OptimizerIrace class that implements iterated racing. Calls
irace::irace() from package irace.
Source
Lopez-Ibanez M, Dubois-Lacoste J, Caceres LP, Birattari M, Stuetzle T (2016). “The irace package: Iterated racing for automatic algorithm configuration.” Operations Research Perspectives, 3, 43--58. doi: 10.1016/j.orp.2016.09.002 .
Parameters
instanceslist()
A list of instances where the configurations executed on.targetRunnerParallelfunction()
A function that executes the objective function with a specific parameter configuration and instance. A default function is provided, see section "Target Runner and Instances".
For the meaning of all other parameters, see irace::defaultScenario(). Note
that we have removed all control parameters which refer to the termination of
the algorithm. Use TerminatorEvals instead. Other terminators do not work
with OptimizerIrace.
Target Runner and Instances
The irace package uses a targetRunner script or R function to evaluate a
configuration on a particular instance. Usually it is not necessary to
specify a targetRunner function when using OptimizerIrace. A default
function is used that forwards several configurations and instances to the
user defined objective function. As usually, the user defined function has
a xs, xss or xdt parameter depending on the used Objective class.
For irace, the function needs an additional instances parameter.
fun = function(xs, instances) {
# function to evaluate configuration in `xs` on instance `instances`
}Archive
The Archive holds the following additional columns:
"race"(integer(1))
Race iteration."step"(integer(1))
Step number of race."instance"(integer(1))
Identifies instances across races and steps."configuration"(integer(1))
Identifies configurations across races and steps.
Result
The optimization result (instance$result) is the best performing elite of
the final race. The reported performance is the average performance estimated
on all used instances.
Dictionary
This Optimizer can be instantiated via the dictionary
mlr_optimizers or with the associated sugar function opt():
mlr_optimizers$get("irace")
opt("irace")Progress Bars
$optimize() supports progress bars via the package progressr
combined with a Terminator. Simply wrap the function in
progressr::with_progress() to enable them. We recommend to use package
progress as backend; enable with progressr::handlers("progress").
Super class
bbotk::Optimizer -> OptimizerIrace
Examples
library(data.table)
search_space = domain = ps(
x1 = p_dbl(-5, 10),
x2 = p_dbl(0, 15)
)
codomain = ps(y = p_dbl(tags = "minimize"))
# branin function with noise
# the noise generates different instances of the branin function
# the noise values are passed via the `instances` parameter
fun = function(xdt, instances) {
a = 1
b = 5.1 / (4 * (pi^2))
c = 5 / pi
r = 6
s = 10
t = 1 / (8 * pi)
data.table(y = (
a * ((xdt[["x2"]] -
b * (xdt[["x1"]]^2L) +
c * xdt[["x1"]] - r)^2) +
((s * (1 - t)) * cos(xdt[["x1"]])) +
unlist(instances)))
}
objective = ObjectiveRFunDt$new(fun = fun, domain = domain, codomain = codomain)
instance = OptimInstanceSingleCrit$new(
objective = objective,
search_space = search_space,
terminator = trm("evals", n_evals = 1000))
# create instances of branin function
instances = rnorm(10, mean = 0, sd = 0.1)
# load optimizer irace and set branin instances
optimizer = opt("irace", instances = instances)
# modifies the instance by reference
optimizer$optimize(instance)
#> # 2022-01-20 04:11:47 UTC: Initialization
#> # Elitist race
#> # Elitist new instances: 1
#> # Elitist limit: 2
#> # nbIterations: 3
#> # minNbSurvival: 3
#> # nbParameters: 2
#> # seed: 1960591573
#> # confidence level: 0.95
#> # budget: 1000
#> # mu: 5
#> # deterministic: FALSE
#>
#> # 2022-01-20 04:11:47 UTC: Iteration 1 of 3
#> # experimentsUsedSoFar: 0
#> # remainingBudget: 1000
#> # currentBudget: 333
#> # nbConfigurations: 55
#> # Markers:
#> x No test is performed.
#> c Configurations are discarded only due to capping.
#> - The test is performed and some configurations are discarded.
#> = The test is performed but no configuration is discarded.
