Scenario

Click to show 
## irace scenario:
scenarioFile = "./scenario.txt"
execDir = "/home/taccgoo"
parameterFile = "/home/taccgoo/parameters.txt"
forbiddenExps = NULL = expression()
forbiddenFile = ""
initConfigurations = structure(list(value_f = "SaDE", value_cr = "SaDE", value_ms = "SaDE",     NP_start = "30", NP_reduce = "0", param_archive = "0", param_ms_LP = "30",     param_ms_k = 0.1), class = "data.frame", row.names = c(NA, -1L))
configurationsFile = ""
logFile = "/home/taccgoo/irace.Rdata"
recoveryFile = ""
instances = structure(0:49, dim = 50L)
trainInstancesDir = "./Instances"
trainInstancesFile = ""
sampleInstances = TRUE
testInstancesDir = ""
testInstancesFile = ""
testInstances = NULL
testNbElites = 1
testIterationElites = FALSE
testType = "friedman"
firstTest = 5
eachTest = 1
targetRunner = function (...) {    .External(".Python", <pointer: (nil)>, ...)}
targetRunnerLauncher = ""
targetRunnerLauncherArgs = "{targetRunner} {targetRunnerArgs}"
targetRunnerRetries = 0
targetRunnerData = ""
targetRunnerParallel = NULL
targetEvaluator = NULL
deterministic = FALSE
maxExperiments = 500
maxTime = 0
budgetEstimation = 0.02
minMeasurableTime = 0.01
parallel = 16
loadBalancing = TRUE
mpi = FALSE
batchmode = "0"
digits = 4
quiet = FALSE
debugLevel = 3
seed = 523573206
softRestart = TRUE
softRestartThreshold = 1e-04
elitist = TRUE
elitistNewInstances = 1
elitistLimit = 2
repairConfiguration = NULL
capping = FALSE
cappingType = "median"
boundType = "candidate"
boundMax = NULL
boundDigits = 0
boundPar = 1
boundAsTimeout = TRUE
postselection = 0
aclib = FALSE
nbIterations = 0
nbExperimentsPerIteration = 0
minNbSurvival = 0
nbConfigurations = 0
mu = 5
confidence = 0.95
## end of irace scenario

Parameters

n_total n_fixed n_int n_real n_cat n_ord n_conditional n_dependent
8 0 0 1 7 0 2 0

Parameters tree

 
├─value_f
│ └─param_ms_LP
├─value_cr
│ └─param_ms_LP
├─value_ms
│ ├─param_ms_LP
│ └─param_ms_k
├─NP_start
├─NP_reduce
└─param_archive

Parameters file

Click to show 
value_f       "" c (jDE,SaDE,EPSDE,SHADE)
value_cr      "" c (jDE,SaDE,EPSDE,SHADE)
value_ms      "" c (jDE,SaDE,EPSDE,SHADE)
NP_start      "" c (30,40,50,60,70,80,90,100,120,140,160,180,200,220,240,260,280,300)
NP_reduce     "" c (0,10,20,30,40,50)
param_ms_LP   "" c (30,40,50,60,70,80,90,100,120,140,160,180,200,220,240,260,280,300) | value_ms %in% c("SaDE", "EPSDE") || value_cr %in% c("SaDE", "SHADE") || value_f %in% c("SHADE")
param_ms_k    "" r (0.001,1)       | value_ms %in% c("SaDE", "EPSDE")
param_archive "" c (0,5,10,15,20,25,30,40,50)

General information

  • irace version: 3.5.6863679
  • Iterations: 6
  • Configurations: 70
  • Instances: 13
  • Experiments: 498
  • Elite configurations: 5
  • Soft restarts: 0
  • Rejected configurations: 0
  • Running time (seconds): 225047 (CPU-user); 303 (CPU-sys); 31526 (Wall-clock)
  • Termination reason: Not enough budget to race more than the minimum configurations

By iteration

By instance

Elite configurations

The final best configurations found by irace are:

Parallel coordinates visualization (only elites)

Sampling model

The frequency of the parameter values sampled by irace:

#> TableGrob (3 x 3) "arrange": 8 grobs
#>   z     cells    name           grob
#> 1 1 (1-1,1-1) arrange gtable[layout]
#> 2 2 (1-1,2-2) arrange gtable[layout]
#> 3 3 (1-1,3-3) arrange gtable[layout]
#> 4 4 (2-2,1-1) arrange gtable[layout]
#> 5 5 (2-2,2-2) arrange gtable[layout]
#> 6 6 (2-2,3-3) arrange gtable[layout]
#> 7 7 (3-3,1-1) arrange gtable[layout]
#> 8 8 (3-3,2-2) arrange gtable[layout]

Testing performance

  • Number of elites tested: 1
  • Iteration elites tested: FALSE

Performance of the elite configurations on the test instances

#> No test instances given.

Final elite configurations on the test instances

#> No test instances given.

Iteration elite configurations on the test instances

#> Iteration elites were not tested.

Training performance

Performance of the final elite configurations on the training instances

Final elite configurations on the training instances

Races overview

Convergence plot

This is a simplified version of the visualization you can obtain with acviz.

Disabled because sections$convergence is FALSE.