simulators#

This module provides Simulator, make_simulator(), and several other kinds of Simulator implementations, which serve as generative mathematical models, or data generating processes, with their primary function being to sample data.

Functions

make_simulator()

This is a dispatch function that will accept a list of simulators (callables) returning dictionaries with simulated outputs.

Classes

BernoulliGLM([T, scale_by_T, rng])

Bernoulli GLM simulated benchmark See: https://arxiv.org/pdf/2101.04653.pdf, Task T.5

BernoulliGLMRaw([T, rng])

Bernoulli GLM raw simulated benchmark.

GaussianLinear([D, prior_scale, n_obs, ...])

Gaussian Linear simulated benchmark See: https://arxiv.org/pdf/2101.04653.pdf, Task T.1

GaussianLinearUniform([D, lower_bound, ...])

Gaussian Linear Uniform simulated benchmark See: https://arxiv.org/pdf/2101.04653.pdf, Task T.2

GaussianMixture([D, lower_bound, ...])

Gaussian Mixture simulated benchmark See: https://arxiv.org/pdf/2101.04653.pdf, Task T.7

HierarchicalSimulator(hierarchy)

Initialize the hierarchical simulator with a sequence of simulators.

InverseKinematics([scales, l1, l2, l3, rng])

Inverse Kinematics simulated benchmark See: https://arxiv.org/pdf/2101.10763.pdf

LambdaSimulator(sample_fn, *[, is_batched])

Implements a simulator based on a sampling function.

LotkaVolterra([X0, Y0, T, subsample, ...])

Lotka Volterra simulated benchmark.

ModelComparisonSimulator(simulators[, p, ...])

Wraps a sequence of simulators for use with a model comparison approximator.

SIR([N, T, I0, R0, subsample, total_count, ...])

SIR simulated benchmark See: https://arxiv.org/pdf/2101.04653.pdf, Task T.9

SLCP([lower_bound, upper_bound, n_obs, ...])

SLCP simulated benchmark See https://arxiv.org/pdf/2101.04653.pdf, Task T.3

SLCPDistractors([lower_bound, upper_bound, ...])

SLCP Distractors simulated benchmark See: https://arxiv.org/pdf/2101.04653.pdf, Task T.4

SequentialSimulator(simulators[, ...])

Combines multiple simulators into one, sequentially.

Simulator()

TwoMoons([lower_bound, upper_bound, rng])

Two moons simulated benchmark.