Applies a function on the results of your finished jobs and thereby collects
them in a list or data.table.
The later requires the provided function to return a list (or data.frame) of scalar values.
See rbindlist for features and limitations of the aggregation.
If not all jobs are terminated, the respective result will be NULL.
Usage
reduceResultsList(
ids = NULL,
fun = NULL,
...,
missing.val,
reg = getDefaultRegistry()
)
reduceResultsDataTable(
ids = NULL,
fun = NULL,
...,
missing.val,
reg = getDefaultRegistry()
)Arguments
- ids
[
data.frameorinteger]
Adata.frame(ordata.table) with a column named “job.id”. Alternatively, you may also pass a vector of integerish job ids. If not set, defaults to the return value offindDone. Invalid ids are ignored.- fun
[
function]
Function to apply to each result. The result is passed unnamed as first argument. IfNULL, the identity is used. If the function has the formal argument “job”, theJob/Experimentis also passed to the function.- ...
[
ANY]
Additional arguments passed to to functionfun.- missing.val
[
ANY]
Value to impute as result for a job which is not finished. If not provided and a result is missing, an exception is raised.- reg
[
Registry]
Registry. If not explicitly passed, uses the default registry (seesetDefaultRegistry).
Value
reduceResultsList returns a list of the results in the same order as the provided ids.
reduceResultsDataTable returns a data.table with columns “job.id” and additional result columns
created via rbindlist, sorted by “job.id”.
Note
If you have thousands of jobs, disabling the progress bar (options(batchtools.progress = FALSE))
can significantly increase the performance.
See also
Other Results:
batchMapResults(),
loadResult(),
reduceResults()
Examples
### Example 1 - reduceResultsList
tmp = makeRegistry(file.dir = NA, make.default = FALSE)
#> No readable configuration file found
#> Created registry in '/tmp/batchtools-example/reg1' using cluster functions 'Interactive'
batchMap(function(x) x^2, x = 1:10, reg = tmp)
#> Adding 10 jobs ...
submitJobs(reg = tmp)
#> Submitting 10 jobs in 10 chunks using cluster functions 'Interactive' ...
waitForJobs(reg = tmp)
#> [1] TRUE
reduceResultsList(fun = sqrt, reg = tmp)
#> [[1]]
#> [1] 1
#>
#> [[2]]
#> [1] 2
#>
#> [[3]]
#> [1] 3
#>
#> [[4]]
#> [1] 4
#>
#> [[5]]
#> [1] 5
#>
#> [[6]]
#> [1] 6
#>
#> [[7]]
#> [1] 7
#>
#> [[8]]
#> [1] 8
#>
#> [[9]]
#> [1] 9
#>
#> [[10]]
#> [1] 10
#>
### Example 2 - reduceResultsDataTable
tmp = makeExperimentRegistry(file.dir = NA, make.default = FALSE)
#> No readable configuration file found
#> Created registry in '/tmp/batchtools-example/reg2' using cluster functions 'Interactive'
# add first problem
fun = function(job, data, n, mean, sd, ...) rnorm(n, mean = mean, sd = sd)
addProblem("rnorm", fun = fun, reg = tmp)
#> Adding problem 'rnorm'
# add second problem
fun = function(job, data, n, lambda, ...) rexp(n, rate = lambda)
addProblem("rexp", fun = fun, reg = tmp)
#> Adding problem 'rexp'
# add first algorithm
fun = function(instance, method, ...) if (method == "mean") mean(instance) else median(instance)
addAlgorithm("average", fun = fun, reg = tmp)
#> Adding algorithm 'average'
# add second algorithm
fun = function(instance, ...) sd(instance)
addAlgorithm("deviation", fun = fun, reg = tmp)
#> Adding algorithm 'deviation'
# define problem and algorithm designs
library(data.table)
prob.designs = algo.designs = list()
prob.designs$rnorm = CJ(n = 100, mean = -1:1, sd = 1:5)
prob.designs$rexp = data.table(n = 100, lambda = 1:5)
algo.designs$average = data.table(method = c("mean", "median"))
algo.designs$deviation = data.table()
# add experiments and submit
addExperiments(prob.designs, algo.designs, reg = tmp)
#> Adding 30 experiments ('rnorm'[15] x 'average'[2] x repls[1]) ...
#> Adding 15 experiments ('rnorm'[15] x 'deviation'[1] x repls[1]) ...
#> Adding 10 experiments ('rexp'[5] x 'average'[2] x repls[1]) ...
#> Adding 5 experiments ('rexp'[5] x 'deviation'[1] x repls[1]) ...
submitJobs(reg = tmp)
#> Submitting 60 jobs in 60 chunks using cluster functions 'Interactive' ...
# collect results and join them with problem and algorithm paramters
res = ijoin(
getJobPars(reg = tmp),
reduceResultsDataTable(reg = tmp, fun = function(x) list(res = x))
)
unwrap(res, sep = ".")
