This function requires the bigdata library section to be loaded.
bd.unstack(data, value.column,
group.columns=character(0),
key.columns=character(0),
names. = NULL,
drop.empty.columns=T,
sort=T)
bdFrame or
data.frame.
group.columns.
Any combination of key column values that are not unique
will lead to loss of data, as the last non-unique value will be used.
TRUE, prune empty columns from the output.
If
group.columns specifies multiple columns,
this function uses the cross-product of the
group.columns factor levels.
This can lead to category combinations that do not actually exist in the data.
The
prune.empty.columns argument determines
whether to remove these unused columns from the output.
FALSE, do not sort the input data by
c(key.columns,group.columns) first.
bdFrame or
data.frame,
of the same type as
data.
# Unstack fuel.frame "Weight" according to "Type" (use mean if there # are more than one occurence of each "Type"). bd.unstack(bd.aggregate(fuel.frame, "Weight", "Type", "mean", "Weight"), "Weight", "Type")