I use lattice package and 'barchart' to build a chart. I have a
problem with setting different x-axes. Some x categories are missing
but they are display and I don't want. I use scales = list(y
"free",x="free") but it works only for y-axes. Simple
example:
package(lattice)
barchart(yield ~ variety | site, data = barley,
groups = year, layout = c(1,6),
scales = list(y = "free",x="free"),
auto.key=list(rectangles = TRUE, space = "bottom"),
)
Let's assume that in the first panel 'Waseca' the category
'Velvet'
are not available in data but in the chart there is a empty place. If
more categories are not available it looks bad. Simple modification to
illustrate my problem:
barley2 <- barley[barley[,2]!="Velvet" |
barley[,4]!="Waseca",]
barchart(yield ~ variety | site, data = barley2,
groups = year, layout = c(1,6),
scales = list(y = "free",x="free"),
auto.key=list(rectangles = TRUE, space = "bottom"),
)
Robert
If I understand you correctly, then what you propose makes very little sense to me. In your barley2 example, I certainly would want to see the empty space for 'Velvet'. If you want a separate plot for each site, with no common 'variety' axis, then just build separate plots. Of course you would lose the ability to easily scan the vertical direction to compare varieties at different sites. -Peter Ehlers Robert Ruser wrote:> I use lattice package and 'barchart' to build a chart. I have a > problem with setting different x-axes. Some x categories are missing > but they are display and I don't want. I use scales = list(y > "free",x="free") but it works only for y-axes. Simple example: > > package(lattice) > barchart(yield ~ variety | site, data = barley, > groups = year, layout = c(1,6), > scales = list(y = "free",x="free"), > auto.key=list(rectangles = TRUE, space = "bottom"), > ) > > Let's assume that in the first panel 'Waseca' the category 'Velvet' > are not available in data but in the chart there is a empty place. If > more categories are not available it looks bad. Simple modification to > illustrate my problem: > > barley2 <- barley[barley[,2]!="Velvet" | barley[,4]!="Waseca",] > barchart(yield ~ variety | site, data = barley2, > groups = year, layout = c(1,6), > scales = list(y = "free",x="free"), > auto.key=list(rectangles = TRUE, space = "bottom"), > ) > > > Robert > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > >-- Peter Ehlers University of Calgary
On Sun, Jan 24, 2010 at 5:24 AM, Robert Ruser <robert.ruser at gmail.com> wrote:> I use lattice package and 'barchart' to build a chart. I have a > problem with setting different x-axes. Some x categories are missing > but they are display and I don't want. I use scales = list(y > "free",x="free") but it works only for y-axes. Simple example: > > package(lattice) > barchart(yield ~ variety | site, data = barley, > ? ? ? ?groups = year, layout = c(1,6), > ? ? ? ?scales = list(y = "free",x="free"), > ? ? ? ?auto.key=list(rectangles = TRUE, space = "bottom"), > ? ? ? ?) > > Let's assume that in the first panel ?'Waseca' ?the category 'Velvet' > are not available in data but in the chart there is a empty place. If > more categories are not available it looks bad. Simple modification to > illustrate my problem: > > barley2 <- barley[barley[,2]!="Velvet" | barley[,4]!="Waseca",] > barchart(yield ~ variety | site, data = barley2, > ? ? ? ?groups = year, layout = c(1,6), > ? ? ? ?scales = list(y = "free",x="free"), > ? ? ? ?auto.key=list(rectangles = TRUE, space = "bottom"), > ? ? ? ?)The short answer is that empty factor levels will be omitted only if they are at the boundary (because what are actually plotted on the x-axis here are the numeric factor codes). For example, barchart(yield ~ reorder(variety, yield, length) | site, data = barley2, groups = year, layout = c(1,6), scales = list(y = "free",x="free")) If you want to omit levels in the middle, you need to actually change the internal codes using 'x[drop=TRUE]' where 'x' is your factor with the missing levels. Lattice has no built-in facility to do that, but you can do it yourself in custom prepanel and panel functions. -Deepayan