Perhaps you’ve come to appreciate the value of a Jupyter (iPython) notebook for certain development, research, or learning scenarios, like I have. I’ll skip listing out all the merits of such notebooks, but suffice it to say I wish I had started with them when I took up learning Python years ago.
Now, I’ve come to find I was really needing a way to access some notebooks in different contexts. Sometimes at the house, working on a side project, sometimes at the office when working on some database migration or analysis, sometimes just wanting to look something up while out of town and without my laptop, but I did have my phone.
In the last week, I have revisited Codeanywhere and have been really enjoying the robust services they offer. I decided to see if Codeanywhere would be a place I could let my Jupyter notebooks live (remote git repo). After several minutes, I was able to get the notebook app serving the Jupyter instance publicly so I could have the same notebooks available in whatever context I found myself. Read on for the process I took.
1. If you haven’t one already, create a free (or paid) Codeanywhere account. Pick one of the Python containers to get started. I chose Python 3.
2. Next, go to Anaconda’s website and get the link for the Linux 64-bit Miniconda installation.
3. Back in your Codeanywhere terminal, you’ll need to use sudo to download Miniconda, then you can install it:
$ sudo wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh $ bash Miniconda3-latest-Linux-x86_64.sh
4. Now that you have Miniconda installed, you now have the excellent
conda package and environment manager at your dispoal. So now you can simply:
$ conda install jupyter
5. Excellent. Almost there. Now you can start your jupyter notebooks like this:
$ jupyter notebook --ip=0.0.0.0
6. Now then, with your jupyter notebook dashboard publicly available, you need to go acquire the URL at which to access what your container is serving. So, in your Codeanywhere IDE, in the left sidebar, right click on your containers name, and choose ‘Info’. Visit the displayed URL and you’ll have a working, remote Jupyter Notebook ready to work.