[R] Loop to extract from variables in the workspace

Wed Apr 23 11:37:24 CEST 2014

Dear Frede,

Sorry for the late reply, yesterday I was out of office.

Thanks so much for your code! It has just compacted in a few lines what 
I had written in several 'paragraphs'!

To be honest, I'm not familiar to work with lists in R. I think one of 
its main advantages is the reduced size of the outputs.

Thanks again and have a nice day,

Bea

On 22.04.2014 07:04, Frede Aakmann Tøgersen wrote:
> Well, sticking to your loop that you seem comfortable with try this:
>
> ## list to hold dataframes
> extracList <- vector("list", length(1981:2010))
>
> cnt <- 0
> for(i in 1981:2010){
> 	cnt <- cnt + 1
> 	rasterObj <-get( c(paste("RR_", i, "_1", sep="")))
>        	extractList[[cnt]]<- extract(rasterObj, coords, df=T)}
>   }
>
> See ?get which is kind of the opposite of assign.
>
> To coerce the list of dataframes into one dataframe several methods are available:
>
> do.call("rbind", extractList)
>
> plyr::rbind.fill(extractList)
>
> plyr::ldply(extractList, data.frame)
>
> as.data.frame(data.table::rbindlist(extractList))
>
> This was taken from http://stackoverflow.com/questions/2851327/converting-a-list-of-data-frames-into-one-data-frame-in-r where you get get some pros and cons on the methods.
>
> The fastest way will probably be to create a huge dataframe and during the loop calculated the row indices of that dataframe for each raster into the result from extract should be put.
>
>
>
> Yours sincerely / Med venlig hilsen
>
>
> Frede Aakmann Tøgersen
> Specialist, M.Sc., Ph.D.
> Plant Performance & Modeling
>
> Technology & Service Solutions
> T +45 9730 5135
> M +45 2547 6050
> frtog at vestas.com
> http://www.vestas.com
>
> Company reg. name: Vestas Wind Systems A/S
> This e-mail is subject to our e-mail disclaimer statement.
> Please refer to www.vestas.com/legal/notice
> If you have received this e-mail in error please contact the sender.
>
>
>> -----Original Message-----
>> From: Beatriz R. Gonzalez Dominguez [mailto:aguitatierra at hotmail.com]
>> Sent: 21. april 2014 16:27
>> To: Frede Aakmann Tøgersen; r-help at r-project.org
>> Subject: Re: [R] Loop to extract from variables in the workspace
>>
>> Hi Frede,
>>
>> Many thanks for your reply.
>> 1. The first argument in extract is a Formal class RasterLayer in the
>> Workspace (e.g RR_1981_1 ).
>>
>> 2. I created an intermediate name to hold the result fromthe extract
>> function because I'd like to create several dataframes with the output
>> of the iterative (loop) extraction. I'd like to get the same result as
>> when I do:
>>
>> PE.coords_01_1981 <- extract(RR_1981_1, coords, df=T)
>> PE.coords_01_1982 <- extract(RR_1982_1, coords, df=T)
>> PE.coords_01_1983 <- extract(RR_1983_1, coords, df=T)
>> PE.coords_01_1984 <- extract(RR_1984_1, coords, df=T)
>> [... this works no problem]
>>
>> 3. 'coords' is a SpatialPointsDataFrame.
>>
>> 4. I used assign in the loop becuase I thought it was the way forward to
>> create new variables out of it. Isn't it?
>>
>> What I'd like to do is to use coordinate points ('coords') to extract
>> raster pixel values (eg. 'RR_1981_1') on which the points are overlying.
>> Then I'd like to build a bigger data frame including the data from all
>> the outputs (i.e. PE.coords_01_1981, PE.coords_01_1982).
>>
>> Hope to have explained myself properly. Please let me know if anything
>> else should be clarified.
>>
>> Best wishes,
>>
>> Bea
>>
>>
