Build linked charts with AI
Let readers select a group of points in one view and see which records survive in the others.
Build linked charts with AI when the question you want to ask travels between two views. A print workshop might plot paper weight against rejected sheets, then use a second plot to show which presses produced those batches. With holoviews, your agent can give the two plots a shared selection. Circle a cluster of heavy stock and inspect the corresponding presses without looking up batch numbers by hand.
Decide what a mark stands for before choosing its display options. A tooltip can show the batch size alongside the rejection count. Putting the units on the axis keeps them visible even before somebody moves the pointer. When several presses share a chart, a legend belongs to the combined view.
What it does
- Link a scatter plot to a second view of the same records.
- Use readable dates, quantities and batch labels in hover details.
- Overlay several series and move their shared legend clear of the points.
- Give a row of small charts common styling across a notebook session.
- Keep a toolbar available on hover when inspecting values matters more than zooming.
Install and use
- Start with Python, HoloViews, Bokeh and a small table that you can inspect.
- Name the record key and the pair of views you want to connect.
- Select a known batch in the result and check its values against the source rows.
FAQ
Source
Made by HoloViz Dev. Original skill: holoviews, published on GitHub under the BSD-3-Clause licence.
View on GitHub