Some LLAMA benchmark results

LLAMA results

All results were produced by using the cross-validation splits in the repository with 10 folds and 1 repetitions.
The best values within a type (i.e., baseline (except for vbs), classif, regr and cluster) and performance measure (i.e., Percentage solved, PAR10, MCP) are colored green. Furthermore, the three best values over all groups within a performance measure are colored pink, the absolute best one is red.

The performance is measured in three different ways.

Percentage solved records the percentage of problem instances in the data set for which the selector selected an algorithm that was able to solve it with runstatus "ok" and the algorithm time plus the feature computation time was at most the timeout.
The penalized average runtime score (PAR10) measures the time required to run on all problem instances. If an instance was solved within the timeout by the algorithm the selector chose, the actual runtime is taken. If a timeout occurred, the timeout value was multiplied by 10.
The misclassification penalty (mcp) measures the additional time required to run on all problems if sub-optimal algorithms were used. That is, if an algorithm is run on a problem instance that is not the best, a performance loss is incurred. There are no additional penalties or factors for timeouts. The virtual best solver always has a misclassification penalty of zero.

algo	model	succ	par10	mcp
baseline	vbs	1.000	227.605	0.000
baseline	singleBest	0.812	7002.907	688.774
baseline	singleBestByPar	0.812	7002.907	688.774
baseline	singleBestBySuccesses	0.812	7002.907	688.774
classif	rpart	0.856	5437.733	537.642
classif	randomForest	0.861	5228.573	512.923
classif	ksvm	0.856	5434.868	534.777
cluster	XMeans	0.852	5534.531	511.480
regr	lm	0.843	5887.326	556.875
regr	rpart	0.843	5916.120	585.669
regr	randomForest	0.846	5748.065	540.574

The following default feature steps were used for model building:

orig, lfa1, lfa2

Number of presolved instances: 0

The cost for using the feature steps (adapted for presolving) is: 0 or on average: NA

The feature steps correspond to the following 22 / 22 instance features:

stacks, tiers, stack.tier.ratio, container.density, empty.stack.pct,
overstowing.stack.pct, overstowing.2cont.stack.pct, group.same.min, group.same.max, group.same.mean,
group.same.stdev, top.good.min, top.good.max, top.good.mean, top.good.stdev,
overstowage.pct, bflb, left.density, tier.weighted.groups, avg.l1.top.left.lg.group,
cont.empty.grt.estack, pct.bottom.pct.on.top

Algorithm and Feature Subset Selection

In order to get a better insight of the scenarios, forward selections have been applied to the solvers and features to determine whether small subsets achieve comparable performances. Following this approach, we reduced the number of solvers from 4 to 4, resulting in a PAR10 score of 5225.288 for the reduced model. Analogously, the model that was generated based on 2 of the originally 22 features resulted in a PAR10 score of 4614.868. Below, you can find the list of the selected features and solvers:

Selected Features:
group.same.max, overstowage.pct

Selected Solvers:
astar.symmulgt.transmul, astar.symmullt.transmul, idastar.symmulgt.transmul, idastar.symmullt.transmul