Shiny for Python is out of Alpha
Shiny for Python has come out of alpha. It is a foss device to construct and share interactive visualizations & dashboards. We predict it compares favorably to different common internet app growth frameworks focused at information scientists. Contemplate having a look.
Announcement from the event crew: [https://shiny.rstudio.com/blog/shiny-python-general-availability.html](https://shiny.rstudio.com/weblog/shiny-python-general-availability.html)
Posit’s common announcement: [https://posit.co/blog/shiny-for-python-is-now-generally-available/](https://posit.co/weblog/shiny-for-python-is-now-generally-available/)
(Full disclosure, I work at Posit PBC.)
Comments ( 9 )
Will this work with rsconnect?
Top for the disclosure
Why do Posit consider this “more developer friendly” than Dash?
Looking at the examples, it looks very un-pythonic to me.
Great! I use to work a lot in shiny with R over 5 years ago, but I had to switch to flask, dash, or streamlit since I needed to work in python. I am excited to build a shiny app again.
Looks great. I used to develop a few shiny apps back when I worked with R. I’m glad that shinywidgets is a package included here as the ipywidgets tool was important from a functionality perspective but not smooth to use. Just need a shinydashboard module now.
Weird that they ported it over. I used it back in 2018, but there’s much better solutions for dashboarding in python. Others have mentioned Dash. I think plotly has their own variant, but I’ve been using Apache Superset which allows CSS formatting, you can change branding/flavor icons, and all the charts/dashboards can be imported/exported from JSON, which makes it much easier for a deployment standpoint. Also has row level security and supports Oauth so is a lot more secure than Shiny.
I normally use flask, streamlit, or dash but will check it out.
As someone that has worked with Shiny (R) as well, the barrier of entry in streamlit is lower.
It’s excellent. Go to shinylive.io, check out some examples with source code, then and only then can you knock it.
It’s aim is to provide a framework for web app development that 1) requires no web dev skills and 2) you’ll never grow out of – if you happen to be proficient with HTML, CSS, etc then you can make your apps that much better.
As a DS lacking solid front end technical chops, Python Shiny is very appealing.
Note on dash: it’s stateless meaning that your data has to be reloaded every time it reacts to user input (or you implement a kludgey workaround that’s rough to maintain). Not so with Shiny – much easier to design a reactive, maintainable, and scalable application.
Have some fun playing around with Shiny, I know I will this weekend!