Please read
http://cran.r-project.org/doc/manuals/R-exts.html
especially Section 1.1.3. Use save to create an Rdata file.
Max
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Wallace Hui
Sent: Wednesday, September 05, 2007 4:18 PM
To:
Rick,
First, please send the results of SessionInfo so I know exactly what
versions you are using. Also, are you using OpenOffice? If so, what
version?
If you use the current version of odfWeave, there is a file with many
details about formatting in the examples subdirectory.
Here is what I
It appears to be an OpenOffice bug:
http://www.openoffice.org/issues/show_bug.cgi?id=73163
Max
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Kuhn, Max
Sent: Friday, August 24, 2007 1:52 PM
To: Rick Bilonick; R Help
Subject: Re: [R] Using odfWeave
Aric,
Can you send me a reproducible example (code and odt file) plus the
results if sessionInfo()?
Thanks,
Max
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Aric Gregson
Sent: Thursday, August 09, 2007 6:56 PM
To: r-help@stat.math.ethz.ch
Subject:
This might work:
numOnly - function(x) gsub([^0-9], , x)
numOnly(lema, rb 2%)
[1] 2
numOnly(rb)
[1]
Max
-Original Message-
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of GOUACHE David
Sent: Monday, July 30, 2007 7:59 AM
To: r-help@stat.math.ethz.ch
Subject: [R]
Florian,
The first thing that you should change is how you call randomForest.
Instead of specifying the model via a formula, use the randomForest(x,
y) interface.
When a formula is used, there is a terms object created so that a model
matrix can be created for these and future observations. That
Rick,
First, can you send details about your versions/OS using sessionInfo()?
If you create plots and use odfWeave to create an odf text document,
the
default for figures is png bitmap graphics. The only way I've found to
insert eps graphics is to first create the eps graphic using the
Shirley,
The contrast package can do this. The method of specifying the contrast
conditions/coefficients is different (and I think easier).
Max
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of shirley zhang
Sent: Monday, July 09, 2007 11:16 AM
To:
, July 09, 2007 11:50 AM
To: Kuhn, Max
Cc: R-help@stat.math.ethz.ch
Subject: Re: [R] similar limma's contrasts.fit() for lme (mixed effect
model)object
Hi Max,
Thanks for your prompt reply. Actually I have already checked contrast
package, but I still could not figure out how to set the contrast
Dieter,
What is the version of odfWeave? If you use those specifications in the
current version (0.5.9):
plotInfo - getImageDefs()
plotInfo$type = eps
plotInfo$device = postscript
setImageDefs(plotInfo)
you will probabiliy need to set
horizontal = FALSE, onefile = FALSE, and paper = special
Des,
Please provide some information (like the results of sessionInfo) and
example code for your model. Did you use gee to fit the model and with
what function? If could be that you need to examine the type argument of
the predict function to make sure that you are getting the scale that
you are
Anup,
There are two ways to pass arguments to functions in R: as named
arguments or by position*.
Users *can* supply arguments that are inconsistent with the order that
you specify in the function definition, but only if they are used as
named arguments:
myfun(X = someMatrix, values =
A new version of odfWeave has been released to CRAN. This is a
significant change to the package internals. It now uses the XML library
instead of a bunch of regular expressions.
New features include:
- Captions for tables and figures
- Functions to insert page breaks and to change the
Greg,
I have looked at ooconvert, but unfortunately the current version is
limited to *nix, and I am currently in a MS windows world. I tried
getting it to run under cygwin at one point, but did not succeed at
the
time. Once it is available more generally, I will just set up
Makefiles
so
Greg,
Thanks for the kind words about odfWeave.
These reports are usually put out as internal webpages for various
people in the organization to look at, so we could either go the
odfWeave approach and generate pdf files (not as automated as I would
like)
I agree that automating the
Jason,
Your best bet is the nnet package. It is part of the VR bundle, which
may be why you missed it.
The code is well documented and the package is closely tied to two books
(see the references in ?nnet).
Also, it has a predict function, which many of the others do not. This
isn't too big
Laurence,
I haven't seen any issues like this. Can you:
1. Send the results of sessionInfo()
2. Try using carte2, echo = FALSE, results= xml, fig = TRUE=. It
sounds like the text being written to the xml file is not valid xml.
