SBArchOpt: Overview of Test Problems
An overview of available test problems in sb_arch_opt.problems (* = recommended; further nomenclature is explained below the table):
| Problem Class | n_discr | n_cont | n_obj | n_con | D | MD | MO | HIER | HC | n_valid_discr | imp_ratio | dependencies | Notes |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Module: continuous |
|||||||||||||
Himmelblau |
2 | 1 | |||||||||||
Goldstein |
2 | 1 | |||||||||||
Branin* |
2 | 1 | |||||||||||
Rosenbrock* |
10 | 1 | |||||||||||
Griewank |
10 | 1 | |||||||||||
Module: discrete |
|||||||||||||
MDBranin* |
2 | 2 | 1 | Y | 4 | ||||||||
AugmentedMDBranin* |
2 | 10 | 1 | Y | 4 | ||||||||
MDGoldstein |
2 | 2 | 1 | Y | 9 | ||||||||
MunozZunigaToy* |
1 | 1 | 1 | Y | 10 | ||||||||
Halstrup04 |
3 | 5 | 1 | Y | 12 | ||||||||
Module: md_mo |
|||||||||||||
MOHimmelblau |
2 | 2 | Y | ||||||||||
MDMOHimmelblau |
1 | 1 | 2 | Y | Y | 10 | |||||||
DMOHimmelblau |
2 | 2 | Y | Y | 100 | ||||||||
MOGoldstein |
2 | 2 | Y | ||||||||||
MDMOGoldstein* |
1 | 1 | 2 | Y | Y | 10 | |||||||
DMOGoldstein* |
2 | 2 | Y | Y | 100 | ||||||||
MORosenbrock* |
10 | 2 | Y | ||||||||||
MDMORosenbrock* |
5 | 5 | 2 | Y | Y | 1025 | |||||||
MOZDT1 |
30 | 2 | Y | ||||||||||
MDZDT1Small |
6 | 6 | 2 | Y | Y | 729 | |||||||
MDZDT1Mid |
10 | 10 | 2 | Y | Y | 59049 | |||||||
MDZDT1 |
15 | 15 | 2 | Y | Y | ~30.5e9 | |||||||
DZDT1 |
30 | 2 | Y | Y | ~931e18 | ||||||||
Module: constrained |
|||||||||||||
ConBraninProd |
2 | 1 | 1 | ||||||||||
ConBraninGomez |
2 | 1 | 1 | ||||||||||
ArchCantileveredBeam* |
4 | 1 | 2 | ||||||||||
MDCantileveredBeam* |
2 | 2 | 1 | 2 | Y | 100 | |||||||
ArchWeldedBeam* |
4 | 2 | 4 | Y | |||||||||
MDWeldedBeam* |
2 | 2 | 2 | 4 | Y | Y | 100 | ||||||
ArchCarside* |
7 | 3 | 10 | Y | |||||||||
MDCarside* |
4 | 3 | 3 | 10 | Y | Y | 100000 | ||||||
ArchOSY |
6 | 2 | 6 | Y | |||||||||
MDOSY |
3 | 3 | 2 | 6 | Y | Y | 1000 | ||||||
MODASCMOP |
30 | 3 | 7 | Y | |||||||||
MDDASCMOP |
15 | 15 | 3 | 7 | Y | Y | ~14.3e6 | ||||||
Module: hierarchical |
|||||||||||||
ZaeffererHierarchical |
2 | 1 | Y | ||||||||||
Jenatton |
3 | 6 | 1 | Y | Y | 4 | 2 | ||||||
NeuralNetwork |
6 | 2 | 1 | Y | Y | 4644 | 2.51 | ||||||
HierarchicalGoldstein* |
6 | 5 | 1 | 1 | Y | Y | 288 | 2.25 | |||||
MOHierarchicalGoldstein* |
6 | 5 | 2 | 1 | Y | Y | Y | 288 | 2.25 | ||||
HierBranin |
8 | 2 | 1 | Y | Y | 200 | 3.2 | ||||||
HierZDT1Small |
3 | 3 | 2 | Y | Y | Y | 10 | 1.8 | |||||
HierZDT1 |
8 | 4 | 2 | Y | Y | Y | 200 | 4.9 | |||||
HierZDT1Large |
9 | 9 | 2 | Y | Y | Y | 2000 | 8.2 | |||||
HierDiscreteZDT1 |
8 | 2 | Y | Y | Y | 2000 | 4.1 | ||||||
HierCantileveredBeam |
6 | 4 | 1 | 2 | Y | Y | 20 | 5.4 | |||||
HierCarside |
6 | 6 | 3 | 10 | Y | Y | Y | 50 | 6.5 | ||||
HierarchicalRosenbrock* |
5 | 8 | 1 | 2 | Y | Y | 32 | 1.5 | |||||
MOHierarchicalRosenbrock* |
5 | 8 | 2 | 2 | Y | Y | Y | 32 | 1.5 | ||||
~~MOHierarchicalTestProblem~~ |
11 | 16 | 2 | 2 | Y | Y | Y | 64 | 72 | ||||
Module: gnc |
|||||||||||||
GNCNoActNrType |
9 | 1 | Y | Y | 265 | 1.9 | |||||||
GNCNoActType |
