The Times or The Blackpool Herald) for digitisation using a range of criteria. We were using a Jupyter notebook to narrow down a large dataset of historical newspaper titles held by the British Library (here a newspaper title means eg.
Before I go on to how, I’ll explain why we ended up looking into this in the first place. Can you put a D3 JavaScript visualisation into a Python Jupyter notebook? Yes! (Though there are some hacky bits). But you may find you need more customisation in your visualisation design than these libraries support.ĭ3 is a low-level JavaScript visualisation library: you’ll write more code, but it has an ‘ expressivity advantage’.
the classic Matplotlib and, with chart interactivity nicely built in with Jupyter, Altair, Bokeh, and Plotly. There are a number of Python libraries for doing visualisation eg. In this post, I explain why we needed to use the JavaScript visualisation library D3 in a Python Notebook, and share the steps I took to get it working for our specific goals. In Living with Machines we’ve found ourselves using Jupyter notebooks widely, largely in Python. Tim Sherratt’s GLAM Workbench, for example, uses Jupyter notebooks to interweave data processing and analysis with visualisations and commentary. They’re becoming increasingly popular in cultural heritage data/digital humanities work.
Jupyter notebooks are a great environment for bringing together code with the outputs of processing, including visualisations.