R help - Apr 2004 - Re: R-help Digest, Vol 14, Issue 3

At 12:01 03/04/04 +0200, you wrote:
>Content-Transfer-Encoding: 8bit
>From: solares at unsl.edu.ar
>Precedence: list
>MIME-Version: 1.0
>Cc:
>To: R-help at stat.math.ethz.ch
>Date: Fri, 2 Apr 2004 12:47:48 -0300 (ART)
>Message-ID: <50155.209.13.250.66.1080920868.squirrel at
inter17.unsl.edu.ar>
>Content-Type: text/plain;charset=iso-8859-1
>Subject: [R] convert excell file to text with RODBC package
>Message: 27
>
>Hi, i can convert excell to list in R with package RODBC ()but i don't
>understand 2 mistake
>1) Don't read the last row of the table excell
>2) Don' t take the hours
See below
>My excell file call prueba4.xls and have the following rows:
>where prueba4.xls was make in excell (office xp) and have one spreadsheet
>call "Hoja1", you see each rows of she:
>D??a     Hora    col1    col2    col3    col4    col5    col6    col7   
col8
>15/12/2003      12:14:59        217     2760    8,2     35      79,6    86,4
>15/12/2003      12:15:00        217     2764    8,2     35      79,6    86,4
>15/12/2003      12:15:01        217     2758    8,3     35      79,6    86,4
>15/12/2003      12:15:02        217     2760    8,3     35      79,6    86,4
>15/12/2003      12:15:03        217     2755    8,3     35      79,6    86,4
>15/12/2003      12:15:04        217     2766    8,3     35      79,6    86,4
>15/12/2003      12:15:05        217,1   2766    8,3     35,1    79,6    86,4
>15/12/2003      12:15:06        217,1   2758    8,3     35,1    79,6    86,4
>15/12/2003      12:15:07        217,1   2768    8,3     35,1    79,6    86,4
That seems to have 9 rows of data. There seem to be either 8 columns or 10 
columns of data.
>My code (i use the R 1.7.1 for windows xp) is the following:
>
>  library(RODBC)
> > canal<-odbcConnectExcel("c:/prueba4.xls")
> > tablas<-sqlTables(canal)
> > tablas
>     TABLE_CAT TABLE_SCHEM TABLE_NAME   TABLE_TYPE REMARKS
>1 c:\\prueba4        <NA>     Hoja1$ SYSTEM TABLE    <NA>
>2 c:\\prueba4        <NA>     Hoja2$ SYSTEM TABLE    <NA>
>3 c:\\prueba4        <NA>     Hoja3$ SYSTEM TABLE    <NA>
> > tbl<-sqlFetch(canal,substr(tablas[1,3],1,nchar(tablas[1,3])-1))
> > tbl[1]
>                   D??a
>1 2003-12-15 00:00:00
>2 2003-12-15 00:00:00
>3 2003-12-15 00:00:00
>4 2003-12-15 00:00:00
>5 2003-12-15 00:00:00
>6 2003-12-15 00:00:00
>7 2003-12-15 00:00:00
>8 2003-12-15 00:00:00
>9 2003-12-15 00:00:00
>
>??Which is the mistake?.
Well that has 9 rows the same as Hoja1 in Pruebas4.xls, so I do not 
understand your question 1.

What exactly do you mean by question 2? Do you mean it is not reading the 
hours column (Hora) or do you mean it takes a long time? It looks to me as 
though column one, despite its label of day (D??a) actually contains some 
much more complex Excel format.


Michael Dewey
m.dewey at iop.kcl.ac.uk

Playing with randomForest, samples run fine. But on real data, no go.

Here's the setup: OS X, same behavior whether I'm using R-Aqua 1.8.1 or
the
Fink compile-of-my-own with X-11, R version 1.8.1.

This is on OS X 10.3 (aka "Panther"), G4 800Mhz with 512M physical
RAM.

I have not altered the Startup options of R.

Data set is read in from a text file with "read.table", and has 46
variables
and 1,855 cases. Trying the following:

The DV is categorical, 0 or 1. Most of the IV's are either continuous, or
correctly read in as factors. The largest factor has 30 levels.... Only the
DV seems to need identifying as a factor to force class trees over
regresssion:
>Mydata$V46<-as.factor(Mydata$V46)
>Myforest.rf<-randomForest(V46~.,data=Mydata,ntrees=100,mtry=7,proximities=FALSE
, importance=FALSE)

5 hours later, R.bin was still taking up 75% of my processor.  When I've
tried this with larger data, I get errors referring to the buffer (sorry,
not in front of me right now).

Any ideas on this? The data don't seem horrifically large. Seems like there
are a few options for setting memory size, but I'm  not sure which of them
to try tweaking, or if that's even the issue.