11 | 2 | Y | Y | Y | 327 | 14.1 | ||||||
GNCNoActNr |
15 | 2 | Y | Y | Y | 26500 | 14.1 | ||||||
GNCNoAct* |
17 | 2 | Y | Y | Y | 29857 | 113 | ||||||
GNCNoNrType |
18 | 1 | Y | Y | 70225 | 3.7 | |||||||
GNCNoType |
21 | 2 | Y | Y | Y | 85779 | 83 | ||||||
GNCNoNr |
27 | 2 | Y | Y | Y | 70225000 | 74 | ||||||
GNC* |
30 | 2 | Y | Y | Y | 79091323 | 1761 | ||||||
MDGNCNoActNr |
9 | 6 | 2 | Y | Y | Y | 265 | 23.2 | |||||
MDGNCNoAct* |
11 | 6 | 2 | Y | Y | Y | 327 | 150 | |||||
SOMDGNCNoAct |
11 | 6 | 1 | Y | Y | 327 | 150 | ||||||
MDGNCNoNr |
18 | 9 | 2 | Y | Y | Y | 70225 | 44.8 | |||||
MDGNC* |
21 | 9 | 2 | Y | Y | Y | 85779 | 925 | |||||
Module: rocket |
rocket |
||||||||||||
RocketArch |
8 | 6 | 2 | 2 | Y | Y | Y | 18522 | 3.7 | rocket |
|||
LCRocketArch* |
8 | 6 | 2 | 2 | Y | Y | Y | 18522 | 3.7 | rocket |
|||
SOLCRocketArch* |
8 | 6 | 1 | 2 | Y | Y | Y | 18522 | 3.7 | rocket |
|||
Module: hidden_constraints |
|||||||||||||
Mueller01 |
5 | 1 | Y | fail_rate: 67% | |||||||||
Mueller02 |
4 | 1 | Y | fail_rate: 40% | |||||||||
MDMueller02 |
2 | 2 | 1 | Y | Y | 36 | fail_rate: 37% | ||||||
Mueller08 |
10 | 1 | Y | fail_rate: 73% | |||||||||
MOMueller08 |
10 | 2 | Y | Y | fail_rate: 73% | ||||||||
MDMueller08 |
2 | 8 | 1 | Y | Y | 100 | fail_rate: 78% | ||||||
MDMOMueller08 |
2 | 8 | 2 | Y | Y | Y | 100 | fail_rate: 78% | |||||
Alimo* |
2 | 1 | Y | fail_rate: 51% | |||||||||
AlimoEdge* |
2 | 1 | Y | fail_rate: 53% | |||||||||
HCBranin* |
2 | 1 | Y | fail_rate: 33% | |||||||||
HCSphere |
2 | 1 | Y | fail_rate: 51% | |||||||||
Tfaily01 |
2 | 1 | Y | fail_rate: 39% | |||||||||
Tfaily02 |
2 | 1 | Y | fail_rate: 80% | |||||||||
Tfaily03 |
4 | 1 | Y | fail_rate: 67% | |||||||||
Tfaily04 |
6 | 1 | Y | fail_rate: 35% | |||||||||
CantileveredBeamHC* |
4 | 1 | 1 | Y | fail_rate: 83% | ||||||||
MDCantileveredBeamHC* |
2 | 2 | 1 | 1 | Y | Y | 100 | fail_rate: 81% | |||||
CarsideHC |
7 | 3 | 8 | Y | Y | fail_rate: 66% | |||||||
CarsideHCLess |
7 | 3 | 9 | Y | Y | fail_rate: 39% | |||||||
MDCarsideHC |
4 | 3 | 3 | 8 | Y | Y | Y | 10000 | fail_rate: 66% | ||||
HierAlimo* |
5 | 2 | 1 | Y | Y | Y | 20 | 5.4 | fail_rate: 51% | ||||
HierAlimoEdge |
5 | 2 | 1 | Y | Y | Y | 20 | 5.4 | fail_rate: 53% | ||||
HierMueller02* |
4 | 4 | 1 | Y | Y | Y | 20 | 5.4 | fail_rate: 37% | ||||
HierMueller08* |
4 | 10 | 1 | Y | Y | Y | 20 | 5.4 | fail_rate: 73% | ||||
MOHierMueller08 |
4 | 10 | 2 | Y | Y | Y | Y | 20 | 5.4 | fail_rate: 73% | |||
HierarchicalRosenbrockHC |
5 | 8 | 1 | 1 | Y | Y | Y | Y | 32 | 1.5 | fail_rate: 21% | ||
MOHierarchicalRosenbrockHC |
5 | 8 | 2 | 1 | Y | Y | Y | Y | 32 | 1.5 | fail_rate: 60% | ||
~~HCMOHierarchicalTestProblem~~ |
11 | 16 | 2 | 1 | Y | Y | Y | Y | 64 | 72 | fail_rate: 60% | ||
Module: turbofan_arch |
ota |
||||||||||||
SimpleTurbofanArch |
6 | 9 | 1 | 5 | Y | Y | Y | 70 | 3.1 | ota |
t_eval: 1-5 min; fail_rate: 51% | ||
SimpleTurbofanArchModel* |
6 | 9 | 1 | 5 | Y | Y | Y | 70 | 3.1 | ota |
fail_rate: 51% | ||
RealisticTurbofanArch |
11 | 30 | 3 | 15 | Y | Y | Y | Y | 142243 | 9.1 | ota |
t_eval: 1-5 min; fail_rate: 67% | |
