Abdelrahman MohamedWork with me

Writing

Building an Interactive Dashboard in a Jupyter Notebook

You don't need a separate BI tool to explore data interactively — ipywidgets gets you dropdowns, buttons, and live updating charts right inside the notebook.

1 min read

Jupyter notebooks are usually static: run a cell, look at a chart, edit code, re-run. For quick exploration, that’s fine. But sometimes you want something closer to a real dashboard — pick a value, see the numbers update — without leaving the notebook or standing up a separate app.

ipywidgets gets you there. Using a League of Legends champion stats dataset as an example:

  1. Load and clean the data — read the CSV with pandas, check for missing values, drop rows for champions with zero picks or presence (they’d just clutter the output).
  2. Fix the types — some columns come in as percentage strings ("52.3%"); convert them to floats so they’re usable in calculations and charts.
  3. Build the controls — a dropdown of champion names and a refresh button, wired up with ipywidgets.
  4. Wire the callback — when the dropdown changes, a function looks up that champion’s picks, bans, wins, and losses, and redraws a bar chart with matplotlib.

The result: pick any champion from the dropdown and the stats and chart update instantly, all inside the notebook — no separate dashboarding tool required.

This pattern generalizes past League of Legends stats — anywhere you’re exploring a dataset interactively and don’t want the overhead of a full BI tool, a notebook with a few widgets gets you most of the way there.