Closed GaryHilares closed 1 year ago
Thanks for the heads up @GaryHilares . I'm planning to do a round of edits on the textbook/worksheets this summer at some point, so I'll make sure to revisit this when I do.
FYI: I've been working on the textbook/worksheets/etc for the past few weeks. Due to various dependencies we have (most notably, nbgrader and autotest), we can't update to JLab 4+ yet. It looks like nbgrader should be compatible shortly, but autotest won't likely be compatible until this PR is merged or at least brought up to speed with nbgrader 0.9.x.
I'll leave this open until that is resolved.
In chapter 13 is recommended to install the R packages in the
environment.yml
file using the commandconda env update --file https://raw.githubusercontent.com/UBC-DSCI/data-science-a-first-intro-worksheets/main/environment.yml
. However, when I tested this method it did not work with the current Jupyter notebook version (it threw an error message Jupyter notebook could not reach the IRkernel). To get a working instance I had to install the latest packages using theconda install -c conda-forge -y r-base r-cowplot r-ggally r-gridextra r-irkernel r-kknn r-rpostgres r-rsqlite r-scales r-testthat r-tidymodels r-tidyverse r-tinytex unixodbc
command.In Binder this does not look like an issue because Binder uses JupyterLab v3.1.12. However, as of 2023-06-06, the instructions for local installation use JupyterLab v4.0.1. I would recommend updating the R packages' versions for compatibility with the latest Jupyter Lab version. Another possible solution would be to add instructions to instructions to install JupyterLab v3.1.12 rather than v4.0.1. I personally prefer the first approach.
Tested in Ubuntu 22.04.1 LTS.