This could be the case if results != xml.
3. If that doesn't
Sarah and Laurence,
A few comments:
1. The default background color for columns is horrible. I've changed
to white it in the upcoming version.
2. In the next version (in 1-2 weeks), I have a fairly long document
that goes into much more detail about the specific styles that can be
changed
Message-
From: Uwe Ligges [mailto:[EMAIL PROTECTED]
Sent: Tuesday, June 05, 2007 3:56 PM
To: Saeed Abu Nimeh
Cc: Kuhn, Max; r-help@stat.math.ethz.ch
Subject: Re: [R] naiveBayes other than e1071
Saeed Abu Nimeh wrote:
Max,
Thanks. I have tried it but i keep getting an error:
Error
Saeed,
There is a version in the klaR package. I recently submitted a change to
the predict function that may be related to your problem.
If:
1. the posterior probabilities (apart from the prior) are being
approximated by the product of the p(x_i|y_j) and
2. a lot of predictors are being
Ted,
This is pretty simple. If you are new to programming with R, please read
the Introduction to R at
http://cran.r-project.org/doc/manuals/R-intro.html
This will answer most of the basic questions, but it requires more time
than typing out an email.
If you still have questions, please
Erin,
There is a snow package (note the case) and also a few others.
Rlsf, is specific to grids/clusters that use the LSF queue system.
More generally, the nws package is more sophisticated and should work on
most systems. Also, there is Rmpi (which I haven't used).
Max
-Original
Ken,
lme.fnl.REML - lme(Max ~ S + Tr + Yr + Tr:Yr, random = ~1 |TID,
method = REML)
You have three variables in the model: S, Tr and Yr.
tst - contrast(lme.fnl.REML, a=list(Yr=levels(Yr), Tr=R), b=list
(Yr=levels(Yr, Tr=T97)))
Error in gendata.default(fit = list(modelStruct =
Ken,
Take a look at the just released contrast package.
Max
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Ken Nussear
Sent: Monday, April 30, 2007 6:12 PM
To: r-help@stat.math.ethz.ch
Subject: [R] Independent contrasts from lme with interactions
Hi
The contrast and desirability packages are available for all platforms
at cran.r-project.org (and coming soon to a mirror near you).
The contrast package extends Frank Harrell's contrast.Design function
for one degree of freedom contrasts of model parameters to other types
of models, such as lm,
Wow.
Some suggestions are below.
1. You should vectorize the for loop in the function g. I would do
something like this:
z1 - rnorm(20 * 100, 0, 1)
Q1=0+1*z1*(1+.83*(1-exp(-G*z1))/(1+exp(-G*z1)))*(1+z1**2)**K
Q1Matrix - matrix(Q1, ncol = 100)
ksStat - function(x) ks.test(x, pnorm, mean =
Sorry, there was a typo in this line:
function(x) ks.test(x, pnorm, mean = mean(x), sd = sd(x))$p.value
Max
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Kuhn, Max
Sent: Wednesday, April 11, 2007 9:55 AM
To: [EMAIL PROTECTED]; r-help
Shuji,
I suspect that bruto blows up because your data are linearly separable.
To see this (if you didn't already know), try
library(lattice)
splom(~x, groups = y)
and look at the first row. If you are trying to do classification, there
are a few methods that would choke on this (logistic
approaches.
Max
-Original Message-
From: ?? ?? [mailto:[EMAIL PROTECTED]
Sent: Monday, April 09, 2007 1:49 PM
To: Kuhn, Max
Cc: r-help@stat.math.ethz.ch
Subject: Re: [R] Could not fit correct values in discriminant analysis
by bruto.
Dear Max,
Thank you very much ! Your sample code is very
At the risk of beating a dead horse...
odfWeave can be used on Linux (or anywhere Open Office is available) to
create reports that can then be converted to Word, rtf, html, pdf etc.
In still in the process of getting Impress (OO's presentation
application) working, but I'll get there.
Max
Fast is a relative term. Tell us more about your version, system, data
and the kernel that you are using (I'm guessing it is a RBF).
