- Pros: Panel allows you to turn jupyter notebooks into dashboards without much additional work. Due to the omnipresence of notebooks this is an easy way to get started fast with a shared dashboard. Also there is built-in support for the many common graph libraries and widgets. You can integrate your panel app into web application frameworks like Django or FastAPI and also use built-in OAuth integration. Panel also supports caching of results on the server side to speed up rendering dashboards.
- Cons: The learning curve of Panel is slightly higher than other dashboarding tools, because of the sheer choice of widgets and libraries that are supported and need to be learned as well. The focus of Panel is more on python and less on HTML/CSS.
- Use (Panel) if: you want to be flexible about existing widgets and libraries that are supported out of the box and want to make a Jupyter dashboard available for everyone as a web application.
- Alternatives: Dash. Jupyter.
- Pros: Streamlit is well suited for fast prototyping. On top of being easy to understand and having a good documentation to get started, the real-time feedback when changing code allows for a fast turnaround. If you do not need a fully scalable application running with thousands of users, then putting a Streamlit application in production should not be a problem.
- Cons: The biggest strength of fast prototyping is also the biggest weakness, as customization is limited. The look and feel can only be customized to a certain degree.
- Use (Streamlit) if: you want a fast way of building applications and visualizations for your users, for example for a small internal data app.
- Alternatives: Dash and Shiny
Firebolt Streamlit Example: https://github.com/spinscale/firebolt-streamlit-demo
%20logo.png){kind=icon}
