How to connect multiple versions of R to jupyterlab

Note : The page is based on ChatGPT response but edited and tested by me.

To add additional versions of R to JupyterLab, you need to install and register each R version as a separate Jupyter kernel using IRkernel. This allows you to switch between different R versions within JupyterLab.

---

Steps to Add Multiple R Versions

0. Set Up JupyterLab Environment

Make sure you install JupyterLab in advance and set or activate the environment where you can launch JupyterLab.

# UTK ISAAG-NG
module load anaconda3/2023.09
# or other environment settings for your jupyter installation

1. Install Mutiple R Versions

Make sure you have all the R versions you need installed on your system. You can install them side-by-side and access them via their respective paths.

For example:

• R-3.6.3 might be installed as /usr/local/bin/R-3.6.3
• R-4.2.0 might be installed as /usr/local/bin/R-4.2.0

---

2. Install IRkernel for Each R Version

Launch each R version individually and install the IRkernel package.

For R Version 3.6.3

Run the following in the R console:

install.packages("IRkernel")
IRkernel::installspec(name = 'ir363', displayname = 'R 3.6.3', user = FALSE)

• installspec function of IRkernel creates configuration files which jupyter reads to set up kernels.
• name: Unique kernel name for Jupyter.
• displayname: The name that will appear in JupyterLab.
• user = TRUE if you do not have root privilege, which will be the most common case.

For R Version 4.2.0

Launch R 4.2.0 and run:

install.packages("IRkernel")
IRkernel::installspec(name = 'ir420', displayname = 'R 4.2.0', user = FALSE)

---

3. Verify Kernel Installation

After installing the kernels for each R version, list the available kernels to verify:

jupyter kernelspec list

The output should look like this:

Available kernels:
  python3      /usr/local/share/jupyter/kernels/python3
  ir363        /usr/local/share/jupyter/kernels/ir363
  ir420        /usr/local/share/jupyter/kernels/ir420

---

4. Launch JupyterLab

Start JupyterLab:

jupyter lab

When creating a new notebook, you will see multiple R options in the “New” dropdown or under “Kernel” > “Change Kernel”:

• R 3.6.3
• R 4.2.0

You can now select the version of R you want to use.

FYI, this is how it looks like in the JupyterLab

---

Notes:

• Each version of R must have IRkernel installed separately.
• Use unique names (name parameter) for each R kernel to distinguish them in Jupyter.
• If you install a new R version later, simply repeat the above steps for that version.