utils.benchmark.vegas_benchmarks module

class VegasBenchmarker(stratified=False, benchmark_time=False)[source]

Bases: utils.benchmark.benchmarker.Benchmarker

Benchmark by comparing with VEGAS

benchmark_method(d, integrand, integrator_config=None, integrand_params=None, n_batch=100000, keep_history=False, device=device(type='cpu'))[source]

Benchmarking class for comparing with VEGAS

Parameters

  • d (int) –

  • integrand (utils.integrands.abstract.Integrand) –

  • integrator_config (dict) –

  • integrand_params (dict) –

  • n_batch (int) –

  • keep_history (bool) –

  • device (torch.device) –

Returns

Return type

utils.benchmark.vegas_benchmarks.VegasBenchmarker

class VegasGridBenchmarker(n_repeat=1, stratified=False, benchmark_time=False)[source]

Bases: utils.benchmark.benchmarker.GridBenchmarker, utils.benchmark.vegas_benchmarks.VegasBenchmarker

Benchmark against VEGAS by sampling parameters on a grid

Parameters

n (int) – Optional: How often the grid is sampled.

class VegasRandomHPBenchmarker(n=5, n_repeat=1, stratified=False, benchmark_time=False)[source]

Bases: utils.benchmark.benchmarker.RandomHyperparameterBenchmarker, utils.benchmark.vegas_benchmarks.VegasBenchmarker

Benchmark against VEGAS by sampling integrator hyperparameters randomly

Parameters

  • n_samples (int) – Number of random integrator configurations to draw

  • n_repeat (int) – Optional: How often the grid is sampled.

class VegasSequentialBenchmarker(n_repeat=1, stratified=False, benchmark_time=False)[source]

Bases: utils.benchmark.benchmarker.SequentialBenchmarker, utils.benchmark.vegas_benchmarks.VegasBenchmarker

Benchmark against VEGAS by testing on a sequence of (dimension, integrand, integrator) triplets

Parameters

n (int) – Optional: How often the grid is sampled.

class VegasSequentialIntegratorBenchmarker(n_repeat=1, stratified=False, benchmark_time=False)[source]

Bases: utils.benchmark.benchmarker.SequentialIntegratorBenchmarker, utils.benchmark.vegas_benchmarks.VegasBenchmarker

Benchmark against VEGAS by testing on a sequence of integrator configurations

Parameters

n (int) – Optional: How often the grid is sampled.