Generator Functions

Here we list many generator functions included with libEnsemble.

Important

See the API for generator functions here.

Sampling

• Basic sampling

  Various generators for sampling a space. The non-persistent function is called as needed.

• Persistent sampling

  Various persistent generators (persists on a worker) for sampling a space. After the initial batch each generator creates p new random points for every p points that are returned.

• Persistent sampling with variable resources

  Various persistent sampling generators that assign different resources to each simulation.

Optimization

• APOSMM

  Asynchronously Parallel Optimization Solver for finding Multiple Minima (APOSMM).

• Ax Multitask

  Bayesian optimization with a Gaussian process driven by an Ax multi-task algorithm.

• DEAP-NSGA-II

  Distributed evolutionary algorithms (community example)

• Distributed optimization

  Distributed optimization methods for minimizing sums of convex functions. (community example)

• uniform_or_localopt

  Samples uniformly in non-persistent mode then runs an NLopt local optimization runs in persistent mode.

• VTMOP

  Multiobjective multidisciplinary design optimization using the VTMOP Fortran package. (community example)

• ytopt

  Bayesian Optimization package for determining optimal input parameter configurations for applications/executables using ytopt. (community example)

Modeling and Approximation

• Finite-difference parameter finder

  Uses ECNoise to determine a suitable finite difference parameters for a mapping F from R^n to R^m.

• gpCAM

  Gaussian Process-based adaptive sampling using gpcam.

• surmise

  Modular Bayesian calibration/inference framework using Surmise (demonstration of cancelling previous issued simulations).

• Tasmanian

  Evaluates points generators by the Tasmanian sparse grid library