#> Key: <job.id>
#> job.id problem algorithm prob.pars.n prob.pars.mean prob.pars.sd
#> <int> <char> <char> <num> <int> <int>
#> 1: 1 rnorm average 100 -1 1
#> 2: 2 rnorm average 100 -1 1
#> 3: 3 rnorm average 100 -1 2
#> 4: 4 rnorm average 100 -1 2
#> 5: 5 rnorm average 100 -1 3
#> 6: 6 rnorm average 100 -1 3
#> 7: 7 rnorm average 100 -1 4
#> 8: 8 rnorm average 100 -1 4
#> 9: 9 rnorm average 100 -1 5
#> 10: 10 rnorm average 100 -1 5
#> 11: 11 rnorm average 100 0 1
#> 12: 12 rnorm average 100 0 1
#> 13: 13 rnorm average 100 0 2
#> 14: 14 rnorm average 100 0 2
#> 15: 15 rnorm average 100 0 3
#> 16: 16 rnorm average 100 0 3
#> 17: 17 rnorm average 100 0 4
#> 18: 18 rnorm average 100 0 4
#> 19: 19 rnorm average 100 0 5
#> 20: 20 rnorm average 100 0 5
#> 21: 21 rnorm average 100 1 1
#> 22: 22 rnorm average 100 1 1
#> 23: 23 rnorm average 100 1 2
#> 24: 24 rnorm average 100 1 2
#> 25: 25 rnorm average 100 1 3
#> 26: 26 rnorm average 100 1 3
#> 27: 27 rnorm average 100 1 4
#> 28: 28 rnorm average 100 1 4
#> 29: 29 rnorm average 100 1 5
#> 30: 30 rnorm average 100 1 5
#> 31: 31 rnorm deviation 100 -1 1
#> 32: 32 rnorm deviation 100 -1 2
#> 33: 33 rnorm deviation 100 -1 3
#> 34: 34 rnorm deviation 100 -1 4
#> 35: 35 rnorm deviation 100 -1 5
#> 36: 36 rnorm deviation 100 0 1
#> 37: 37 rnorm deviation 100 0 2
#> 38: 38 rnorm deviation 100 0 3
#> 39: 39 rnorm deviation 100 0 4
#> 40: 40 rnorm deviation 100 0 5
#> 41: 41 rnorm deviation 100 1 1
#> 42: 42 rnorm deviation 100 1 2
#> 43: 43 rnorm deviation 100 1 3
#> 44: 44 rnorm deviation 100 1 4
#> 45: 45 rnorm deviation 100 1 5
#> 46: 46 rexp average 100 NA NA
#> 47: 47 rexp average 100 NA NA
#> 48: 48 rexp average 100 NA NA
#> 49: 49 rexp average 100 NA NA
#> 50: 50 rexp average 100 NA NA
#> 51: 51 rexp average 100 NA NA
#> 52: 52 rexp average 100 NA NA
#> 53: 53 rexp average 100 NA NA
#> 54: 54 rexp average 100 NA NA
#> 55: 55 rexp average 100 NA NA
#> 56: 56 rexp deviation 100 NA NA
#> 57: 57 rexp deviation 100 NA NA
#> 58: 58 rexp deviation 100 NA NA
#> 59: 59 rexp deviation 100 NA NA
#> 60: 60 rexp deviation 100 NA NA
#> job.id problem algorithm prob.pars.n prob.pars.mean prob.pars.sd
#> prob.pars.lambda algo.pars.method result.res
#> <int> <char> <num>
#> 1: NA mean -1.092018851
#> 2: NA median -0.863780644
#> 3: NA mean -1.084890423
#> 4: NA median -1.505171392
#> 5: NA mean -1.381319138
#> 6: NA median -1.341051423
#> 7: NA mean -1.188083630
#> 8: NA median 0.029939562
#> 9: NA mean -0.987419910
#> 10: NA median -1.673002281
#> 11: NA mean 0.007267359
#> 12: NA median 0.127401909
#> 13: NA mean -0.179617350
#> 14: NA median 0.046125736
#> 15: NA mean -0.057929853
#> 16: NA median 0.140366680
#> 17: NA mean -0.388849478
#> 18: NA median -1.207701427
#> 19: NA mean 0.042334184
#> 20: NA median -0.381653696
#> 21: NA mean 0.907293957
#> 22: NA median 1.033215601
#> 23: NA mean 1.020619322
#> 24: NA median 0.876104940
#> 25: NA mean 0.679306393
#> 26: NA median 0.977987955
#> 27: NA mean 0.675594915
#> 28: NA median 1.366615131
#> 29: NA mean 1.189950292
#> 30: NA median 0.263092706
#> 31: NA <NA> 1.040412625
#> 32: NA <NA> 2.392300899
#> 33: NA <NA> 3.042692900
#> 34: NA <NA> 4.263774294
#> 35: NA <NA> 5.296092785
#> 36: NA <NA> 1.071510659
#> 37: NA <NA> 2.072637870
#> 38: NA <NA> 3.153765268
#> 39: NA <NA> 3.419489935
#> 40: NA <NA> 5.558785921
#> 41: NA <NA> 1.012589238
#> 42: NA <NA> 2.015986064
#> 43: NA <NA> 3.170339048
#> 44: NA <NA> 4.281061044
#> 45: NA <NA> 4.988645174
#> 46: 1 mean 0.945622941
#> 47: 1 median 0.619859008
#> 48: 2 mean 0.446385905
#> 49: 2 median 0.420682183
#> 50: 3 mean 0.397613162
#> 51: 3 median 0.306781776
#> 52: 4 mean 0.277069910
#> 53: 4 median 0.165324176
#> 54: 5 mean 0.182106656
#> 55: 5 median 0.144752066
#> 56: 1 <NA> 0.942306695
#> 57: 2 <NA> 0.453601669
#> 58: 3 <NA> 0.292487926
#> 59: 4 <NA> 0.210416655
#> 60: 5 <NA> 0.189556888
#> prob.pars.lambda algo.pars.method result.res