If you are building a classification model and there are two classes,
the svmpath package can be used to get quick predictions over the
cost/regularization
Aimin,
Please see color.palette in ?plot.tune.
Sorry to be a broken record, but if you want people to help you, you
should:
- read the help pages before sending an email
- run sessionInfo() and send it in the email
You run the risk of people ignoring your emails. These steps will help
Aimin,
I haven't used that function and you haven't old us anything about your
data or the system that you are on, but here is a guess: when the number
of votes are tied, a random choice is usually made between the classes.
0.63 and 0.67 are relatively close and this might account for the
: Kuhn, Max
Subject: RE: [R] RDA and trend surface regression
Thanks a lot for your answers,
I am concerned by your advice not to use polynomial constraints, or to
use
QDA instead of RDA. My final goal is to perform variation partitioning
using
partial RDA to assess the relative importance
Helene,
You will have to give us more information, such as your system/versions
and a small reproducible example. We try to stress that questions are
more easily answered when there are a lot of specific details given and
a reproducible case can be tested.
Here are two comments though:
1. The
).
Max
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Kuhn, Max
Sent: Tuesday, February 06, 2007 5:10 PM
To: Doran, Harold; R-help@stat.math.ethz.ch
Subject: Re: [R] R in Industry
As someone who has (reluctantly) sent job postings to R Help, I think
Chris,
You might be able to get it using format. You would have to convert the
data frame to a matrix if you want the row names to be dots too.
foo - function(x, top = 3, ...)
{
if(dim(x)[1] top + 3) stop(not enough rows)
charX - format(x, ...)
charX - charX[c(1:(top+2),
As someone who has (reluctantly) sent job postings to R Help, I think
that a SIG would be a good idea.
Max
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Doran, Harold
Sent: Tuesday, February 06, 2007 2:08 PM
To: R-help@stat.math.ethz.ch
Subject: [R] R
You can use the anova function a la:
anova(model1, model2)
Analysis of Variance Table
Model 1: y ~ x
Model 2: y ~ x + z
Res.DfRSS Df Sum of Sq F Pr(F)
1 13 4.4947
2 12 4.4228 10.0720 0.1952 0.6665
I would suggest
Laurent,
Since you question was not about odfWeave (despite the message title),
it would have been better to send this question to an OpenOffice mailing
list.
That said, I think that the issues was your workflow (odt - html -
doc). HTML is plain text, so when you saved the file in that format
I've verified this bug and will be releasing a new version shortly.
Max
-Original Message-
From: Abhijit Dasgupta [mailto:[EMAIL PROTECTED]
Sent: Saturday, December 23, 2006 12:03 AM
To: r-help@stat.math.ethz.ch; Kuhn, Max
Subject: bug in odfWeave
Hi,
I think there is a minor bug
How about:
apply(iris[, 1:4], 2, summary)
Sepal.Length Sepal.Width Petal.Length Petal.Width
Min. 4.300 2.0001.000 0.100
1st Qu.5.100 2.8001.600 0.300
Median 5.800 3.0004.350 1.300
Mean 5.843
Francesco,
odfTable writes out a table in xml format directly into an xml file. The
issues is that the characters and (among others) have to be
converted to gt; and lt;.
I will build a new version where this happens automatically. In the
meantime, you should be able to use the colnames
Pfizer Global Research and Development is seeking non-clinical
statisticians to support several areas of the drug development pipeline.
Positions available in Connecticut are:
- a non-clinical statistician to support non-clinical areas of RD,
including: Discovery, Drug Safety RD, and the
Whoops. I hit the wrong button! Here's the whole message:
And on that note, here is a function that I use to get around it:
startPNG - function(filename, width = 680, height = 990, ...)
{
nativePNG - capabilities(png)
if(nativePNG)
{
png(file = filename, width = width,
And on that note, here is a function that I use to get around it:
-Original Message-
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Jeffrey Horner
Sent: Friday, November 03, 2006 10:01 AM
To: [EMAIL PROTECTED]
Cc: r-help@stat.math.ethz.ch
Subject: Re: [R] R CMD BATCH:
Alfonso,
It sounds like unzip and zip are not in your path (as opposed to odfWeave
needing additional configuration). The error message that you received is
generated when odfWeave is invoked. It checks to see if unzip can be used at
the command line.
