Hi list. I was saving my modified pairs graphic using a custom panel from
the R Graphics site, and I got an interesting difference in the final image
when I save it as eps or png.
This custom panel make possible to show at the left side of the pairs plot
the p-value as symbol and the correlation r value with its cex proportional
to the r value itself.
Well, when saving as a png file everything is pretty fine. But, if I save it
as an eps, one of the r values (0.0056) gets extremely bigger than it should
, and it's the only one presenting this behavior.
P.S.: I use GIMP 2.6.11 (x86) and GhostScript 9.00 (x86) to render the eps
file, I don't know if it could be an issue of one of these other softwares.
Below are the script that generates everything, a link to an example matrix
(csv) and a Sys.info result.
The link to an example data
http://dl.dropbox.com/u/2613625/Groups.csv
The script without the part used to import the data
# Analysis
cor.prob <- function(X, dfr = nrow(X) - 2) {
R <- cor(X)
above <- row(R) < col(R)
r2 <- R[above]^2
Fstat <- r2 * dfr / (1 - r2)
R[above] <- 1 - pf(Fstat, 1, dfr)
R
}
cor.prob(Grupos[,-7])
# Custom upper pairs panel
panel.cor <- function(x, y, digits=2, prefix="", cex.cor)
{
usr <- par("usr"); on.exit(par(usr))
par(usr = c(0, 1, 0, 1))
r <- abs(cor(x, y))
txt <- format(c(r, 0.123456789), digits=digits)[1]
txt <- paste(prefix, txt, sep="")
if(missing(cex.cor)) cex <- 0.8/strwidth(txt)
test <- cor.test(x,y)
# borrowed from printCoefmat
Signif <- symnum(test$p.value, corr = FALSE, na = FALSE,
cutpoints = c(0, 0.001, 0.01, 0.05, 0.1, 1),
symbols = c("***", "**", "*",
".", " "))
text(0.5, 0.5, txt, cex = cex * r)
text(.8, .8, Signif, cex=cex, col=2)
}
# Saving the Graphics
postscript('CorMatrix.eps',paper='special',onefile=F,horizontal=F,bg='white'
,width=7,height=7,pointsize=9) # Here the issue pops up
par(pty='m')
pairs(Grupos[,-7],lower.panel=panel.smooth,upper.panel=panel.cor)
dev.off()
png('CorMatrix.png',bg='white',width=2800,height=2800,res=300) #
Here
everything goes fine
par(pty='m')
pairs(Grupos[,-7],lower.panel=panel.smooth,upper.panel=panel.cor)
dev.off()
# System Information
Sys.info()
sysname release version nodename machine
login
"Windows" "7 x64" "build 7600"
"RODRIGO-PC" "x86-64"
"Rodrigo"
user
"Rodrigo"
Hope you can help me find out what is going on!
Regards
-------------------------------------------------------------------
MSc. <mailto:r.aluizio@gmail.com> Rodrigo Aluizio
[[alternative HTML version deleted]]
Hi, a correction in the previous script. Anyone who tries to use it must
ignore (remove) the [,-7] after the Grupos object.
Sorry.
De: Rodrigo Aluizio [mailto:r.aluizio@gmail.com]
Enviada em: terça-feira, 11 de janeiro de 2011 11:01
Para: R Help
Assunto: Postscript function Bug at R x64 2.12.1?
Hi list. I was saving my modified pairs graphic using a custom panel from
the R Graphics site, and I got an interesting difference in the final image
when I save it as eps or png.
This custom panel make possible to show at the left side of the pairs plot
the p-value as symbol and the correlation r value with its cex proportional
to the r value itself.
Well, when saving as a png file everything is pretty fine. But, if I save it
as an eps, one of the r values (0.0056) gets extremely bigger than it should
, and it’s the only one presenting this behavior.
P.S.: I use GIMP 2.6.11 (x86) and GhostScript 9.00 (x86) to render the eps
file, I don’t know if it could be an issue of one of these other softwares.
Below are the script that generates everything, a link to an example matrix
(csv) and a Sys.info result.
The link to an example data
http://dl.dropbox.com/u/2613625/Groups.csv
The script without the part used to import the data
# Analysis
cor.prob <- function(X, dfr = nrow(X) - 2) {
R <- cor(X)
above <- row(R) < col(R)
r2 <- R[above]^2
Fstat <- r2 * dfr / (1 - r2)
R[above] <- 1 - pf(Fstat, 1, dfr)
R
}
cor.prob(Grupos[,-7])
# Custom upper pairs panel
panel.cor <- function(x, y, digits=2, prefix="", cex.cor)
{
usr <- par("usr"); on.exit(par(usr))
par(usr = c(0, 1, 0, 1))
r <- abs(cor(x, y))
txt <- format(c(r, 0.123456789), digits=digits)[1]
txt <- paste(prefix, txt, sep="")
if(missing(cex.cor)) cex <- 0.8/strwidth(txt)
test <- cor.test(x,y)
# borrowed from printCoefmat
Signif <- symnum(test$p.value, corr = FALSE, na = FALSE,
cutpoints = c(0, 0.001, 0.01, 0.05, 0.1, 1),
symbols = c("***", "**", "*",
".", " "))
text(0.5, 0.5, txt, cex = cex * r)
text(.8, .8, Signif, cex=cex, col=2)
}
# Saving the Graphics
postscript('CorMatrix.eps',paper='special',onefile=F,horizontal=F,bg='white'
,width=7,height=7,pointsize=9) # Here the issue pops up
par(pty='m')
pairs(Grupos[,-7],lower.panel=panel.smooth,upper.panel=panel.cor)
dev.off()
png('CorMatrix.png',bg='white',width=2800,height=2800,res=300) #
Here
everything goes fine
par(pty='m')
pairs(Grupos[,-7],lower.panel=panel.smooth,upper.panel=panel.cor)
dev.off()
# System Information
Sys.info()
sysname release version nodename machine
login
"Windows" "7 x64" "build 7600"
"RODRIGO-PC" "x86-64"
"Rodrigo"
user
"Rodrigo"
Hope you can help me find out what is going on!
Regards
-------------------------------------------------------------------
MSc. Rodrigo Aluizio <mailto:r.aluizio@gmail.com>
[[alternative HTML version deleted]]