Hvplot .interactive - The Easiest Way to Create an Interactive Dashboard

🐍🐼📈❤️

This repo supports the blog post

The Easiest Way to Create an Interactive Dashboard in Python. Turn Pandas pipelines into a dashboard using hvplot .interactive

by Sophia Yang and Marc Skov Madsen

Check out the notebook and app on Github page, Binder, or Colab

Github Page	Jupyter Notebook	Jupyter Labs	Panel Apps	VS Code	Colab
Github

.

Installation

You can use either conda or pip. Here we provide the instructions for conda.

Basic

conda install -c conda-forge hvplot panel pandas jupyterlab

Advanced

Includes the Jupyter Preview and jupyter-panel-proxy server. Will also work in VS Code.

conda create --name hvplot_interactive -c conda-forge hvplot panel pandas jupyterlab jupyter-panel-proxy jupyter_bokeh
conda activate hvplot_interactive
jupyter serverextension enable panel.io.jupyter_server_extension

Run notebook locally

To run the notebook

jupyter lab hvplot_interactive.ipynb

If you installed jupyter-panel-proxy, then (on unix based systems) you can open the app at http://localhost:8888/panel/

Serve the app locally

To serve the app with auto reload you can run

panel serve hvplot_interactive.ipynb --autoreload --show

The app will appear at http://localhost:5006/hvplot_interactive

Use VS Code on Binder

To work with Panel and hvplot in VS Code in Binder you will need to set the python and Jupyter interpreter to notebook before you open the notebook.

Issues

On the back of this work we created or updated the following Issues

• holoviz/hvplot#696
• Holoviews/ hvplot not working in Notebook in code server on Binder.

Feature Requests

On the back of this work we created or updated the following Features

• Tabulator fast theme not working in notebook
• Enable users to easily get running environment: notebook, lab, colab, vscode, server, na
• Auto determine and configure the environment
• Could hvplot.interactive please support outputting to responsive hvplots
• pn.widgets not working in a list

License: MIT