defaultScenario: defaultScenario
In MLopez-Ibanez/iracelhs: Iterated Racing for Automatic Algorithm Configuration

Description Usage Arguments Value Author(s) See Also

'defaultScenario' returns the default scenario settings of irace.

1	defaultScenario(scenario = list())

scenario

A list where tagged elements correspond to scenario settings of irace.

A list indexed by the irace parameter names, containing the default values for each parameter, except for those already present in the scenario passed as argument. The scenario list contains the following elements:

General:

scenarioFile: Path of the file that describes the configuration scenario setup and other irace settings. (Default: "./scenario.txt")
debugLevel: Debug level of the output of irace. Set this to 0 to silence all debug messages. Higher values provide more verbose debug messages. (Default: 0)
seed: Seed of the random number generator (must be a positive integer, NA means use a random seed). (Default: NA)
execDir: Directory where the experiments will be run. (Default: "./")
logFile: Path of the file to save tuning results as an R dataset, either absolute path or relative to execDir. (Default: "./irace.Rdata")
repairConfiguration: User-defined R function that takes a configuration generated by irace and repairs it. (Default: "")

Elitist irace:

elitist: Enable/disable elitist irace. (Default: 1)
elitistNewInstances: Number of instances added to the execution list before previous instances in elitist irace. (Default: 1)
elitistLimit: Limit for the elitist race, number statistical test without elimination peformed. Use 0 for no limit. (Default: 2)

irace internal: For most of these parameters is adviced to use the default settings.

sampleInstances: Enable/disable the sampling of instances. (Default: 1)
nbIterations: Number of iterations to be performed by irace.
nbExperimentsPerIteration: Number of experiments (runs) per iteration.
nbConfigurations: Number of configurations to be sampled and evaluated at each iteration.
mu: Parameter used to define the number of configurations sampled and evaluated at each iteration. (Default: 5)
minNbSurvival: Minimum number of configurations needed to continue executing a race.
softRestart: Enable/disable the soft restart strategy that avoids premature convergence of the probabilistic model. (Default: 1)
softRestartThreshold: Soft restart threshold value for numerical parameters. If NA, it computed as 10^-digits. (Default: NA)

Target algorithm parameters:

parameterFile: Path to the file that contains the description of the parameters to be tuned. See the template. (Default: "./parameters.txt")
digits: Indicates the number of decimal places to be considered for the real parameters. (Default: 4)
forbiddenExps: MANUEL
forbiddenFile: Path to a file that contains a list of logical expressions that cannot be TRUE for any evaluated configuration. If empty or NULL, do not use forbidden expressions. (Default: "")

Target algorithm execution:

targetRunner: Path to the script called for each configuration that launches the algorithm to be tuned. See templates. (Default: "./target-runner")
targetRunnerRetries: Number of times to retry a call to targetRunner if the call failed. (Default: 0)
targetRunnerData: Optional data passed to targetRunner. This is ignored by the default targetRunner function, but it may be used by custom targetRunner functions to pass persistent data around.
targetRunnerParallel: Optional R function to provide custom parallelization of targetRunner.
targetEvaluator: Path to the script that provides a numeric value for each configuration. See templates. (Default: "")
deterministic: If the target algorithm is deterministic, configurations will be evaluated only once per instance. (Default: 0)
parallel: Number of calls to targetRunner to execute in parallel. 0 or 1 mean disabled. (Default: 0)
loadBalancing: Enable/disable load-balancing when executing experiments in parallel. Load-balancing makes better use of computing resources, but increases communication overhead. If this overhead is large, disabling load-balancing may be faster. (Default: 1)
mpi: Enable/disable MPI. Use Rmpi to execute targetRunner in parallel (parameter parallel is the number of slaves). (Default: 0)
batchmode: Specify how irace waits for jobs to finish when targetRunner submits jobs to a batch cluster: sge, pbs, torque or slurm (targetRunner must submit jobs to the cluster using. for example, qsub). (Default: 0)

Initial configurations:

configurationsFile: Path to a file that contains a set of initial configurations. If empty or NULL do not use a initial configurations. (Default: "")

Training instances:

instances: Array of the instances to be used in the targetRunner.
instancesList: Data frame that contains the instance and seeds used in the tuning. The order of this data frame indicates the order in which the instances have been used.
- instance: The instance index (corresponding to the instance array).
- seed: Seed to be used with the corresponding instance.
instances.extra.params: Array that contains the extra parameters defined per instance, the name of each element corresponds to the instance name in the instances array. When no extra parameters are provided this element is NULL.
trainInstancesDir: Directory where tuning instances are located; either absolute path or relative to current directory.. If no trainInstancesFiles is provided all the files in trainInstancesDir will be listed as instances. (Default: "./Instances")
trainInstancesFile: Path to a file that contains a list of instances and optionally additional parameters for them. If trainInstancesDir is provided irace will search for the files in this folder. (Default: "")

Tuning budget:

maxExperiments: Maximum number of runs (invocations of targetRunner) that will be performed. It determines the maximum budget of experiments for the tuning. (Default: 0)
maxTime: Maximum total execution time for the executions of targetRunner. targetRunner must return two values: quality time. (Default: 0)
budgetEstimation: Percentage of the time budget used to estimate the mean computation time of a configuration. Only used when maxTime is provided. (Default: 0.02)

Statistical test:

testType: Statistical test used for elimination. (Default: "F-test")
confidence: Confidence level of the statistical test. (Default: 0.95)
firstTest: Number of instances evaluated before the first elimination test. (Default: 5)
eachTest: Number of instances evaluated between elimination tests. (Default: 1)

Recovery:

recoveryFile: Path to an irace log file used to recover the irace execution. (Default: "")

Testing:

testNbElites: Number of elite configurations returned by irace to be tested if test instances are provided. (Default: 1)
testIterationElites: Enable/disable testing the elite configurations found at each iteration. (Default: 0)
testInstancesDir: Directory where testing instances are located, either absolute or relative to current directory. (Default: "")
testInstancesFile: Path to a file containing a list of test instances and optionally additional parameters for them. (Default: "")
testInstances: Array of the instances to be used in the targetRunner when executing the testing.
testInstances.extra.params: Array that contains the extra parameters defined per instance, the name of each element corresponds to the instance name in the testInstances array. When no extra parameters are provided this element is NULL.

Manuel López-Ibáñez and Jérémie Dubois-Lacoste

readScenario for reading a configuration scenario from a file. checkScenario to check that the scenario is valid. printScenario prints the given scenario.

MLopez-Ibanez/iracelhs documentation built on May 15, 2019, 1:57 a.m.