SALib.sample.radial.radial_sobol module

SALib.sample.radial.radial_sobol.sample(problem: Dict, N: int, R=4, skip_num: int = 0, seed: int | None = None)

Generates N sobol samples for a Radial OAT approach.

Notes

Compatible with:

• SALib.analyze.sobol_jansen.analyze() (the Jansen sensitivity estimator)

• SALib.analyze.radial_ee.analyze() (Elementary Effects)

References

[1]

Campolongo, F., Saltelli, A., Cariboni, J., 2011. From screening to quantitative sensitivity analysis: A unified approach. Computer Physics Communications 182, 978–988. https://www.sciencedirect.com/science/article/pii/S0010465510005321 DOI: 10.1016/j.cpc.2010.12.039

Parameters:

• problem (dict) – SALib problem specification

• N (int) – The number of sample sets to generate. It is assumed here that N = r, where r is the number of points/trajectories.

• R (int) – Number of rows in Sobol random matrix to shift downwards. Defaults to 4 (as given in [1])

• skip_num (int) – Number of sobol sequence values to skip When conducting a sequential sensitivity analysis, this is the previous number of samples used

• seed (int) – Seed value to use for np.random.seed

• Example (Usage)

• -------------

• ```python –

• sample(problem (>>> X =)

• N

• seed)

• ``` –

• hold (X will now)

• [ – [x_{1,1}, x_{1,2}, …, x_{1,p}] [b_{1,1}, x_{1,2}, …, x_{1,p}] [x_{1,1}, b_{1,2}, …, x_{1,p}] [x_{1,1}, x_{1,2}, …, b_{1,p}] … [x_{N,1}, x_{N,2}, …, x_{N,p}] [b_{N,1}, x_{N,2}, …, x_{N,p}] [x_{N,1}, b_{N,2}, …, x_{N,p}] [x_{N,1}, x_{N,2}, …, b_{N,p}]

• ]

• and (where p denotes the number of parameters as specified in problem)

• values. (b represents perturbed)

• baseline. (The first parameter set in each sample set acts as the)

• X. (We can now run the model using the values in)

• N(p+1). (The total number of model evaluations will be)

Returns:

numpy.ndarray

Return type:

An array of samples