[R] 3-D-Plot

Edwin Sendjaja edwin7 at web.de
Sat Apr 19 23:16:42 CEST 2008


Hi Uwe,

With lab= c(5,hostLength*2), the host names will be shown twice(double) in 
y-axis, which is wrong.

Both lab= c(5,hostLength) and  lab= c(5,hostLength*2) produces plots on the 
wrong postion.

As Attachment you can find the graphic with lab= c(5,hostLength*2).

Is there any other way to solve this?does scatterplot3d actually support this?



Edwin


Am Samstag, 19. April 2008 19:59:17 schrieb Uwe Ligges:
> Edwin,
>
> the problem is that the underlying R code treats the "lab" argument as
> something that should be approximate in some way.
>
> In your case, you can say, e.g.:
>
> par(lwd=3)
> scatterplot3d(.....
>                lab= c(5,hostLength*2),
>                .....)
>
> Best wishes;
> uwe
>
> Edwin Sendjaja wrote:
> > Hi Uwe,
> >
> > In your example, you have a numeric value in dataset.
> >
> > But my problem is that I have character value, like:uni Dortmund, uni
> > Berlin, etc.
> >
> >
> > --------------------------------------------
> > For example:
> >
> >   Name Student Ranking
> >
> > 1 Uni Dortmund 100000 1
> > 2. Uni Berlin 20000 3
> > ...etc
> >
> > ---------------------------------------
> >
> > I want that "Name" coloumn as y-axis.
> >
> >
> >
> > ########
> >
> > For detail I send you as attachment my data. In that graphic you can see
> > that I want the Host-name as y-axis. But unfortunately it doesn't plot
> > correctly at the right place. Besides that, the length of y-label doesn't
> > match like what I've configured. It is strange, because I have actually
> > hostLength= 10. but in graphic:
> > peeramidion.irisa.fr-Messung-19.04.08_14.57-Timeout50-Transmission500-Eur
> >opa-1-grafik-RTT_bigger_40000_3D.pdf
> >
> > Unfortunately, it shows only 9 Hosts.
> > -------------------------------------------------------------------------
> >----------------------------------
> >
> > The Code:
> >
> >
> > data <-
> > read.table("peeramidion.irisa.fr-Messung-TEST-50-500-Europa-1-registrar.d
> >ata") mapping <-
> > read.table("Messung-TEST-50-500-Europa-1-mappingfile-PE.data") mergeXY<-
> > merge(data,mapping, by.x="PE_ID", by.y="ID")
> >
> > mergeXY <- subset(mergeXY, (mergeXY$Ereignis == "EndpointKeepAlive") &
> > (mergeXY$Zusatz > 40000))
> >
> >
> > xSet <- mergeXY$RelTime
> > xTitle <- "Zeit [s]"
> >
> >
> > ySet <- mergeXY$Host
> > yTitle <- "PE-Hosts"
> >
> > zSet <- mergeXY$Zusatz / 1000.0
> > zTitle <- "Round Trip Time [ms]"
> >
> >
> > hostList <- levels(factor(ySet))
> > hostLength <- length(hostList)
> >
> > pdf("RESULT.pdf", width=17, height=25, onefile=TRUE, family="Helvetica",
> > pointsize=22)
> >
> > par(lwd=3)
> > scatterplot3d(xSet, ySet, zSet,
> >               xlab=xTitle, ylab=yTitle, zlab=zTitle,
> >               axis=TRUE, tick.marks=TRUE, label.tick.marks=TRUE,
> >               grid=TRUE, box=TRUE, highlight.3d=TRUE,
> >
> >               main="Round Trip Time - EndpointKeepAlive",
> >               sub="Pool Registrar: peeramidion.irisa.fr",
> >               color=par("col"), pch=16, type="h",
> >               scale.y=1, angle=40,
> >               mar=c(5,5,5,5), y.margin.add=7,
> >               font.axis=2, font.lab=2,
> >               lty.grid=2,
> >
> >               y.ticklabs=hostList,
> >               lab= c(5,hostLength),
> >               lab.z=5)
> >
> > dev.off()
> >
> > #++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> >++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++#
> >
> >
> > I hope you can help me to solve this problem.
> >
> >
> >
> >
> > Kind regards,
> >
> >
> >
> >
> > Edwin
> >
> > Am Freitag, 18. April 2008 13:58:33 schrieb Uwe Ligges:
> >> Edwin Sendjaja wrote:
> >>> Hi Uwe,
> >>>
> >>> I decided to use scatterplot3d, because it looks better.
> >>> I have some questions:
> >>>
> >>> Is it possible to get 1 axis( for example: z-axis) not as numeric, but
> >>> as character.
> >>>
