search

lux-org / lux

Automatically visualize your pandas dataframe via a single print! 📊 💡
Apache License 2.0
5.14k stars 364 forks source link

New Action: Automatic geographic attribute visualizations #123

Closed dorisjlee closed 3 years ago

dorisjlee commented 3 years ago

For geo-related attributes (longitude, latitude, state, countries, etc), we should detect it as a special data type and display a choropleth automatically.

micahtyong commented 3 years ago

Hi @dorisjlee, I'm thinking of potentially picking this one up! From my understanding, Lux depends on Altair. If that's the case, I would use the Altair Cloropleth Map to display a choropleth for each non-geo-related quantitative group. More precisely, would a suitable solution to this be a repeated choropleth map like this example?

dorisjlee commented 3 years ago

Hi @micahtyong, This is a fun issue to work on! What you suggested is a good starting point. We can have a choropleth like this for states, where we have a color choropleth for each of the quantitative variables. We probably first need to detect geo-related attribute types first, then implement this as a custom action.