CODEWITHSUNDEEP
rrasterrds
Cruz
Cruz

Reputation: 65

Reading in multiple .rds files and creating one object

I have seen other examples of this on stackexchange but cannot seem to adapt them to my code.

Problem: I have a folder of .rds files that I would like to read into R, then stack all the files together so I can take the mean and standard deviation. All the .rds files are 'Formal class RasterLayer' when brought into R, if that is pertinent.

Example code:

 # file path to folder where .rds files are stored
   path = "~/Predictions/"
   # create place to store files
   stack <-""
   # create vector of all .rds files in folder
   pred.dates <- dir(path, pattern =".rds")
   # loop to bring in each .rds file
   for(i in 1:length(pred.dates)){
   file <- readRDS(file.names[i],".rds")
   stack <- rbind(stack, file)
   }

   # take mean of all .rds files stacked together and plot 
   pred_mean <- mean(stack, na.rm=T)
   plot(pred_mean)

   # take sd of all .rds files stacked together and plot 
   pred_sd <- sd(stack, na.rm = T)
   plot(pred_sd)

However, it returns the error:

Error in gzfile(file, "rb") : cannot open the connection
In addition: Warning message:
In gzfile(file, "rb") :
  cannot open compressed file 'Pred_.rds', probable reason 'No such file or directory'

Seems like this should be straightforward, but perhaps I'm not using the correct function. Thanks!

Upvotes: 1

Views: 4576

Answers (2)

Robert Hijmans
Robert Hijmans

Reputation: 47156

I think the problem with the solution by F. Privé is that they use rbind instead of stack. I would suggest doing 

library(raster)
files <- list.files(path = path, pattern = "\\.rds$", full.names = TRUE)
r <- lapply(files, readRDS)
s <- stack(r)

Upvotes: 5

F. Priv&#233;
F. Priv&#233;

Reputation: 11728

You can do:

files <- list.files(path = path, pattern = "\\.rds$", full.names = TRUE)
stack <- do.call("rbind", lapply(files, readRDS))

Upvotes: 2

Related Questions