#> ! The test is performed and configurations could be discarded but elite configurations are preserved.
#> . All alive configurations are elite and nothing is discarded
#>
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> | | Instance| Alive| Best| Mean best| Exp so far| W time| rho|KenW| Qvar|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> |x| 1| 55| 53| -8.056789966| 55|00:00:00| NA| NA| NA|
#> |x| 2| 55| 53| -8.136412105| 110|00:00:00|+1.00|1.00|0.0000|
#> |x| 3| 55| 53| -8.210824082| 165|00:00:00|+1.00|1.00|0.0000|
#> |x| 4| 55| 53| -8.261989296| 220|00:00:00|+1.00|1.00|0.0000|
#> |-| 5| 1| 53| -8.267936286| 275|00:00:00| NA| NA| NA|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> Best-so-far configuration: 53 mean value: -8.267936286
#> Description of the best-so-far configuration:
#> .ID. x1 x2 .PARENT.
#> 53 53 -3.6743 13.5351 NA
#>
#> # 2022-01-20 04:11:47 UTC: Elite configurations (first number is the configuration ID; listed from best to worst according to the sum of ranks):
#> x1 x2
#> 53 -3.6743 13.5351
#> # 2022-01-20 04:11:47 UTC: Iteration 2 of 3
#> # experimentsUsedSoFar: 275
#> # remainingBudget: 725
#> # currentBudget: 362
#> # nbConfigurations: 52
#> # Markers:
#> x No test is performed.
#> c Configurations are discarded only due to capping.
#> - The test is performed and some configurations are discarded.
#> = The test is performed but no configuration is discarded.
#> ! The test is performed and configurations could be discarded but elite configurations are preserved.
#> . All alive configurations are elite and nothing is discarded
#>
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> | | Instance| Alive| Best| Mean best| Exp so far| W time| rho|KenW| Qvar|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> |x| 6| 52| 103| -9.325996915| 52|00:00:00| NA| NA| NA|
#> |x| 1| 52| 103| -9.193478123| 103|00:00:00|+1.00|1.00|0.0000|
#> |x| 4| 52| 103| -9.268870182| 154|00:00:00|+1.00|1.00|0.0000|
#> |x| 5| 52| 103| -9.275626039| 205|00:00:00|+1.00|1.00|0.0000|
#> |-| 3| 2| 103| -9.293264311| 256|00:00:00|+1.00|1.00|0.0000|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> Best-so-far configuration: 53 mean value: -8.276918163
#> Description of the best-so-far configuration:
#> .ID. x1 x2 .PARENT.
#> 53 53 -3.6743 13.5351 NA
#>
#> # 2022-01-20 04:11:48 UTC: Elite configurations (first number is the configuration ID; listed from best to worst according to the sum of ranks):
#> x1 x2
#> 53 -3.6743 13.5351
#> 103 -3.4009 12.8144
#> # 2022-01-20 04:11:48 UTC: Iteration 3 of 3
#> # experimentsUsedSoFar: 531
#> # remainingBudget: 469
#> # currentBudget: 469
#> # nbConfigurations: 60
#> # Markers:
#> x No test is performed.
#> c Configurations are discarded only due to capping.
#> - The test is performed and some configurations are discarded.
#> = The test is performed but no configuration is discarded.
#> ! The test is performed and configurations could be discarded but elite configurations are preserved.
#> . All alive configurations are elite and nothing is discarded
#>
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> | | Instance| Alive| Best| Mean best| Exp so far| W time| rho|KenW| Qvar|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> |x| 7| 60| 126| -9.452692610| 60|00:00:00| NA| NA| NA|
#> |x| 5| 60| 126| -9.493920062| 118|00:00:00|+1.00|1.00|0.0000|
#> |x| 4| 60| 126| -9.548916110| 176|00:00:00|+1.00|1.00|0.0000|
#> |x| 3| 60| 126| -9.562454910| 234|00:00:00|+1.00|1.00|0.0000|
#> |-| 1| 3| 126| -9.510006575| 292|00:00:00|+1.00|1.00|0.0000|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> Best-so-far configuration: 53 mean value: -8.267254045
#> Description of the best-so-far configuration:
#> .ID. x1 x2 .PARENT.
