Save and load R data

Save and load data in .RData or .rds file format

( see → read & write for saving data as text files )

save()   

- to save one or several variables (including their variable names)

- useful to save all variables of a data analysis project

saveRDS()

- to save the pure data content of a single variable (without keeping the variable name)

- useful to save different results, one by one, and later combine the results by importing the individual data in a controlled way

Save multiple variables

# save multiple variables (objects)

x = 23

y = 14

save(x, y, file="data_xy.RData")

# load data into new R environment "ev" to avoid overwriting variables having the same name

x = 44

load("data_xy.RData", envir = (ev = new.env()) ) 

ev$x

 [1] 23

ev$y

 [1] 14

x

 [1] 44

# load data (without new environment): overwrites all variables in R that have the same name

 x = 44 # will be overwritten by identical variable "x" in data_xy.RData

load("data_xy.RData")

x

  [1] 23

y

  [1] 14

Save all variables

# save all variables (data objects) from the current R working session

save.image(file = "data_all.RData")

 q()   # exit R

# load the saved R variables (same as above)

load("data_all.RData")

# list all variables present in R

ls()

  [1]  "x"  "y"

Save data content of a single variable (RDS)

While load() by default re-loads the variables into the global R environment (and potentially overwriting identical variable names, if not protected by defining a new environment), readRDS() can be used to better control individual variable names of imported data.

Use saveRDS to save a single variable and to re-load the content into any new variable name.

# save content of variable "x"

x = 23

saveRDS(x, "data_x.rds")

# load data (of former variable "x") into a new variable "a"

a = readRDS("data_x.rds")

ls()

  [1] "a"  "x"

 a   # new variable

  [1] 23

 x      # original variable

  [1] 23   

Get list of saved R data files

# list all .RData files in current directory (path=".")

list.files(path = '.', pattern=".RData")

   [1]  "data_all.RData"  "data_xy.RData"