After googling, I found a basic link for
Version 0.4.4 of odfWeave is available from CRAN. A Windows binary
should be available shortly.
This version requires base R version 2.3.1 or greater.
Changes from the last version include
- Non-English character sets are handled better. For example, Chinese
characters can be included in R
I was hoping that someone could try to reproduce an error that I am
getting. The R Site Search keeps timing out on me, so apologies of this
has already come up.
I'm using
R.version
_
platform i386-pc-mingw32
arch i386
A new version of odfWeave is on CRAN. Changes include:
- handling of locales. Errors were being produced when locales were set
to anything but C. The fix changes the locale to C and changes back
to the original locale when the user's code is executed.
- a bug fix for default plot device units
Dr. Harrell,
I tried odfWeave to create an OpenOffice file and found that it
exhausted the memory of my large linux machine and took a long time to
run.
Do you have any details about the problem that you encountered? A bug
that someone else had pointed out might be the culprit. I have the
On page 2 of the manual, Section 2 (Requirements) has
The package also requires a utility to zip and unzip compressed
files,
such as unzip, Winzip or jar.
With a footnote that unzip can be found at
http://www.info-zip.org/
for free. You can use any utility that can zip/unzip pkzip
The odfWeave package is now available on CRAN at
http://lib.stat.cmu.edu/R/CRAN/src/contrib/Descriptions/odfWeave.html
and your local mirror.
The package extends Sweave to Open Document Format (ODF) text document
files. Latex-style code chunks and in-line Sexpr commands can be used
to embed R
I've been meaning send an announcement for this package, but Greg Snow
beat me to the punch today.
Max
snip
The odfWeave package is now available on CRAN at
http://lib.stat.cmu.edu/R/CRAN/src/contrib/Descriptions/odfWeave.html
and your local mirror.
The package extends Sweave to Open
One correction... since you are fitting a logistic model, it is
technically correct to say the mean value of the linear predictor,
instead of mean response.
20 lashes for me.
Max
-Original Message-
From: Kuhn, Max
Sent: Friday, July 07, 2006 4:11 PM
To: 'r-help@stat.math.ethz.ch
Hi,
I'm creating some lattice plots that have a key and I'd like to put a
label on the key. The problem is that the text label for the key
prevents the values of the group variables from being shown (see example
below). I don't think that this is a feature, but I might be abusing the
title arg
Peter and Sundar,
Thanks, I did miss that. I used:
key = list(
title = More Text,
cex.title = 1,
columns = 2,
text=list(
to get what I wanted.
Max
-Original Message-
From: P Ehlers [mailto:[EMAIL PROTECTED]
Sent: Thursday, June 01, 2006 4:14 PM
To: Kuhn, Max
Cc: r
The family arg appears to be the problem. Either bernoulli or adaboost
are appropriate for classification problems.
Max
Perhaps by following the Posting Guide you're likely to get more
helpful
responses. You have not shown an example that others can reproduce,
not
given version information
KJ,
In my job I write custom computer programs for data analysis, which
are used in our company's consulting business. Whenever I've needed
statistical analyses I've coded the algorithms myself, but my boss
wants me to start learning and using R, to speed up development.
I'm curious as to what
Emma,
One way is to use lattice:
test - data.frame(y = rnorm(5 * 50),
groupings = rep(letters[1:5], 50))
library(lattice)
densityplot( ~ y, group = groupings, data =test)
HTH,
Max
emma hartnett emma_hartnett at yahoo.co.uk
Wed Sep 21 17:57:44 CEST 2005
* Previous message:
Hi,
I am trying to implement the Adaboost.M1. algorithm as described in
The Elements of Statistical Learning p.301
I don't use Dtettling 's library boost because :
- I don't understande the difference beetween Logitboost and L2boost
- I 'd like to use larger trees than stumps.
It also
Jenny,
It didn't work and They worked aren't very specific. Also, the package
name is ipred and the function is errorest.
The estimator entry on the man page for errorest has:
'cv' cross-validation, 'boot' bootstrap or '632plus' bias corrected
bootstrap (classification only).
Note the
62 matches
Mail list logo