How to share data frames between scripts in R

Question

I've got multiple R scripts; one that cleans my original data and produces a tidy data frame, and several others that performs functions on that data frame.

When I wrote them, the data frame produced by the first script was in my RStudio environment and the other scripts referenced the resulting data frame without trouble.

Now that I'm trying to run them from the console, the data frame produced by the first script isn't reference-able for the others.

What's the best way to share a data frame between scripts?

It sounds like the data frame had been stored in the R environment. Works in RStudio, but the console starts a new, empty environment for each script, so there’s no data to access. Save your data frame to a file and then load that file in subsequent scripts. — jdobres
– jdobres, Commented Mar 17, 2019 at 17:50
I would write a csv at the end of the first script and and read it as csv in the second. The fastest way to do that is fwrite and fread from package data.table — Pierre Lapointe
– Pierre Lapointe, Commented Mar 17, 2019 at 18:00
I would recommend saveRDS() and readCSV(). Faster, smaller (compressed), no need serialize and de-serialize to text. — Dirk is no longer here
– Dirk is no longer here, Commented Mar 17, 2019 at 18:05

Diego Queiroz · Accepted Answer · 2019-03-17 18:18:18Z

2

You could try using the commands save.image() and load() to save your workspace to a file and then load it onto your console environment as it's likely that your console instance and RStudio each have their own independent environments.

Doing this way, you would have access to all objects that the previous scripts executed. However, if you're only interested in the generated data, you could save your data.frame using save() and open it using load().

As mentioned by @Dirk Eddelbuettel, there are also plenty good functions to save single variables like saveRDS() and readRDS() (which provides a better serialization than save()) and write.csv() and read.csv().

edited Mar 17, 2019 at 18:18

answered Mar 17, 2019 at 18:01

Diego Queiroz

3948 silver badges11 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Dirk is no longer here Over a year ago

I would recommend saveRDS() and readCSV() for single variables. Easier to handle than the images, and more focussed.

Diego Queiroz Over a year ago

Indeed, I had forgotten about both, will update answer accordingly.

Collectives™ on Stack Overflow

How to share data frames between scripts in R

1 Answer 1

2 Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Related