>> On 21/04/2014 15:17, Frede Aakmann Tøgersen wrote:
>>> Hi Beatriz
>>>
>>> Did you read the help for extract{raster} carefully?
>>>
>>> Several things can be wrong.
>>>
>>> 1) First argument to extract is not a file name but a raster object.
>>> 2) In the loop you name an object extract as an intermediate name to hold
>> the result from the extract function. Do you think there could be a name
>> clash? R is clever but perhaps not clever enough.
>>> 3) coords are of the right class (see ?extract).
>>> 4) assign can be useful from time to time. But in a loop?
>>>
>>> I think the things you are doing are some intermediate results that needs
>> more processing. Do you think this is the right way to do that. For instance
>> instead of storing the immediate result as separate objects why not store
>> those in a list.
>>> Perhaps if you tell us what you would like to do  overall, i.e. from first to
>> last, then we will be able to help you to become more efficient.
>>>
>>> Yours sincerely / Med venlig hilsen
>>>
>>>
>>> Frede Aakmann Tøgersen
>>> Specialist, M.Sc., Ph.D.
>>> Plant Performance & Modeling
>>>
>>> Technology & Service Solutions
>>> T +45 9730 5135
>>> M +45 2547 6050
>>> frtog at vestas.com
>>> http://www.vestas.com
>>>
>>> Company reg. name: Vestas Wind Systems A/S
>>> This e-mail is subject to our e-mail disclaimer statement.
>>> Please refer to www.vestas.com/legal/notice
>>> If you have received this e-mail in error please contact the sender.
>>>
>>>
>>>> -----Original Message-----
>>>> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
>> project.org]
>>>> On Behalf Of Beatriz R. Gonzalez Dominguez
>>>> Sent: 21. april 2014 14:53
>>>> To: r-help at r-project.org
>>>> Subject: [R] Loop to extract from variables in the workspace
>>>>
>>>> Dear all,
>>>>
>>>> I'm starting to work with loops and I'm stucked on something.
>>>> I've been searching and trying different possibilities but I don't get
>>>> to the solution.
>>>> I'd be very grateful if you could share any ideas that you think may help.
>>>>
>>>> library("raster")
>>>>
>>>> # All my variables are in the workspace
>>>>
>>>> # This is what I'd like to obtain, but with a loop (I'm working with
>>>> several years and variables).
>>>> PE.coords_01_1981 <- extract(RR_1981_1, coords, df=T)
>>>> PE.coords_01_1982 <- extract(RR_1982_1, coords, df=T)
>>>> PE.coords_01_1983 <- extract(RR_1983_1, coords, df=T)
>>>> PE.coords_01_1984 <- extract(RR_1984_1, coords, df=T)
>>>> PE.coords_01_1985 <- extract(RR_1985_1, coords, df=T)
>>>> PE.coords_01_1986 <- extract(RR_1986_1, coords, df=T)
>>>> PE.coords_01_1987 <- extract(RR_1987_1, coords, df=T)
>>>> PE.coords_01_1988 <- extract(RR_1988_1, coords, df=T)
>>>> PE.coords_01_1989 <- extract(RR_1989_1, coords, df=T)
>>>> PE.coords_01_1990 <- extract(RR_1990_1, coords, df=T)
>>>>
>>>>
>>>> # This is one of the things I've tried.
>>>>
>>>> for(i in 1981:2010){
>>>> file <- c(paste("RR_", i, "_1", sep=""))
>>>> extract <- extract(file, coords, df=T)}
>>>> names.a <- paste("PE.coords_01_", i, sep="")
>>>> assign(names.a, value=extract)
>>>> }
>>>>
>>>> # I get the following error.
>>>> Error in (function (classes, fdef, mtable) :
>>>> unable to find an inherited method for function 'extract'
>>>> for signature '"character", "SpatialPointsDataFrame"'
>>>> # I think the error must be something when I'm defining 'file'
>>>>
>>>> Thanks a lot for any help!
>>>>
>>>> ______________________________________________
>>>> 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.
>
>