#> 53 53 -3.6743 13.5351 NA
#>
#> # 2022-01-20 04:11:48 UTC: Elite configurations (first number is the configuration ID; listed from best to worst according to the sum of ranks):
#> x1 x2
#> 53 -3.6743 13.5351
#> 103 -3.4009 12.8144
#> 126 -3.2256 12.7154
#> # 2022-01-20 04:11:48 UTC: Iteration 4 of 4
#> # experimentsUsedSoFar: 823
#> # remainingBudget: 177
#> # currentBudget: 177
#> # nbConfigurations: 22
#> # Markers:
#> x No test is performed.
#> c Configurations are discarded only due to capping.
#> - The test is performed and some configurations are discarded.
#> = The test is performed but no configuration is discarded.
#> ! The test is performed and configurations could be discarded but elite configurations are preserved.
#> . All alive configurations are elite and nothing is discarded
#>
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> | | Instance| Alive| Best| Mean best| Exp so far| W time| rho|KenW| Qvar|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> |x| 8| 22| 183| -9.342026312| 22|00:00:00| NA| NA| NA|
#> |x| 1| 22| 183| -9.334022665| 41|00:00:00|+1.00|1.00|0.0000|
#> |x| 4| 22| 183| -9.450919773| 60|00:00:00|+1.00|1.00|0.0000|
#> |x| 3| 22| 183| -9.495409102| 79|00:00:00|+1.00|1.00|0.0000|
#> |-| 2| 4| 183| -9.493379941| 100|00:00:00|+1.00|1.00|0.0000|
#> |!| 7| 4| 183| -9.490899682| 101|00:00:00|+1.00|1.00|0.0000|
#> |-| 5| 3| 183| -9.500907341| 102|00:00:00|+1.00|1.00|0.0000|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> Best-so-far configuration: 103 mean value: -9.247116311
#> Description of the best-so-far configuration:
#> .ID. x1 x2 .PARENT.
#> 103 103 -3.4009 12.8144 53
#>
#> # 2022-01-20 04:11:48 UTC: Elite configurations (first number is the configuration ID; listed from best to worst according to the sum of ranks):
#> x1 x2
#> 103 -3.4009 12.8144
#> 53 -3.6743 13.5351
#> 183 -3.2215 12.6519
#> # 2022-01-20 04:11:48 UTC: Iteration 5 of 5
#> # experimentsUsedSoFar: 925
#> # remainingBudget: 75
#> # currentBudget: 75
#> # nbConfigurations: 9
#> # Markers:
#> x No test is performed.
#> c Configurations are discarded only due to capping.
#> - The test is performed and some configurations are discarded.
#> = The test is performed but no configuration is discarded.
#> ! The test is performed and configurations could be discarded but elite configurations are preserved.
#> . All alive configurations are elite and nothing is discarded
#>
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> | | Instance| Alive| Best| Mean best| Exp so far| W time| rho|KenW| Qvar|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> |x| 9| 9| 187| -9.449906541| 9|00:00:00| NA| NA| NA|
#> |x| 4| 9| 187| -9.568279210| 15|00:00:00|+1.00|1.00|0.0000|
#> |x| 8| 9| 187| -9.493507541| 21|00:00:00|+1.00|1.00|0.0000|
#> |x| 5| 9| 187| -9.510853453| 27|00:00:00|+1.00|1.00|0.0000|
#> |-| 1| 4| 187| -9.474274144| 33|00:00:00|+1.00|1.00|0.0000|
#> |!| 6| 4| 187| -9.494060869| 35|00:00:00|+1.00|1.00|0.0000|
#> |!| 3| 4| 187| -9.513597171| 36|00:00:00|+1.00|1.00|0.0000|
#> |!| 7| 4| 187| -9.509452060| 37|00:00:00|+1.00|1.00|0.0000|
#> |-| 2| 1| 187| -9.506979741| 38|00:00:00| NA| NA| NA|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> Best-so-far configuration: 187 mean value: -9.506979741
#> Description of the best-so-far configuration:
#> .ID. x1 x2 .PARENT.
