Recently, I wanted to try out a library that goes along with a paper that was just published. I didn’t want to actually install the library, because it comes with a bunch of specific dependencies, so I used a virtual environment for my testing. This is easy

python3 -m venv env

Which creates a virtual environment called env. The library came with a bunch of examples in Jupyter notebooks. I usually just run Jupyter notebooks with standard Python installation on my machine, but in this case I needed to use the virtual environment that had all of the dependencies I needed. It’s easy to set this up. First, activate your new virtual environment

source env/bin/activate

Then, in the environment, install ipykernel

pip install ipykernel

Finally, add the virtual environment as a kernel.

python -m ipykernel install --name=env

Then, when you create a new notebook you see env as an option and can have sets of notebooks with different dependencies. When you want to go back to your default Python installation just type deactivate