AJF
May 25, 2020, 2:17pm
1
Hi all,
I have a question about exposing user-defined R functions to Python code.
I have created R functions in my utils.R
file. Currently, in my R Markdown document, I just run
source("utils.R")
and use the functions. The functions generally bring in relatively standard data types (data frames, numbers, dates).
I have some colleagues who use Python who want to use the functions, and I would (ideally) not like to rewrite them in Python.
Is there a way to use the reticulate
package to turn them into files to be run from python directly (not from python through R).
Thanks!
There is the rpy2 interface but I could not get it working with R 4.0.0, Windows 10 and Jupyter.
Maybe you have more luck with a different configuration.
1 Like
AJF
May 25, 2020, 8:54pm
3
Thanks -- I've found rpy2 finicky in the past, so I had hoped to avoid it, and just rely on reticulate's magic - but this might have to do.
Today I saw that version 3.3.3 of rpy2 solved a problem that I had with an earlier version. So maybe worth a second try?
system
Closed
June 3, 2020, 11:58pm
5
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.