MLMC package¶
The mlmc package provides tools to realize the Multilevel Monte Carlo method.
Subpackages¶
plot |
Subpackage provides plot functions to display pdf, violinplot, … |
quantity |
Subpackage provides methods to represent and handle a quantity of interest. |
random |
Subpackage provides random field generation and GSTools library interface |
sim |
Contains a parent simulation class and a specific synthetic simulation |
tool |
Contains classes that provide an interface to other resources such as HDF5, Gmsh, PBS, … |
Classes¶
Sampler¶
Sampler(sample_storage, sampling_pool, …) |
Manages samples scheduling, results collection, and result storage. |
SamplingPool¶
SamplingPool([work_dir, debug]) |
Determining the runtime environment of samples, eg single process, multiple processes, running PBS, … |
OneProcessPool([work_dir, debug]) |
|
ProcessPool(n_processes[, work_dir, debug]) |
Suitable for local parallel sampling for simulations WITHOUT external program call |
SamplingPoolPBS¶
SamplingPoolPBS(work_dir[, debug]) |
Sampling pool PBS (Portable batch system) runtime environment |
SampleStorage¶
SampleStorage |
Provides methods to store and retrieve sample’s data |
Memory() |
Sample’s data are stored in the main memory |
SampleStorageHDF¶
SampleStorageHDF(file_path) |
Sample’s data are stored in a HDF5 file |
Estimate¶
Estimate(quantity, sample_storage[, moments_fn]) |
Provides wrapper methods for moments estimation, pdf approximation, … |
Moments¶
Moments(size, domain[, log, safe_eval]) |
Class for calculating moments of a random variable |
Monomial(size[, domain, ref_domain, log, …]) |
Monomials generalized moments |
Fourier(size[, domain, ref_domain, log, …]) |
Fourier functions generalized moments |
Legendre(size, domain[, ref_domain, log, …]) |
Legendre polynomials generalized moments |
LevelSimulation¶
LevelSimulation(config_dict, Any], …) |
This class is used to pass simulation data at a given level between a Sampler and a SamplingPool User shouldn’t change this class |