[R] [R-pkgs] new version of abind()
Tony Plate
tplate at acm.org
Sat Mar 20 22:21:34 CET 2004
There is a new version of the abind package on CRAN (abind_1.1-0). abind()
is a multi-dimensional generalization of cbind() and rbind() -- it can bind
multiple 2-d matrices into a 3-d array, or bind 3-d arrays together, etc.
In this new version the behavior of the function abind() has been enhanced
slightly (it can now accept a list as the first argument, removing the need
to use do.call() in most situtations), and changed slightly (to no longer
by default creates dimension names if none exist, because these are usually
useless and mostly annoying). There is also a new function in the package:
adrop(), which does approximately what drop() does, but allows control over
which dimensions are dropped.
Note also that the abind() function runs under S-plus as well (I believe
adrop() does as well, but it is not as well tested).
Details:
2004-03-12
* R/abind.R man/abind.Rd
allow first argument of abind() to be a list of objects to be
bound -- this avoids the need for do.call() when one wants to bind
a list of objects
* R/abind.R man/abind.Rd
changed argument name 'use.anon.names' to the more intuitive
'make.names' (the argument 'use.anon.names' still works)
* R/abind.R man/abind.Rd
changed default value for 'make.names' to FALSE (now more closely
behaves like rbind() and cbind()). This means that dimension
names for dimensions that have no names are only constructed
when requested, not by default.
* R/adrop.R man/adrop.Rd
added new function adrop(). This is a function like drop(), but it
allows to user to specify which of the dimensions with extent one
will be dropped.
Please let me know of any problems.
-- Tony Plate
Tony Plate tplate at acm.org
_______________________________________________
R-packages mailing list
R-packages at stat.math.ethz.ch
https://www.stat.math.ethz.ch/mailman/listinfo/r-packages
More information about the R-help
mailing list