Quick Dashboarding With Panel

Updated: Mar 26

A bespoke, polished data science dashboard can be a beautiful thing for anyone looking to make data-driven decisions. And yet, not every project can afford setting up elaborate dashboards that cost money and developer time.


In this post, we show you how to construct a quick dashboard using Panel & Python without ever leaving the comfort of your Jupyter notebook.


For this example, we want to create a dynamic plot dashboard displaying historical trends of popular (user-supplied) baby names.

While the historical trends of name registrations are not at the forefront of business decisions, similar name-indexed data queries could involve stock indexes, product names in your catalog (or a competitor's), or perhaps predictions of future trends for your named-apparel business.


This baby-name data originally comes from the US Social Security open data; we have modified it for ease of use and to off-load some pre-processing needed for the final plots.


For each observation, we have preserved Year, Name, and Gender as features from the original data. We also have a feature Normalized that represents the percentage of all names within a given year.


We could add names common to M & F genders—say Dillon (M) & Dillon (F)—but we leave them distinct here to preserve their individual trends.


We present a random slice of the data below to build intuition.


from IPython.core.display import display, HTML
display(HTML("<style>.container { width:90% !important; }</style>"))
import pandas
import hvplot.