[R] Error: vector memory exhausted (limit reached?)

[Ivan Krylov]

Hello Olivia!

On Tue, 23 Nov 2021 20:26:07 +0000
Olivia Keefer <okeefer using monigle.com> wrote:

> We have been able to run other regressions, but when we go to run a
> certain set of variables, we both are getting this message each time
> we try.

It would help if you provided more information about how exactly you
run this regression (e.g. "I'm running `reg <- lm(y ~ X)`") and what
are the sizes of the matrices involved (e.g. "dim(X) is c(10000,
1000)"). A reproducible example we could run ("When I copy&paste the
following code into my R session, I always get the error") would be
ideal, but it's not always possible.

I'm assuming you run relaimpo::boot.relimp, but with what kind of data?

(Perhaps you had attached an example, but the mailing list ate the
attachment. It eats most kinds of attachments:
https://www.r-project.org/posting-guide.html)

Adding more RAM to the computer running R would probably help. If the
code makes excessive copies of data (by taking a large variable,
changing something in it and keeping it referenced, maybe indirectly
via an environment), perhaps it could be restructured to avoid doing
that. I'm afraid it's hard to be more specific without more information.

> 	[[alternative HTML version deleted]]

P.S. Your message reached us almost intact, but in order to be sure
that we'd be seeing your messages exactly as you compose them, please
compose them in plain text, not HTML.

-- 
Best regards,
Ivan