Skip to content

Python benchmark artifact

Source JSON: Notebooks/Python/benchmark_literature_comparison_results.json

Metadata

  • Dimensions: 10
  • Independent runs: 30
  • Algorithms: GIVP-full, GRASP-only
  • Functions: Sphere, Rosenbrock, Rastrigin, Ackley

Regenerate

python -m benchmarks.publishing --repo-root .. --output-dir ../docs/examples/benchmark-reports

Generated charts

Python mean objective value chart

Python mean runtime chart

Function-level summary

Function GIVP-full mean +- std GRASP-only mean +- std Runtime ratio (GIVP-full/GRASP-only)
Sphere 1.8690e-04 +- 5.7689e-05 2.1568 +- 0.482714 158.16x
Rosenbrock 0.513512 +- 0.325223 5.4412e+03 +- 2.2323e+03 157.07x
Rastrigin 0.849192 +- 0.624725 28.0927 +- 4.9224 158.22x
Ackley 0.152539 +- 0.027741 8.9242 +- 1.0037 150.01x

Detailed summary rows

Function Algorithm Mean +- std Median Mean nfev Mean time (s)
Ackley GIVP-full 0.152539 +- 0.027741 0.150154 4.7784e+05 30.0020
Ackley GRASP-only 8.9242 +- 1.0037 9.2427 4.8520e+03 0.2
Rastrigin GIVP-full 0.849192 +- 0.624725 1.0144 5.2482e+05 29.7450
Rastrigin GRASP-only 28.0927 +- 4.9224 29.2184 4.8980e+03 0.188
Rosenbrock GIVP-full 0.513512 +- 0.325223 0.509999 5.7176e+05 30.0010
Rosenbrock GRASP-only 5.4412e+03 +- 2.2323e+03 5.3735e+03 4.8340e+03 0.191
Sphere GIVP-full 1.8690e-04 +- 5.7689e-05 1.9862e-04 6.0563e+05 29.2600
Sphere GRASP-only 2.1568 +- 0.482714 2.1947 4.8280e+03 0.185