#> 187 187 -3.2277 12.6472 183
#>
#> # 2022-01-20 04:11:49 UTC: Elite configurations (first number is the configuration ID; listed from best to worst according to the sum of ranks):
#> x1 x2
#> 187 -3.2277 12.6472
#> # 2022-01-20 04:11:49 UTC: Iteration 6 of 6
#> # experimentsUsedSoFar: 963
#> # remainingBudget: 37
#> # currentBudget: 37
#> # nbConfigurations: 4
#> # Markers:
#> x No test is performed.
#> c Configurations are discarded only due to capping.
#> - The test is performed and some configurations are discarded.
#> = The test is performed but no configuration is discarded.
#> ! The test is performed and configurations could be discarded but elite configurations are preserved.
#> . All alive configurations are elite and nothing is discarded
#>
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> | | Instance| Alive| Best| Mean best| Exp so far| W time| rho|KenW| Qvar|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> |x| 10| 4| 190| -9.446019379| 4|00:00:00| NA| NA| NA|
#> |x| 6| 4| 190| -9.521448742| 7|00:00:00|+1.00|1.00|0.0000|
#> |x| 1| 4| 190| -9.458246002| 10|00:00:00|+1.00|1.00|0.0000|
#> |x| 5| 4| 190| -9.485378201| 13|00:00:00|+1.00|1.00|0.0000|
#> |-| 4| 2| 190| -9.526409659| 16|00:00:00|+1.00|1.00|0.0000|
#> +-+-----------+-----------+-----------+---------------+-----------+--------+-----+----+------+
#> Best-so-far configuration: 187 mean value: -9.500495343
#> Description of the best-so-far configuration:
#> .ID. x1 x2 .PARENT.
#> 187 187 -3.2277 12.6472 183
#>
#> # 2022-01-20 04:11:49 UTC: Elite configurations (first number is the configuration ID; listed from best to worst according to the sum of ranks):
#> x1 x2
#> 187 -3.2277 12.6472
#> 190 -3.2284 12.6347
#> # 2022-01-20 04:11:49 UTC: Stopped because there is not enough budget left to race more than the minimum (3)
#> # You may either increase the budget or set 'minNbSurvival' to a lower value
#> # Iteration: 7
#> # nbIterations: 7
#> # experimentsUsedSoFar: 979
#> # timeUsed: 0
#> # remainingBudget: 21
#> # currentBudget: 21
#> # number of elites: 2
#> # nbConfigurations: 3
#> x1 x2 configuration x_domain y
#> 1: -3.2277 12.6472 187 <list[2]> -9.500495
# best scoring configuration
instance$result
#> x1 x2 configuration x_domain y
#> 1: -3.2277 12.6472 187 <list[2]> -9.500495
# all evaluations
as.data.table(instance$archive)
#> x1 x2 y race step instance configuration
#> 1: 2.2738 3.0264 -5.995785 1 1 2 1
#> 2: 5.1642 13.2338 148.587791 1 1 2 2
#> 3: 2.1558 6.0020 2.933354 1 1 2 3
#> 4: 4.1562 13.3102 131.879233 1 1 2 4
#> 5: 2.6207 7.1171 11.250214 1 1 2 5
#> ---
#> 975: -3.2244 12.6557 -9.559869 6 1 4 191
#> 976: -3.2184 12.6582 -9.558008 6 1 4 192
#> 977: -3.2284 12.6347 -9.690535 6 1 7 190
#> 978: -3.2244 12.6557 -9.683630 6 1 7 191
#> 979: -3.2184 12.6582 -9.681769 6 1 7 192
#> timestamp batch_nr
#> 1: 2022-01-20 04:11:47 1
#> 2: 2022-01-20 04:11:47 1
#> 3: 2022-01-20 04:11:47 1
#> 4: 2022-01-20 04:11:47 1
#> 5: 2022-01-20 04:11:47 1
#> ---
#> 975: 2022-01-20 04:11:49 35
#> 976: 2022-01-20 04:11:49 35
#> 977: 2022-01-20 04:11:49 36
#> 978: 2022-01-20 04:11:49 36
#> 979: 2022-01-20 04:11:49 36