data.frame,
resamp, and
series
colMins(x, na.rm=F, dims=1, n=NULL) colMaxs(x, na.rm=F, dims=1, n=NULL) colRanges(x, na.rm=F, dims=1, n=NULL) rowMins(x, na.rm=F, dims=1, n=NULL) rowMaxs(x, na.rm=F, dims=1, n=NULL) rowRanges(x, na.rm=F, dims=1, n=NULL)
bdFrame.
FALSE, missing values (
NA) in the input result in
missing values in corresponding elements of the output.
TRUE, missing values are omitted from calculations.
x has dimension higher than 2,
dims determines what dimensions are summarized.
If
dims=3 and
x is a 5-dimensional array,
the result of
rowMeans is a 3 dimensional array consisting of the means
across the remaining 2 dimensions,
and the result of
colMeans is a 2 dimensional array consisting of the means
across the last 3 dimensions.
x as a matrix with
n rows.
For a matrix,
colMins(x) is equivalent to
apply(x, 2, min) except possibly for
trivial differences in how
dimnames are stored. Similarly,
rowMins(x)
matches
apply(x, 1, min). Corresponding relationships hold for the other functions.
If there are any missing values, then
apply is used for the calculations,
and computations are slower.
The
dims and
n arguments should not be used for a data frame or
a
bdFrame.
If
n is supplied then the result has no
names or
dimnames, and
the
dims argument is ignored.
n is useful for working with
vectors without converting them to matrices.
x <- matrix(1:12, 4) colMins(x) rowMins(x) colRanges(x) ## Summaries for regular subsets of a vector x <- 1:10 colMins(x, n=5) # like colMins(matrix(x, 5)) ## Higher-dimensional array x <- array(runif(24), dim=c(2,3,4)) rowMins(x) # vector of length 2. rowMins(x, dims=2) # 2x3 matrix. apply(x, 1:2, min) # same as previous colMins(x) # 3x4 matrix. colMins(x, dims=2) # vector of length 4. colMins(aperm(x, c(2,1,3))) # 2x4 matrix