Module: assignment |
assignment |
dist_corr: 100% if not specified | |||||||||||
Assignment |
12 | 2 | Y | Y | 4096 | assignment |
|||||||
AssignmentLarge |
16 | 2 | Y | Y | 65536 | assignment |
|||||||
AssignmentInj |
5 | 2 | Y | Y | 7776 | assignment |
dist_corr: 31% | ||||||
AssignmentInjLarge |
18 | 2 | Y | Y | 117649 | assignment |
dist_corr: 31% | ||||||
AssignmentRepeat |
8 | 2 | Y | Y | 6561 | assignment |
|||||||
AssignmentRepeatLarge |
10 | 2 | Y | Y | 59049 | assignment |
|||||||
Partitioning |
12 | 2 | Y | Y | 4096 | assignment |
dist_corr: 74% | ||||||
PartitioningLarge |
7 | 2 | Y | Y | 78125 | assignment |
dist_corr: 67% | ||||||
PartitioningCovering |
12 | 2 | Y | Y | Y | 2401 | 1.71 | assignment |
|||||
PartitioningCoveringLarge |
16 | 2 | Y | Y | Y | 50625 | 1.29 | assignment |
|||||
Downselecting |
12 | 2 | Y | Y | 4096 | assignment |
|||||||
DownselectingLarge |
15 | 2 | Y | Y | 32768 | assignment |
|||||||
Connecting |
12 | 2 | Y | Y | 4096 | assignment |
|||||||
ConnectingUndirected |
15 | 2 | Y | Y | 32768 | assignment |
|||||||
Permuting |
6 | 2 | Y | Y | 5040 | assignment |
dist_corr: 43% | ||||||
PermutingLarge |
7 | 2 | Y | Y | 40320 | assignment |
dist_corr: 39% | ||||||
UnordNonReplComb |
14 | 2 | Y | Y | Y | 6435 | 2.55 | assignment |
|||||
UnordNonReplCombLarge |
18 | 2 | Y | Y | Y | 92378 | 2.84 | assignment |
|||||
UnorderedComb |
1 | 2 | Y | Y | 2002 | assignment |
dist_corr: 34% | ||||||
AssignmentGNCNoActType |
11 | 2 | Y | Y | Y | 327 | 14.1 | assignment |
|||||
AssignmentGNCNoAct |
15 | 2 | Y | Y | Y | 29857 | 39.5 | assignment |
|||||
AssignmentGNCNoType |
21 | 2 | Y | Y | Y | 85779 | 82.5 | assignment |
|||||
AssignmentGNC |
27 | 2 | Y | Y | Y | 79091323 | 367 | assignment |
Nomenclature:
- n_discr: number of discrete (integer or categorical) design variables
- n_cont: number of continuous design variables
- n_obj: number of objectives
- n_con: number of (inequality) constraints
- D: whether the problem contains discrete design variables
- MD: whether the problem is a mixed-discrete problem or not
- MO: whether the problem is a multi-objective problem or not
- HIER: whether the problem contains hierarchical variables
- HC: whether the problem contains hidden constraints (i.e. some points might fail to evaluate); see also failure_rate
- n_valid_discr: number of valid discrete design points (i.e. ignoring continuous dimensions)
- imp_ratio: imputation ratio; ratio between the number of declared and valid discrete design points (1 means there are no invalid design vectors)
- dependencies: name of the optional dependencies list to install to use the test problem
- For assignment and ota, use
pip install -r requirements-ota.txt(same for assignment)! - Otherwise:
pip install sb-arch-opt[name] - fail_rate: fraction of randomly-sampled points that fail to evaluate
- t_eval: rough estimate of the time it takes to evaluate one design point (practically instantaneous if left empty)
- dist_corr: distance correlation between design vectors and assignment patterns (higher is better)