KFold

class KFold(*args)

K-fold.

Parameters:
kpositive integer

Number of folds in which the sample is split. If not provided, default is k = 10.

Methods

getClassName()

Accessor to the object's name.

getK()

Accessor to the number of folds.

getName()

Accessor to the object's name.

hasName()

Test if the object is named.

run(x, y, weight, basis, indices)

Run the algorithm.

setK(p)

Accessor to the number of folds.

setName(name)

Accessor to the object's name.

See also

FittingAlgorithm, CorrectedLeaveOneOut

Notes

KFold inherits from FittingAlgorithm.

Examples

>>> import openturns as ot
>>> size = 100
>>> xuniform = ot.Uniform(0.9, 1.1)
>>> x = xuniform.getSample(size)
>>> yuniform = ot.Uniform(1.9, 2.1)
>>> y = yuniform.getSample(size)
>>> w = [1.0] * size
>>> f = ot.SymbolicFunction(['x'], ['2.0 * x'])
>>> basis = [f]
>>> indices = [0]
>>> fittingAlgo = ot.KFold()
>>> result = fittingAlgo.run(x, y, w, basis, indices)
__init__(*args)
getClassName()

Accessor to the object’s name.

Returns:
class_namestr

The object class name (object.__class__.__name__).

getK()

Accessor to the number of folds.

Returns:
kint

Number of folds in which the sample is split.

getName()

Accessor to the object’s name.

Returns:
namestr

The name of the object.

hasName()

Test if the object is named.

Returns:
hasNamebool

True if the name is not empty.

run(x, y, weight, basis, indices)

Run the algorithm.

Parameters:
x2-d sequence of float

Input sample

y2-d sequence of float

Output sample

weightsequence of float

Weights associated to the outputs

psisequence of Function

Basis

indicessequence of int

Indices of the basis

Returns:
measurefloat

Fitting measure

setK(p)

Accessor to the number of folds.

Parameters:
kint

Number of folds in which the sample is split.

setName(name)

Accessor to the object’s name.

Parameters:
namestr

The name of the object.

Examples using the class

Trend computation

Trend computation