Dataset sizes in OpenML

dataset_size_openML.R

# OpenML Benchmarking Suites and the OpenML100
# https://arxiv.org/abs/1708.03731
# https://www.openml.org/s/14/data

library(OpenML)
ids <- getOMLStudy('OpenML100')$data$data.id
dsall <- listOMLDataSets()
sum(dsall$data.id %in% ids)   ## 96???
ds <- dsall[dsall$data.id %in% ids,]

median(ds$number.of.instances)
sum(ds$number.of.instances>=10000)
sum(ds$number.of.instances>=40000)
ds[ds$number.of.instances>=40000,]

#> median(ds$number.of.instances)
#[1] 2054.5
#> sum(ds$number.of.instances>=10000)
#[1] 21
#> sum(ds$number.of.instances>=40000)
#[1] 8

#> ds[ds$number.of.instances>=40000,c(2,10:12)]
#                      name number.of.classes number.of.features
#101            electricity                 2                  9
#413              mnist_784                10                785
#955         KDDCup09_churn                 2                231
#957     KDDCup09_upselling                 2                231
#1165        bank-marketing                 2                 17
#1207 tamilnadu-electricity                20                  4
#1287                 adult                 2                 15
#2550                 higgs                 2                 29
#     number.of.instances
#101                45312
#413                70000
#955                50000
#957                50000
#1165               45211
#1207               45781
#1287               48842
#2550               98050