[R] save() and load(): *prefer* saveRDS() and readRDS()

[AbouEl-Makarim Aboueissa]

Hi Martin: good morning

Thank you very much for your detailed explanation. Got it now.


With many thanks
Abou


On Tue, Sep 26, 2023, 3:24 AM Martin Maechler <maechler using stat.math.ethz.ch>
wrote:

> >>>>> Jeff Newmiller via R-help
> >>>>>     on Mon, 25 Sep 2023 18:46:02 -0700 writes:
>
>     > You never created any object in R called irisdataTest. Objects in
> the global environment have names that are unrelated to the names of files
> on disk.
>     > The load function modifies an environment to create a variable named
> as it was named in the environment from which it was saved. Thus, you
> cannot simply load an object that was saved with one name into an object
> named something else. It is possible to create a new environment to put the
> loaded objects into, but I wouldn't recommend trying to explain how to do
> that to a beginner.  Rather, I would instead recommend using saveRDS and
> readRDS instead to save/load exactly one object at a time without storing
> the object name.
>
>     > saveRDS( mtcars, "my_mtcars.rds" )
>     > new_obj <- readRDS( "my_mtcars.rds" )
>
>     > I would also guide them to never save their environment when
> prompted by R... the .RData file this creates will remember mistakes made
> in previous sessions making troubleshooting very difficult later. Instead
> they should focus on making a top-to-bottom script that has all their
> analysis steps so they can start from scratch.
>
> Yes!
>
> And just re-iterating what Jeff mentioned above:
> Notably when teaching, the use of  saveRDS()  and  readRDS()
> should be emphasized as safer / self-documenting, ...
> for the case where there's just one object to save/load.
>
> *and* you can always put several objects into a list and
> saveRDS() / readRDS() that.
>
> Note: Our pkg {sfsmisc} nowadays contains a nice utility function
> list_()  ==> help page online e.g. here
>
> https://search.r-project.org/CRAN/refmans/sfsmisc/html/list_named.html
>
> which comes were handy when you want to easily create a *named*
> list from a bunch of objects, as e.g. above to be able to nicely use
>
>   saveRDS(list_(obj1, obj2, table3, grob4, data5),
>           file = "allthings.rds")
>
> The cute utility is very simply defined as
>
> ##' list_(a, b, cc)  creates a *named* list  using the actual arguments'
> names
> list_ <- function(...) `names<-`(list(...), vapply(sys.call()[-1L],
> as.character, ""))
>
>
>     > On September 25, 2023 6:23:01 PM PDT, AbouEl-Makarim Aboueissa <
> abouelmakarim1962 using gmail.com> wrote:
>     >> Dear ALL:
>     >>
>     >> I am teaching statistical packages class this semester, in R
> programing I
>     >> am trying to explain the use of save() and load() with an example
> using the
>     >> iris data. It seems that the save() function works, BUT when I
> tried to
>     >> load the data back to R, it seems that there is a problem(s), I
> could not
>     >> figure out what went wrong.
>     >>
>     >> Any help would be highly appreciated.
>     >>
>     >>
>     >> I saved the iris data in my computer in the text format,
> "iris.with.head.txt
>     >> ".
>     >>
>     >> Here are my R codes:
>     >>
>     >>> irisdata<-read.table("G:/iris.with.head.txt", header=T)
>     >>>
>     >>> head(irisdata)
>     >> Sepal.Length Sepal.Width Petal.Length Petal.Width Species
>     >> 1          5.1         3.5          1.4         0.2  setosa
>     >> 2          4.9         3.0          1.4         0.2  setosa
>     >> 3          4.7         3.2          1.3         0.2  setosa
>     >> 4          4.6         3.1          1.5         0.2  setosa
>     >> 5          5.0         3.6          1.4         0.2  setosa
>     >> 6          5.4         3.9          1.7         0.4  setosa
>     >>
>     >>
>     >>
>     >> *##### saving the data as an .rda*
>     >>
>     >> save(irisdata,file="G:/irisdataTest.rda")
>     >>
>     >> *##### load the data back to R*
>     >>
>     >> load(file="G:/irisdataTest.rda")
>     >>
>     >>
>     >>> head(irisdataTest)
>     >> Error in head(irisdataTest) : object 'irisdataTest' not found
>     >>
>     >>> irisdataTest
>     >> Error: object 'irisdataTest' not found
>     >>
>     >>
>     >>
>     >> with many thanks
>     >> abou
>     >> ______________________
>     >>
>     >>
>     >> *AbouEl-Makarim Aboueissa, PhD*
>     >>
>     >> *Professor, Mathematics and Statistics*
>     >> *Graduate Coordinator*
>     >>
>     >> *Department of Mathematics and Statistics*
>     >> *University of Southern Maine*
>     >>
>     >> [[alternative HTML version deleted]]
>     >>
>     >> ______________________________________________
>     >> R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
>     >> 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.
>
>     > --
>     > Sent from my phone. Please excuse my brevity.
>
>     > ______________________________________________
>     > R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
>     > 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.
>

	[[alternative HTML version deleted]]