Compact

class Compact(*args)

Compact history storage strategy.

Parameters:
Nint

minimum number of points to store.

Notes

The compact strategy stores a regularly spaced sub-sample where the minimum size of the stored numerical sample is N. OpenTURNS proceeds as follows :

  1. it stores the first 2N simulations : the size of the stored sample is 2N,

  2. it selects only 1 out of 2 of the stored simulations : then the size of the stored sample decreases to N (this is the compact step),

  3. it stores the next N simulations when selecting 1 out of 2 of the next simulations : the size of the stored sample is 2N,

  4. it selects only 1 out of 2 of the stored simulations : then the size of the stored sample decreases to N,

  5. it stores the next N simulations when selecting 1 out of 4 of the next simulations : the size of the stored sample is 2N,

  6. then it keeps on until reaching the stopping criteria.

The stored numerical sample will have a size within N and 2N if at least one cycle has been done, else it will be at most N.

Methods

clear()

Clear the stored points.

getClassName()

Accessor to the object's name.

getHalfMaximumSize()

Accessor to the half maximum number of points to store.

getIndex()

Accessor to the index.

getName()

Accessor to the object's name.

getSample()

Accessor to the stored sample.

hasName()

Test if the object is named.

setDimension(dimension)

Set the dimension of points to store.

setName(name)

Accessor to the object's name.

store(*args)

Store points or samples.

__init__(*args)
clear()

Clear the stored points.

Notes

It erases the previously stored points

getClassName()

Accessor to the object’s name.

Returns:
class_namestr

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

getHalfMaximumSize()

Accessor to the half maximum number of points to store.

Returns:
Nint

The half maximum number of points to store.

getIndex()

Accessor to the index.

Returns:
indexint

The number of the stored points.

getName()

Accessor to the object’s name.

Returns:
namestr

The name of the object.

getSample()

Accessor to the stored sample.

Returns:
sampleSample

Numerical sample which is the collection of points stored by the history strategy.

hasName()

Test if the object is named.

Returns:
hasNamebool

True if the name is not empty.

setDimension(dimension)

Set the dimension of points to store.

This method must be called before calling the store method.

Parameters:
dimension: int

Dimension of points to store

Notes

It erases the previously stored points

setName(name)

Accessor to the object’s name.

Parameters:
namestr

The name of the object.

store(*args)

Store points or samples.

Parameters:
datasequence of float or 2-d sequence of float

Point or sample to store.

Notes

It adds a unique point or all the point of the sample in the natural order to the history.

Examples using the class

Specify a simulation algorithm

Specify a simulation algorithm

Exploitation of simulation algorithm results

Exploitation of simulation algorithm results