Save objects resulting from a for loop in R

Question

I am trying to save different objects resulting from loop. I am using the funcion raster to charge multiple images in different directories,the loop generetes these objects:

b : it generates the different directories where I have my images

[1] "C:/Users/franc/Documents/Fran/Tesis/Sin sincronizar/Imagenes Landsat/1985/Invierno/band3.tif"

[1] "C:/Users/franc/Documents/Fran/Tesis/Sin sincronizar/Imagenes Landsat/1986/Invierno/band3.tif"

[1] "C:/Users/franc/Documents/Fran/Tesis/Sin sincronizar/Imagenes Landsat/2004/Invierno/band3.tif"

name: the different names that I want to use to save the outputs

[1] "1985Banda3"

[1] "1986Banda3"

[1] "2004Banda3"

Then I want to use the directory b to charge the images, and save each one in each value of name

Here is my code:

library(raster)
a<-c(1985,1986,2004)
i<-1

while(i<=(length(a)))
{
  b<-paste("C:/Users/franc/Documents/Fran/Tesis/Sin sincronizar/Imagenes Landsat/",a[[i]],
       "/Invierno/band3.tif", sep = '')
  name<-(paste(a[[i]],"Banda3", sep =''))
  name<- raster(b)
  i<-i+1

}

I want to generete this

1985Banda3 <- raster( "C:/Users/franc/Documents/Fran/Tesis/Sin sincronizar/Imagenes Landsat/1985/Invierno/band3.tif")

1986Banda3 <- raster( "C:/Users/franc/Documents/Fran/Tesis/Sin sincronizar/Imagenes Landsat/1986/Invierno/band3.tif")

2004Banda3 <- raster( "C:/Users/franc/Documents/Fran/Tesis/Sin sincronizar/Imagenes Landsat/2004/Invierno/band3.tif")

lookup `assign()`. You can use it to assign values within the loop pretty easily — morgan121, Dec 13 '18 at 02:33

score 2 · Answer 1 · answered Dec 13 '18 at 02:34

If you really want seperate variables, you can use assign, which gives you the full call assign(name, raster(b)).

However, I doubt whether it's really useful to have different variables for many objects that you want to handle the same way (I'm assuming your real data has more than 3 items). Because in a next step you want to do something with them, and you need those names again...

In my experience, working with a (named) list is a lot easier, which you can get this way:

library(raster)
a<-c(1985,1986,2004)

Bandas <- lapply(a, function(name) {
    b <- paste("C:/Users/franc/Documents/Fran/Tesis/Sin sincronizar/Imagenes Landsat/",name)
    raster(b)
})
names(Bandas) <- paste("Banda3_", a, sep ='')

Now you have a list, called Bandas, with 3 named elements.
To access e.g. the 2004 one, you can use Bandas$Banda3_2004, or name <- 'Bandar_2004'; Banda[['name']]

But not only do you have a cleaner workspace, it's also easier to do things to your objects:

ModifiedBandas <- lapply(Bandas, function(ban) {
    _do something with an individual file_
})

for (ban in Bandas) {
   print(ban)
}

And I've changed the names a bit, because from the documentation:

Identifiers consist of a sequence of letters, digits, the period (‘.’) and the underscore. They must not start with a digit or an underscore, or with a period followed by a digit.

You can circumvent it (and assign complies with other names), but that means you have to use backticks (" ` ") around the name to refer to it.

Save objects resulting from a for loop in R

1 Answers1