVariable using the default domain size (second argument of the functions).
• with a[10], the variable is a pyagrum.RangeVariable using 10 as domain size (from 0 to 9)
• with a[3,7], the variable is a pyagrum.RangeVariable using the integer values from 3 to 7.
• with a[1,3.14,5,6.2], the variable is a pyagrum.DiscretizedVariable using the given ticks (at least 3 values).
• with a[1:6.15:9], the variable is a pyagrum.DiscretizedVariable using the ticks from 1 to 6.15 in 9 steps (the domain size is 9).
• with a+[1,3.14,5,6.2], the variable is a pyagrum.DiscretizedVariable using the given ticks (at least 3 values), empirical.
• with a+[1:6.15:9], the variable is a pyagrum.DiscretizedVariable using the ticks from 1 to 6.15 in 9 steps (the domain size is 9), empirical.
• with a{top|middle|bottom}, the variable is a pyagrum.LabelizedVariable using the given labels (here : top, middle and bottom).
• with a{-1|5|0|3}, the variable is a pyagrum.IntegerVariable using the sorted given values.
• with a{-0.5|5.01|0|3.1415}, the variable is a pyagrum.NumericalDiscreteVariable using the sorted given values.
• with a{-0.5:3.1415:5}, the variable is a pyagrum.NumericalDiscreteVariable using the values from -0.5 to 3.1415 in 5 steps (the domain size is 5).

pyagrum.fastVariable(*args)

Use fast syntax to add a variable in the BayesNet.

• Raises: pyagrum.NotAllowed –
• Parameters:
  • fast_description (str) – string following fast syntax description
  • default_nbrmod (int) – nbr of modality if fast_description does not indicate it. default_nbrmod=1 is the way to create a variable with only one value (for instance for reward in influence diagram).
• Return type: DiscreteVariable

Examples

>>> print(pyagrum.fastVariable('A{On|Off|Defun}'))
A:Labelized({On|Off|Defun})
>>> print(pyagrum.fastVariable('A{3.14|0|1.15}'))
A:NumericalDiscrete({0|1.15|3.14})
>>> print(pyagrum.fastVariable('A{1.2:5.2:5}}'))
A:NumericalDiscrete({1.2|2.2|3.2|4.2|5.2})
>>> print(pyagrum.fastVariable('A{1|3|9}'))
A:Integer({1|3|9})
>>> print(pyagrum.fastVariable('A[4,6]'))
A:Range([4,6])
>>> print(pyagrum.fastVariable('A[5]'))
A:Range([0,4])
>>> print(pyagrum.fastVariable('A[4,6,10]'))
A:Discretized(<[4;6[,[6;10]>)
>>> print(pyagrum.fastVariable('A[1:6:5]'))
A:Discretized(<[1;2[,[2;3[,[3;4[,[4;5[,[5;6]>)

Quick specification of (randomly parameterized) graphical models

These fastPrototype aGrUM’s methods have also been wrapped in functions of pyagrum.

pyagrum.fastBN("A[10]->B<-C{top|middle|bottom};B->D")

Note

• If the dot-like string contains such a specification more than once for a variable, the first specification will be used.
• the CPTs are randomly generated.

pyagrum.fastBN(structure, domain=‘[2]’)

Create a Bayesian network with a dot-like syntax which specifies: : - the structure ‘a->b->c;b->d<-e;’,

• the type of the variables with different syntax (cf documentation).

Examples

>>> import pyagrum as gum
>>> bn=pyagrum.fastBN('A->B[1,3]<-C{yes|No}->D[2,4]<-E[1,2.5,3.9]',6)

• Parameters:
  • structure (str) – the string containing the specification
  • domain (int or str) – the default domain size (int) or domain specification (str) for variables (default is “[2]”
• Returns: the resulting bayesian network
• Return type: pyagrum.BayesNet

pyagrum.fastMRF(structure, domain=‘[2]’)

Create a Markov random field with a modified dot-like syntax which specifies: : - the structure ‘a-b-c;b-d;c-e;’ where each chain ‘a-b-c’ specifies a factor,

• the type of the variables with different syntax (cf documentation).

Examples

>>> import pyagrum as gum
>>> bn=pyagrum.fastMRF('A--B[1,3]--C{yes|No};C--D[2,4]--E[1,2.5,3.9]',6)

• Parameters:
  • structure (str) – the string containing the specification
  • domain (int or str) – the default domain size (int) or domain specification (str) for variables (default is “[2]”
• Returns: the resulting Markov random field
• Return type: pyagrum.MarkovRandomField

pyagrum.fastID(structure, domain=‘[2]’)

Create an Influence Diagram with a modified dot-like syntax which specifies: : - the structure and the type of the variables following fast syntax,

• a prefix for the type of node (chance/decision/utiliy nodes):
  • a : a chance node named ‘a’ (by default)
  • $a : a utility node named ‘a’
  • *a : a decision node named ‘a’

Examples

>>> import pyagrum as gum
>>> bn=pyagrum.fastID('A->B[1,3]<-*C{yes|No}->$D<-E[1,2.5,3.9]',6)

• Parameters:
  • structure (str) – the string containing the specification
  • domain (int or str) – the default domain size (int) or domain specification (str) for variables (default is “[2]”
• Returns: the resulting Influence Diagram
• Return type: pyagrum.InfluenceDiagram