Right now, there is a loss of information when one melts a list of columns simultaneously since the variable column only gives the number of the group (as a factor by default), not a name.
It would be great if the functionality of melting a list of columns simultaneously would be extended such that it is possible to provide labels for the factor in the variable ###column. I would suggest an additional argument, e.g. variable.labels which makes it possible to provide custom labels.
# Minimal reproducible example
library(data.table)
DT <- data.table(val_loss = 1:2,
val_acc = 3:4,
loss = 5:6,
acc = 7:8,
id = c("A", "B"))
DT_melted <-
melt(DT,
measure.vars = list(c("loss", "acc"), c("val_loss", "val_acc")),
# the following line shows what I think would be ideal
# variable.labels = c("group 1", "group 2"),
value = c("training", "validation"))
# the following step should become obsolete by providing the labels during the melt call above
DT_melted[, variable := factor(variable, levels = 1:2, labels = c("group 1", "group 2"))]
# Output of sessionInfo()
R version 3.3.3 (2017-03-06)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 7 x64 (build 7601) Service Pack 1
locale:
[1] LC_COLLATE=German_Germany.1252 LC_CTYPE=German_Germany.1252 LC_MONETARY=German_Germany.1252
[4] LC_NUMERIC=C LC_TIME=German_Germany.1252
attached base packages:
[1] stats graphics grDevices utils datasets methods base
other attached packages:
[1] data.table_1.10.4-3
loaded via a namespace (and not attached):
[1] tools_3.3.3
I know that the R Version is not up to date but at my Company it takes a while until we get the most recent version.
Sorry, for not labeling the issue. Somehow I could not figure out, how to set them.
Right now, there is a loss of information when one melts a list of columns simultaneously since the variable column only gives the number of the group (as a factor by default), not a name.
It would be great if the functionality of melting a list of columns simultaneously would be extended such that it is possible to provide labels for the factor in the variable ###column. I would suggest an additional argument, e.g. variable.labels which makes it possible to provide custom labels.
# Minimal reproducible example
library(data.table) DT <- data.table(val_loss = 1:2, val_acc = 3:4, loss = 5:6, acc = 7:8, id = c("A", "B")) DT_melted <- melt(DT, measure.vars = list(c("loss", "acc"), c("val_loss", "val_acc")), # the following line shows what I think would be ideal # variable.labels = c("group 1", "group 2"), value = c("training", "validation")) # the following step should become obsolete by providing the labels during the melt call above DT_melted[, variable := factor(variable, levels = 1:2, labels = c("group 1", "group 2"))]# Output of sessionInfo()
I know that the R Version is not up to date but at my Company it takes a while until we get the most recent version.
Sorry, for not labeling the issue. Somehow I could not figure out, how to set them.