PyGWalker Gradio API Guide

This guide explains how to integrate PyGWalker with Gradio applications, enabling interactive data visualization capabilities within your Gradio UI.

Quick Start

To render PyGWalker in a Gradio app, use the get_html_on_gradio function:

import gradio as gr
import pandas as pd
from pygwalker.api.gradio import PYGWALKER_ROUTE, get_html_on_gradio
 
with gr.Blocks() as demo:
    df = pd.read_csv("data.csv")
    pyg_html = get_html_on_gradio(df, spec="./gw_config.json", spec_io_mode="rw")
    gr.HTML(pyg_html)
 
app = demo.launch(app_kwargs={"routes": [PYGWALKER_ROUTE]})

Key Function: get_html_on_gradio

This function generates the HTML for PyGWalker to be embedded in a Gradio application.

Important Parameters

Best Practices

1. Data Loading: Load your data efficiently, considering caching for large datasets.

2. Configuration: Use a separate JSON file for chart configurations (spec parameter) to easily manage and update visualizations.

3. Performance: For large datasets, set kernel_computation=True to enhance performance.

4. Theming: Adjust the appearance parameter to match your Gradio app's theme.

5. Interactivity: Set spec_io_mode="rw" to allow users to modify and save chart configurations.

Advanced Usage

Custom Routing

Always include the PYGWALKER_ROUTE in your Gradio app's routes:

app = demo.launch(app_kwargs={"routes": [PYGWALKER_ROUTE]})

This ensures proper communication between PyGWalker and your Gradio app.

Combining with Other Gradio Components

You can combine PyGWalker with other Gradio components for a more comprehensive data analysis tool:

with gr.Blocks() as demo:
    gr.Markdown("# Data Visualization with PyGWalker")
 
    with gr.Tab("PyGWalker Explorer"):
        pyg_html = get_html_on_gradio(df, spec="./gw_config.json")
        gr.HTML(pyg_html)
 
    with gr.Tab("Data Summary"):
        gr.DataFrame(df.describe())
 
    with gr.Tab("Raw Data"):
        gr.DataFrame(df)

Troubleshooting

• If visualizations don't render, check your data format and ensure PYGWALKER_ROUTE is properly included.
• For performance issues with large datasets, try enabling kernel_computation and consider data sampling techniques.

For more detailed information and advanced features, refer to the PyGWalker documentation (opens in a new tab).