Skip to content

Reading text files in Spatstat

9 messages · Taylor, RB, Don MacQueen, ONKELINX, Thierry

Original
Taylor, RB
# 
Hello all,

I have x,y data with an associated mark. Is there a simple way of 
getting Spatstat to read this data? I have tried putting the x, y data 
in separate columns in Excel and ignoring the marks, saving it as a 
tab-delim text file (called "test"). However the as.ppp function fails 
to convert it into a ppp object.


 > test <- read.table("C:/temp/Rtest.txt")
 > as.ppp(test)
Error in as.ppp.default(test) : Can't interpret X as a point pattern


Perhaps someone can answer my simple question and tell me what I am 
doing wrong?

Thanks,
Rob
--
Don MacQueen
# 
The help page for as.ppp() explains that in a case like yours (or 
what I assume is yours from your description), you also need to 
supply the W argument.

Try looking at the help page for the ppp() function.

-Don
At 1:13 PM +0100 7/3/07, Taylor, RB wrote:

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Taylor, RB
    

    

      
      #
    

  


  

      
Right, I have supplied the W argument now. Still no joy though....

 > as.ppp(test,c(0,10,0,10),fatal=TRUE)
Error in as.ppp.default(test, c(0, 10, 0, 10), fatal = TRUE) :
         Can't interpret X as a point pattern

Any ideas anyone?

Rob

        
Don MacQueen wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      ONKELINX, Thierry
    

    

      
      #
    

  


  

      
The problem is probably in "test". Try to give us a (small) sample of
test so we can reproduce your problem. Otherwise it will be hard to
discover the problem.

Thierry

------------------------------------------------------------------------
----
ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature
and Forest
Cel biometrie, methodologie en kwaliteitszorg / Section biometrics,
methodology and quality assurance
Gaverstraat 4
9500 Geraardsbergen
Belgium
tel. + 32 54/436 185
Thierry.Onkelinx at inbo.be
www.inbo.be 

Do not put your faith in what statistics say until you have carefully
considered what they do not say.  ~William W. Watt
A statistical analysis, properly conducted, is a delicate dissection of
uncertainties, a surgery of suppositions. ~M.J.Moroney

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Taylor, RB
    

    

      
      #
    

  


  

      
Ok, "test" is just a small test data set of x, y coords.

3.45	5.34
2.78	5.67
3.89	1.345
2.34	3.78
3.45	5.34
2.78	5.67
3.89	1.345
2.34	3.78


You could make it easily in Excel or a similar program. x coords are in 
the first column, y coords are in the 2nd column. It is saved as a tab 
delimited text file called "test".


Cheers,
Rob

        
ONKELINX, Thierry wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        
      

    

    

      
      

        


  

        

      ONKELINX, Thierry
    

    

      
      #
    

  


  

      
I've found your problem. Test must be a matrix and you probably have a
data.frame.
I hope you see the importance of a reproducibel example of your code.
You MUST define the data too.


This works.

            

      
      
      
    

        
planar point pattern: 10 points 
window: rectangle = [0, 10] x [0, 10] units 

This doesn't work.

            

      
      
      
    

        
Error in as.ppp.default(test, c(0, 10, 0, 10), fatal = TRUE) : 
        Can't interpret X as a point pattern


Cheers,

Thierry

------------------------------------------------------------------------
----
ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature
and Forest
Cel biometrie, methodologie en kwaliteitszorg / Section biometrics,
methodology and quality assurance
Gaverstraat 4
9500 Geraardsbergen
Belgium
tel. + 32 54/436 185
Thierry.Onkelinx at inbo.be
www.inbo.be 

Do not put your faith in what statistics say until you have carefully
considered what they do not say.  ~William W. Watt
A statistical analysis, properly conducted, is a delicate dissection of
uncertainties, a surgery of suppositions. ~M.J.Moroney

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Taylor, RB
    

    

      
      #
    

  


  

      
ah, thank you Thierry. As a fairly new R user, this is most helpful :)

Cheers,
Rob

        
ONKELINX, Thierry wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Taylor, RB
    

    

      
      #
    

  


  

      
Just to finally finish off the thread, here is the code I used.

# read in the text data into a dataframe. "Rtest" is just x,y coords
 > test <- read.table("C:/temp/Rtest.txt")

# convert the data frame to a numeric matrix
 > test.matrix <- data.matrix(test)

# coerce the data into a ppp object in Spatstat
 > as.ppp(test.matrix,c(0,10,0,10),fatal=TRUE)
planar point pattern: 8 points
window: rectangle = [0, 10] x [0, 10] units

        
Taylor, RB wrote: