With Jupyter notebooks now being opened in a browser window (as of Canopy 1.6.1), you are now able to edit and run notebooks in Julia, Python, or R kernels. Below are the steps needed to install an R kernel to be used in Canopy's Jupyter installation.
For all platforms, R must be installed. Mirrors to R installers can be found here:
https://cran.r-project.org/mirrors.html
Instructions for installing can be found here:
http://irkernel.github.io/installation/
After installation you will be able to create a new notebook from within the Jupyter browser window in the R kernel. You will also be chance between a Python and R kernels using the Kernel --> Change Kernel menu.
Platform specific caveats:
Mac (tested on OSX 10.11):
All of the commands below are executed from a Canopy Terminal (accessible from the Tools menu in Canopy).
From a Canopy terminal, install zmq using the instructions from the "Source" tab in the above installation instructions.
Homebrew can be installed by running:
ruby -e "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install)"To use Homebrew on OSX 10.11, you must first change the permissions on /usr/local by running:
sudo chown -R $(whoami):admin /usr/local
Macports can be installed by downloading installers from:
https://www.macports.org/install.php
After installing ZMQ, open an R Console in the Canopy Terminal by running: R
In the R Console, run the following (as a single, long command):
install.packages(c('rzmq','repr','IRkernel','IRdisplay'), repos = c('http://irkernel.github.io/', getOption('repos')), type = 'source')
If this command fails to install (which it likely will), certain directories and symbolic file links will need to be created.
Take this error message for example:
Warning message: In doTryCatch(return(expr), name, parentenv, handler): unable to load shared object '/Library/Frameworks/R.framework/Resources/modules/R_X11.so':dlopen(/Library/Frameworks/R.framework/Resources/modules/R_X11.so, 6): Library not loaded: /opt/X11/lib/libICE.6.dylib
Referenced from: /Library/Frameworks/R.framework/Resources/modules/R_X11.so
Reason: image not found
This indicates that R is looking for the file "libICE.6.dylib" in the directory "/opt/X11/lib/"
Quit the R Console by running: q()
It is likely that such a file exists on your computer, but in another location. In this instance, you can find the location of "libICE.6.dylib" by running:
sudo find / -name libICE.6.dylib
On my system, libICE.6.dylib was found in "/usr/X11/lib". Once this file is found, you can create a symbolic link to this file in the location that R is searching for it. For libICE.6.dylib, this can be accomplished by running:
sudo mkdir -p /opt/X11/lib/
sudo ln -s /usr/X11/lib/libICE.6.dylib /opt/X11/lib/libICE.6.dylib
After creating this link, run the "install.packages..." command in R again. If the install fails again with a similar error, repeat the above procedure to create a symbolic link to the file indicated by the new error message.
After the installation succeeds, run the following in the R Console to install the IRkernel to Jupyter:
IRkernel::installspec()
Installation should now be complete and the R kernel available in Jupyter notebooks.
Comments