You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've tried installing mxnorm on a fresh environment. Here is what my setup was so that you have a better idea.
I'm using a freshly installed Linux container installed on a pixelbook (which is a docker container of a fairly bare bones Debian, I can provide the specific version if necessary). I installed miniconda, and made a separate environment for R using the following command (also installed mamba, because mamba is amazing):
mamba create -n R -c conda-forge r-base r-essentials
Inside the R environment, I followed the instructions to install mxnorm install.packages("mxnorm")
On my end, it gave an error that lme4 and nloptr needed to be installed. I used the following command to get those installed through conda:
mamba install -c conda-forge r-lme4 r-nloptr
Now, my macbook already has a well established R environment setup, so I doubt I would have any problems installing it there, I just wanted to see how it would work on a fresh environment.
I just wanted to bring those dependencies to your attention. I think it should be a pretty easy thing to either include in the README or on CRAN setup side of things.
I also wanted to mention, in case you haven't considered it already, to at some point make mxnorm available through conda. It seems like it wouldn't be too difficult, and most of the dependencies should already be available through there.
The text was updated successfully, but these errors were encountered:
After doing some digging, it looks like both lme4 (and nloptr since it's one of the dependencies) are not included in the r-essentials bundle from conda. lme4 is listed as an "Import" in the DESCRIPTION file, so I would expect that R would either prompt you to install those packages or go ahead and install them by default when installing mxnorm.
Can you share more info about the error you encountered? Was it the R environment flagging that you needed to install those packages, and instead of installing them within the R environment you installed them via conda?
So I was able to setup a similar Docker container and re-create this issue, which turned out to be a missing dependency for CMake installed on the machine. I've added this to the Dependencies section of the README in case users hit the same error.
I've tried installing mxnorm on a fresh environment. Here is what my setup was so that you have a better idea.
I'm using a freshly installed Linux container installed on a pixelbook (which is a docker container of a fairly bare bones Debian, I can provide the specific version if necessary). I installed miniconda, and made a separate environment for R using the following command (also installed mamba, because mamba is amazing):
Inside the R environment, I followed the instructions to install mxnorm
install.packages("mxnorm")
On my end, it gave an error that
lme4
andnloptr
needed to be installed. I used the following command to get those installed through conda:Now, my macbook already has a well established R environment setup, so I doubt I would have any problems installing it there, I just wanted to see how it would work on a fresh environment.
I just wanted to bring those dependencies to your attention. I think it should be a pretty easy thing to either include in the README or on CRAN setup side of things.
I also wanted to mention, in case you haven't considered it already, to at some point make mxnorm available through conda. It seems like it wouldn't be too difficult, and most of the dependencies should already be available through there.
The text was updated successfully, but these errors were encountered: