using R as a tool for high quality scientific visualisations

1. Scientific Visualisation using R
f.jamitzky@lrz.de

2. Schedule
1.Introduction to R
2.Graphics library
3.texmacs notebook
4.iplots library
5.rgl
overview for MATLAB and
IDL convertites
standard plotting
Mathematica style notebook
with latex output

3. Introduction to R

4. Starting R
In the terminal type:
$ R
R version 2.5.1 (2007-06-27)
Copyright (C) 2007 The R Foundation for Statistical Computing
ISBN 3-900051-07-0
>
$ R --no-save < myfile.R
$ R CMD BATCH myfile.R

5. Running programs
Open myfile.R in your favorite editor and type:
print("Hello World from R!")
Load R source file and execute
source("myfile.R")
Define a function
hello <- function(x) paste("Hello","World","from",x)
hello("R")

6. Getting Information
Sys.info()
system("pwd",int=T)
system.time(sum(1:100))
x=runif(100)
str(x)
print(x)
fix(x)
summary(x)
get system info
call shell command
get elapsed time
random uniform numbers
structure of x
print
quick edit
statistical summary

7. Vectors
a = c(4,5,3,4,3)
a=1:5
a=rep(1:5,2)
a=rep(1:4,1:4)
a[-(1:3)]
a[a > 2]
define vector (concat)
define range
repeat data
repeat individual
delete indices
select elements

8. Matrices
a = matrix(runif(9),c(3,3))
a= matrix(1, c(3,3))
b = a %*% a
b = a %x% a
b = a %o% a
d=diag(1:3)+1
d[2,]
d[,3]
square matrix 3x3
matrix product
kronecker product
outer product
diagonal matrix
2nd row
3rd column

9. Matrices II
t(a)
solve(a)
a %*% solve(a)
data.matrix(a)
transpose matrix
inverse of matrix
get data as matrix

10. Arrays
a=array(runif(24),c(2,3,4))
apply(a,2,sum)
sum(a)
dim(a)
as.vector(a)
c(a)
3-dim array
sum array over 2nd index
total sum of a
information about a
map to 1d vector

11. Lists
a=list(x=1:3,y=c("a","b","c"))
names(a)
unlist(a)
str(a)
c(a,a)
define list
get names of a
make vector
structure of a
concatenate

12. Factors
factors are also called
categories or enunms
a=c("a","b","a","c","a")
a.f = factor(a)
levels(a.f)
str(a.f)
summary(a.f)
vector
define factor
get unique levels of a
structure of factor
histogram

13. Data Frames
data frames are "excel" data
sheets
a=data.frame(x=1:3,
y=c("a","b","c"),z=runif(3))
str, summary
read.table()
read.csv()
read.delim()
read.fwf()
definition like a list
information about data frame
read data from source

14. Loops
a=matrix(runif(100),c(10,10))
apply(a,2,sum)
apply(a,2,function(x) sum(x))
lapply(1:10, function(x) runif(10))
sapply(1:10, function(x) runif(10))
mapply, vapply, tapply
define matrix
sum columns
apply function to columns
apply function to list
apply function to vector

15. Standard graphics lib

16. Plotting
x=1:100
y=runif(100)
plot(x,y)
lines(y/2, col="green")
points(y/3, col="red")
type="p"
"l"
"b"
"h"
"s"
define vectors
x-y dot plot
overplot lines
overplot points
points
lines
both
histogram
stairs

17. Keywords
plot(x,y,...)
main="The Main Title"
sub="the subtitle"
xlab="x axis"
ylab="y axis"
log="x"
par()
main title
figure caption
labels x and y
additional parameters

18. 2D Plots
data(volcano)
image(volcano)
contour(volcano,add=T)
image(volcano,
col=terrain.colors(100))
filled.contour(volcano)
load example data
image map
contour plot
different color table
rainbow, hsv, heat.colors,
topo.colors
nice contours

19. 3D Plots
wire frame (simple version)
persp(volcano)
wire frame (nice version)
z <- 2 * volcano
x <- 10 * (1:nrow(z))
y <- 10 * (1:ncol(z))
persp(x, y, z, theta = 135, phi = 30, col = "green3",
scale = FALSE, ltheta = -120, shade = 0.75,
border = NA, box = FALSE)

20. TeXmacs

21. Overview
GNU TeXmacs is a free wysiwyw (what you see is what you want)
editing platform with special features for scientists.
TeXmacs runs on all major Unix platforms and Windows.
Documents can be saved in TeXmacs, Xml or Scheme format and
printed as Postscript or Pdf files. Converters exist for TeX/LaTeX
and Html/Mathml.

22. WYSIWYG
TeXmacs supports
math notation and
can export to Latex.
you can run computer
algebra sessions inside
TeXmacs

23. Using R inside TeXmacs
log in to gvs
module load texmacs
> Insert > Session > R
> library(TeXmacs)
> data(volcano)
> image(volcano)
> contour(volcano,add=T)
> v()

24. iPlots and JGR

25. Installing JGR
goto
http://rforge.net/JGR/
> Downloads

26. Load Data to Object Browser
library(MASS)
data(Cars93)
attach(Cars93)
Ctrl-B or
select
Tools >
Object Browser
from menu

27. Scatterplot
iplot(Horsepower, MPG.city)

28. Histogram
ihist(Horsepower)

29. Mixing them together
Select some points in the scatterplot, then a sub-region
of the histogram is selected

30. Multiplot
imosaic(data.frame(AirBags,Cylinders,Origin))
ipcp(Cars93[c(4:8,12:15,17,19:25)])
iplot(Horsepower,EngineSize)
ihist(Horsepower)
ibox(Cars93[4:6])

31. Rainbow plot
Select from
menu:
View >
Set Colors
(rainbow)

32. Get Indices
You can obtain information about the selected points:
and select points
by the command
iset.select

33. Overview
●iset.brush()
●iset.col()
●iset.cur()
●iset.df()
●iset.list()
●iset.new()
●iset.next()
●iset.prev()
●iset.sel.changed()
●iset.select()
●iset.selectAll()
●iset.selected()
●iset.selectNone()
●iset.set()
● iplot.cur()
● iplot.data()
● iplot.list()
● iplot.new()
● iplot.next()
● iplot.off()
● iplot.opt()
● iplot.prev()
● iplot.set()
● iobj.cur()
● iobj.get()
● iobj.list()
● iobj.next()
● iobj.opt()
● iobj.prev()
● iobj.rm()
● iobj.set()

34. opengl binding

35. opengl bindings for R
●library(rgl)
●abstraction for opengl primitives:
osurfaces, spheres, triangles
opoints, lines, text, sprites
olights, background, materials, textures

36. Examples
example(bg3d)
demo(regression)
example(particles3d)

37. Primitives

38. Appearance

39. Simple Example: colored spheres
rgl.clear("all")
rgl.bg(sphere = T, back = "lines")
rgl.light()
rgl.bbox()
x=runif(1000)
y=runif(1000)
z=runif(1000)
col=rainbow(1000)
rad=runif(1000)/10
rgl.spheres(x,y,z,col=col,radius=rad)
clear space
set background
some light
bounding box
some coordinates
set colors
set radii
generate the spheres

40. rgl.spheres(x,y,z,col=col,radius=rad)

41. Simple Example: surface
rgl.clear("shapes")
data(volcano)
y = 2*volcano
x=10 * (1:nrow(y))
z=10 * (1:ncol(y))
rgl.surface(x, z, y, back="lines")
clear space
some data
define coordinates
plot surface

42. rgl.surface(x, z, y, back="lines")