Shiny-Python
is a great framework to build interactive web applications using Python, and combining it with pygwalker
opens up new avenues for creating data-driven interactive web applications. In this tutorial, we'll walk you through the process of integrating pygwalker
with Shiny-Python
.
- Ensure you have both
shiny
andpygwalker
libraries installed. You can do this using pip:
pip install shiny pygwalker
Start by importing the required modules:
from shiny import App, ui
import pygwalker as pyg
from datasets import load_dataset
For this tutorial, we're leveraging the NYC-Airbnb dataset, but you can replace it with any dataset you prefer:
dataset = load_dataset("gradio/NYC-Airbnb-Open-Data", split="train")
df = dataset.to_pandas()
Define the user interface (UI) of the Shiny application using ui
components:
app_ui = ui.page_fluid(
ui.h1("Using pygwalker with Shiny"),
ui.markdown("This is a demo of using [pygwalker](https://github.com/Kanaries/pygwalker) with Shiny."),
ui.HTML(pyg.walk(df, spec="./viz-config.json", return_html=True, debug=False)),
)
Here's a quick breakdown of the UI components:
ui.h1()
: Adds a header to the Shiny app.ui.markdown()
: Allows you to include markdown-formatted text for descriptions or additional information.ui.HTML()
: This is wherepygwalker
comes into play. Thepyg.walk()
function transforms the dataset into an interactive visualization, which is then rendered in the Shiny app through this HTML component.
Though in this example, the server function is empty, it's where any reactive behavior would be defined. As the app grows more complex, you might find yourself populating this section with more interactive logic.
def server(input, output, session):
...
Finally, create the Shiny app instance and then run it:
app = App(app_ui, server)
If you want to immediately launch the app, you can add the line:
app.run()
or
shiny run --reload
By integrating pygwalker
with Shiny-Python
, you can harness the power of both tools to build impressive and interactive data visualization web apps. This not only provides a rich user experience but also allows a broader audience to engage with your data insights. Get started and unleash the potential of your datasets!