Description Usage Arguments Value See Also Examples
View source: R/classification.R
Random Forest Thresholdout, which is TO with the feature selection and classifier replaced with Random Forest.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | privateRF(
train.ds = NULL,
holdout.ds = NULL,
validation.ds = NULL,
label = "class",
is.simulated = TRUE,
rf.importance.measure = "MeanDecreaseGini",
rf.ntree = 500,
rf.mtry = NULL,
pec.file = NULL,
update.freq = 50,
threshold = 4/sqrt(nrow(train.ds)),
tolerance = 1/sqrt(nrow(train.ds)),
signal.names = NULL,
save.file = NULL,
verbose = FALSE
)
|
train.ds |
A data frame with training data and outcome labels |
holdout.ds |
A data frame with holdout data and outcome labels |
validation.ds |
A data frame with validation data and outcome labels |
label |
A character vector of the outcome variable column name |
is.simulated |
Is the data simulated (or real?) |
rf.importance.measure |
A character vector for the random forest importance measure |
rf.ntree |
An integer the number of trees in the random forest |
rf.mtry |
An integer the number of variables sampled at each random forest node split |
pec.file |
A character vector filename of privateEC results |
update.freq |
A integer for the number of steps before update |
threshold |
A numeric, default 4 / sqrt(n) suggested in the thresholdout’s supplementary material (Dwork, et al.,2015) |
tolerance |
A numeric, default 1 / sqrt(n) suggested in the thresholdout’s supplementary material (Dwork, et al.,2015) |
signal.names |
A character vector of signal names in simulated data |
save.file |
A character vector for results filename or NULL to skip |
verbose |
A flag indicating whether verbose output be sent to stdout |
A list containing:
data frame of results, a row for each update
melted results data frame for plotting
number of variables detected correctly in each data set
total elapsed time
Other classification:
epistasisRank()
,
getImportanceScores()
,
originalThresholdout()
,
privateEC()
,
standardRF()
,
xgboostRF()
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 | num.samples <- 100
num.variables <- 100
pct.signals <- 0.1
label <- "class"
temp.pec.file <- tempfile(pattern = "pEc_temp", tmpdir = tempdir())
sim.data <- createSimulation(num.variables = num.variables,
num.samples = num.samples,
pct.signals = pct.signals,
label = label,
sim.type = "mainEffect",
pct.train = 1 / 3,
pct.holdout = 1 / 3,
pct.validation = 1 / 3,
verbose = FALSE)
pec.results <- privateEC(train.ds = sim.data$train,
holdout.ds = sim.data$holdout,
validation.ds = sim.data$validation,
label = sim.data$label,
is.simulated = TRUE,
signal.names = sim.data$signal.names,
save.file = temp.pec.file,
verbose = FALSE)
prf.results <- privateRF(train.ds = sim.data$train,
holdout.ds = sim.data$holdout,
validation.ds = sim.data$validation,
label = sim.data$label,
is.simulated = TRUE,
signal.names = sim.data$signal.names,
pec.file = temp.pec.file,
verbose = FALSE)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.