> >>> Because I have date set like this:
> >>>
> >>> x=relative Time: 0,3 ms; 0,5ms, etc
> >>> y=Delay:10 ms, 20 ms, etc
> >>> z= Host: cnn.com, heise.de,etc
> >>
> >> See ?scatterplot3d. You can use arguments x.ticklabs, y.ticklabs,
> >> z.ticklabs in order top specify character strings.
> >>
> >>> If no, can you tell me little how to modify the code to get this. Thank
> >>> you.
> >>>
> >>> another question:
> >>>
> >>> How can I change the width line of the axis (x,y,z). The default line
> >>> is for me too thin.
> >>
> >> par(lwd=2) or more, for example.
> >>
> >>
> >> First example from ?scatterplot3d  but with some other labels on x axis
> >> and thick lines:
> >>
> >>    z <- seq(-10, 10, 0.01)
> >>    x <- cos(z)
> >>    y <- sin(z)
> >>    par(lwd=3)
> >>    scatterplot3d(x, y, z, highlight.3d=TRUE, col.axis="blue",
> >>        col.grid="lightblue", main="scatterplot3d - 1", pch=20,
> >>        x.ticklabs=paste(seq(0.3, 1.1, by=0.2), "ms"),
> >>        xlab="relative Time")
> >>
> >> Best wishes,
> >> Uwe
> >>
> >>> Thank you in advance,
> >>>
> >>>
> >>> Edwin Sendjaja
> >>>
> >>> Am Donnerstag, 17. April 2008 14:43:39 schrieb Uwe Ligges:
> >>>> Edwin Sendjaja wrote:
> >>>>> Hi Uwe,
> >>>>>
> >>>>> Thanks for your answer.
> >>>>>
> >>>>> What is the different between rgl and scatterplot3d? I dont need a
> >>>>> graphik like vulcano. I just need 3D-"dot"-plot.
> >>>>
> >>>> Sure, rgl can do it as well.
> >>>> Difference is that scatterplot3d is based on R's standard devices
> >>>> while rgl is based on an OpenGL device. Hence rgl is much more
> >>>> flexible and can rotate things and easily draws nice transparent
> >>>> forms, but scatterplot3d is nice for printing to 2D in different
> >>>> formats at the end.
> >>>>
> >>>> Uwe
> >>>>
> >>>>> Am Donnerstag, 17. April 2008 09:03:17 schrieb Uwe Ligges:
> >>>>>> Edwin Sendjaja wrote:
> >>>>>>> Hello,
> >>>>>>>
> >>>>>>> I've got some problems. I hope someone can help me.
> >>>>>>>
> >>>>>>> First question:
> >>>>>>> I am trying to get grid on scatterplot3d (from scatterplot3d
> >>>>>>> package). It seems that scatterplot3d draw on grip on X and Z side.
> >>>>>>> Is it possible to get Grid on the whole Box?
> >>>>>>
> >>>>>> At least there is no build in function to do it.
> >>>>>>
> >>>>>>> Second question:
> >>>>>>> Is it possible to use the standard package drawing 3d-plot(without
> >>>>>>> scatterplot3d) ?because now I have already 2d-plot. I just want to
> >>>>>>> add the z-axis.
> >>>>>>
> >>>>>> See, for example, ?cloud in package "lattice", ?plot3d in package
> >>>>>> "rgl",
> >>>>>>
> >>>>>>> Third question:
> >>>>>>> What is the best 3d-plot? it seems that scatterplot3 doesn't
> >>>>>>> support anything( like turning the box vertically. I think, angle
> >>>>>>> parameter just turn the box horisontally. I might wrong with this.
> >>>>>>
> >>>>>> You are right.
> >>>>>>
> >>>>>> Best wishes,
> >>>>>> Uwe
> >>>>>>
> >>>>>>> Thank a lot in advance
> >>>>>>>
> >>>>>>> Edwin
> >>>>>>>
> >>>>>>> ______________________________________________
> >>>>>>> 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.
> >>>>>>
> >>>>>> ______________________________________________
> >>>>>> 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.
> >>>>>
> >>>>> ______________________________________________
> >>>>> 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.
> >>>>
> >>>> ______________________________________________
> >>>> 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.


-------------- nächster Teil --------------
Ein Dateianhang mit Bin?rdaten wurde abgetrennt...
Dateiname   : peeramidion.irisa.fr-Messung-19.04.08_21.56-Timeout50-Transmission500-Europa-1-grafik-RTT_bigger_40000_3D.pdf
Dateityp    : application/pdf
Dateigr??e  : 71259 bytes
Beschreibung: nicht verf?gbar
URL         : https://stat.ethz.ch/pipermail/r-help/attachments/20080419/3900d693/attachment.pdf 


More information about the R-help mailing list