Linux:
$ sudo apt install python3-pip python3-venv python3-tk
Mac OS X (via Homebrew):
$ brew install python@3
$ pip3 install virtualenv
$ python3 -m venv /path/to/your/venv
$ source /path/to/your/venv/bin/activate
$ pip3 install --no-cache-dir -r requirements.txt
$ jupyter nbextension enable --py widgetsnbextension
Some docs online state that you need to create a separate kernel from your virtualenv, and this is how you can go about that:
$ python -m ipykernel install --user --name=fmp
Though if you don't create a kernel, so long as you used the pip from your activated virtualenv to install jupyter, the default Python kernel should already be good to go:
$ jupyter kernelspec list
Available kernels:
fmp /home/jmf/.local/share/jupyter/kernels/fmp
python3 /home/jmf/src/venv/fmp/share/jupyter/kernels/python3
You can select which kernel to use from a drop-down menu in the jupyter UI.
To start out a local jupyter instance, you can run the command below from this directory:
$ jupyter notebook <notebook_file.ipynb>
And then open your browser at http://localhost:8888 to access it. Click on any of the .ipynb files to open a notebook.
If you're running this notebook server remotely, however, and trying to access the it from a different machine, you can create a ssh tunnel for that.
From your client (the machine you're ssh'ing from):
$ ssh -L 8080:localhost:8888 [email protected]From your server (the machine you ssh'ed to, and from which you'll be running the jupyter notebook server):
$ ./run.sh # starts out the jupyter server
Then from your client machine, direct your browser to http://localhost:8080
to access the jupyter server.