Dakota variable types

Dakota variables are the parameter sets to be iterated by a particular analysis method.

Variables base class

An abstract base class for all Dakota variable types.

class dakotathon.variables.base.VariablesBase(variables='continuous_design', descriptors=(), **kwargs)

Bases: object

Describe features common to all Dakota variable types.

__init__(variables='continuous_design', descriptors=(), **kwargs)

Create default variables parameters.

descriptors : str or tuple or list of str, optional

Labels for the variables.

variables : str, optional

The type of parameter set (default is ‘continuous_design’).

__str__()

Define the variables block of a Dakota input file.

descriptors

Labels attached to Dakota variables.

Continuous design

Implementation of a Dakota continous design variable.

class dakotathon.variables.continuous_design.ContinuousDesign(descriptors=('x1', 'x2'), initial_point=None, lower_bounds=None, upper_bounds=None, **kwargs)

Bases: dakotathon.variables.base.VariablesBase

Define attributes for Dakota continous design variables.

Continuous variables are defined by a real interval and are changed during the search for the optimal design.

__init__(descriptors=('x1', 'x2'), initial_point=None, lower_bounds=None, upper_bounds=None, **kwargs)

Create the parameter set for a continuous design variable.

descriptors : str or tuple or list of str, optional

Labels for the variables.

initial_point : tuple or list of numbers

Start points used by study variables.

lower_bounds : tuple or list of numbers

Minimum values used by the study variables.

upper_bounds : tuple or list of numbers

Maximum values used by the study variables.

**kwargs

Optional keyword arguments.

Create a default ContinuousDesign instance with:

>>> v = ContinuousDesign()

__str__()

Define the variables block for continous design variables.

Display the variables block created by a default instance of ContinuousDesign:

>>> v = ContinuousDesign()
>>> print(v)
variables
  continuous_design = 2
    descriptors = 'x1' 'x2'
    initial_point = -0.3 0.2
<BLANKLINE>
<BLANKLINE>

dakotathon.variables.base.VariablesBase.__str__

initial_point

Start points used by study variables.

lower_bounds

Minimum values of study variables.

upper_bounds

Maximum values of study variables.

Uniform uncertain

Implementation of a Dakota uniform uncertain variable.

class dakotathon.variables.uniform_uncertain.UniformUncertain(descriptors=('x1', 'x2'), lower_bounds=(-2.0, -2.0), upper_bounds=(2.0, 2.0), initial_point=None, **kwargs)

Bases: dakotathon.variables.base.VariablesBase

Define attributes for Dakota uniform uncertain variables.

The distribution lower and upper bounds are required specifications; the initial point is optional.

__init__(descriptors=('x1', 'x2'), lower_bounds=(-2.0, -2.0), upper_bounds=(2.0, 2.0), initial_point=None, **kwargs)

Create the parameter set for a uniform uncertain variable.

descriptors : str or tuple or list of str, optional

Labels for the variables.

initial_point : tuple or list of numbers, optional

Start points used by study variables.

lower_bounds : tuple or list of numbers

Minimum values used by the study variables.

upper_bounds : tuple or list of numbers

Maximum values used by the study variables.

**kwargs

Optional keyword arguments.

Create a default instance of UniformUncertain with:

>>> v = UniformUncertain()

__str__()

Define the variables block for a uniform uncertain variable.

Display the variables block created by a default instance of UniformUncertain:

>>> v = UniformUncertain()
>>> print(v)
variables
  uniform_uncertain = 2
    descriptors = 'x1' 'x2'
    lower_bounds = -2.0 -2.0
    upper_bounds = 2.0 2.0
<BLANKLINE>
<BLANKLINE>

dakotathon.variables.base.VariablesBase.__str__

initial_point

Start points used by study variables.

lower_bounds

Minimum values of study variables.

upper_bounds

Maximum values of study variables.

Normal uncertain

Implementation of a Dakota normal uncertain variable.

class dakotathon.variables.normal_uncertain.NormalUncertain(descriptors=('x1', 'x2'), means=(0.0, 0.0), std_deviations=(1.0, 1.0), lower_bounds=None, upper_bounds=None, initial_point=None, **kwargs)

Bases: dakotathon.variables.base.VariablesBase

Define attributes for Dakota normal uncertain variables.

The means and standard deviations are required specifications; the initial point, and the distribution lower and upper bounds are optional.

For vector and centered parameter studies, an inferred initial starting point is needed for uncertain variables. These variables are initialized to their means for these studies.

__init__(descriptors=('x1', 'x2'), means=(0.0, 0.0), std_deviations=(1.0, 1.0), lower_bounds=None, upper_bounds=None, initial_point=None, **kwargs)

Create the parameter set for a normal uncertain variable.

descriptors : str or tuple or list of str, optional

Labels for the variables.

means : tuple or list of numbers

First parameter of Gaussian distribution.

std_deviations : tuple or list of numbers

Second parameter of Gaussian distribution.

lower_bounds : tuple or list of numbers, optional

Minimum values used by the study variables.

upper_bounds : tuple or list of numbers, optional

Maximum values used by the study variables.

initial_point : tuple or list of numbers, optional

Start points used by study variables.

**kwargs

Optional keyword arguments.

Create a default instance of NormalUncertain with:

>>> v = NormalUncertain()

__str__()

Define the variables block for a normal uncertain variable.

Display the variables block created by a default instance of NormalUncertain:

>>> v = NormalUncertain()
>>> print(v)
variables
  normal_uncertain = 2
    descriptors = 'x1' 'x2'
    means = 0.0 0.0
    std_deviations = 1.0 1.0
<BLANKLINE>
<BLANKLINE>

dakotathon.variables.base.VariablesBase.__str__

initial_point

Start points used by study variables.

lower_bounds

Minimum values of study variables.

means

Mean values of study variables.

std_deviations

Standard deviations of study variables.

upper_bounds

Maximum values of study variables.