Skip to content

quality-attributes/autoML-models

Repository files navigation

Iterated Racing Configuration for Differential Evolutionary Algorithms

Hyperparameter tuning for software requirements categorization using Differential Evolution, the configuration results are:

DecisionTreeClassifier

Instance Alive Best Mean best Exp so far W time rho KenW Qvar
x 1 8 2 -58.12429379 8 00:07:39 NA NA NA
x 2 8 2 -58.28813559 16 00:07:07 +0.81 0.90 0.2449
x 3 8 2 -58.22598870 24 00:06:59 +0.76 0.84 0.1995
x 4 8 2 -58.02824859 32 00:07:03 +0.80 0.85 0.1602
- 5 2 2 -58.04180791 40 00:07:01 +0.20 0.36 0.4000

Best-so-far configuration: 2 mean value: -58.04180791

Description of the best-so-far configuration:

.ID. population generations crossover factor .PARENT.
2 2 28 97 0.6056 1.0345 NA

Best configurations (first number is the configuration ID; listed from best to worst according to the sum of ranks):

population generations crossover factor
2 28 97 0.6056 1.0345
7 34 78 0.1657 0.7599

MultinomialNB

Instance Alive Best Mean best Exp so far W time rho KenW Qvar
x 1 8 1 -63.18644068 8 00:00:51 NA NA NA
x 2 8 1 -63.35593220 16 00:00:49 +0.39 0.70 0.6552
x 3 8 1 -63.30131827 24 00:00:52 +0.08 0.38 0.7576
x 4 8 1 -63.27683616 32 00:00:49 +0.14 0.36 0.7402
= 5 8 1 -62.92203390 40 00:00:43 +0.04 0.23 0.8468
= 6 8 1 -62.74293785 48 00:00:49 +0.05 0.21 0.8784

Best-so-far configuration: 1 mean value: -62.74293785

Description of the best-so-far configuration:

.ID. population generations crossover factor .PARENT.
1 1 50 48 0.0778 1.1948 NA

Best configurations (first number is the configuration ID; listed from best to worst according to the sum of ranks):

population generations crossover factor
1 50 48 0.0778 1.1948
7 32 88 0.0998 1.6417
3 18 18 0.1956 1.5302
2 12 11 0.3356 1.6721

SVC

Instance Alive Best Mean best Exp so far W time rho KenW Qvar
x 1 5 3 -67.23163842 5 00:45:25 NA NA NA
x 2 5 5 -66.38983051 10 00:48:50 +0.41 0.71 0.2080
x 3 5 3 -66.78154426 15 00:48:01 +0.34 0.56 0.2418
x 4 5 3 -66.64124294 20 00:48:52 +0.28 0.46 0.2893
= 5 5 3 -66.55593220 25 00:46:40 +0.28 0.42 0.2687
- 6 4 4 -66.16666667 30 00:44:43 -0.19 0.01 0.8539

Best-so-far configuration: 2 mean value: -66.05178908

Description of the best-so-far configuration:

.ID. population generations crossover factor .PARENT.
2 2 6 75 0.5504 0.5324 NA

Best configurations (first number is the configuration ID; listed from best to worst according to the sum of ranks):

population generations crossover factor
2 6 75 0.5504 0.5324
4 4 81 0.2506 1.2559
3 10 41 0.6035 1.3334
5 26 18 0.9067 0.3387

RandomForestClassifier

Instance Alive Best Mean best Exp so far W time rho KenW Qvar
x 1 5 3 -67.23163842 5 00:45:25 NA NA NA
x 2 5 5 -66.38983051 10 00:48:50 +0.41 0.71 0.2080
x 3 5 3 -66.78154426 15 00:48:01 +0.34 0.56 0.2418
x 4 5 3 -66.64124294 20 00:48:52 +0.28 0.46 0.2893
= 5 5 3 -66.55593220 25 00:46:40 +0.28 0.42 0.2687
- 6 4 4 -66.16666667 30 00:44:43 -0.19 0.01 0.8539

Best-so-far configuration: 2 mean value: -66.05178908

Description of the best-so-far configuration:

.ID. population generations crossover factor .PARENT.
2 2 6 75 0.5504 0.5324 NA

Best configurations (first number is the configuration ID; listed from best to worst according to the sum of ranks):

population generations crossover factor
2 6 75 0.5504 0.5324
4 4 81 0.2506 1.2559
3 10 41 0.6035 1.3334
5 26 18 0.9067 0.3387