I am getting the following error when I try to load the
reticulatepackage I built in a local conda environment:
Error: package or namespace load failed for ‘reticulate’ in dyn.load(file, DLLpath = DLLpath, ...): unable to load shared object '~/Gits/nlp_r_in_python_testing/nlp_py_env/lib/R/library/reticulate/libs/reticulate.so': /lib64/libstdc++.so.6: version `GLIBCXX_3.4.20' not found (required by ~/Gits/nlp_r_in_python_testing/nlp_py_env/lib/R/library/reticulate/libs/reticulate.so)
The RStudio instance just seems to keep defaulting to the server base. Even happens with the library paths:
.libPaths(c("./nlp_py_env/lib/R/library", "./nlp-modeling/renv/library/R-4.2/x86_64-pc-linux-gnu")) > .libPaths()  "./nlp_r_in_python_testing/nlp_py_env/lib/R/library"  "/opt/R/4.2.1/lib/R/library"
The same error described in the posts here:
And in this previous topic:
The workaround to this (proposed in the first link) seems fairly straightforward.
However, I don't quite understand how to call the RStudio IDE from my browser.
Specifically, if I should use the format :<port#>, or :<port#>, as neither seem to have worked.
Any guidance or suggestions would be greatly appreciated!