pygwalker

What it is and what it does

PyGWalker is a Python library that embeds Graphic Walker, an open-source Tableau alternative, into Jupyter Notebooks as an interactive widget. It converts a pandas DataFrame into a drag-and-drop visual interface where you can create charts, filter data, and explore patterns without writing code. The library supports multiple chart types, real-time visualization updates, and includes a data table with profiling and type-change capabilities.

The package integrates deeply with the Jupyter ecosystem through ipywidgets and ipylab, and optionally uses duckdb as a computation engine for handling larger datasets. It allows you to save chart configurations to JSON files, export visualizations as SVG or PNG, and maintain your analysis state across sessions. The runtime dependencies—including pandas, numpy, pyarrow, sqlalchemy, and sqlglot—provide the data manipulation and query capabilities underlying the interactive interface.

Use it for:

• Exploratory data analysis in Jupyter without writing visualization code—load a CSV and drag dimensions/measures to build charts interactively.
• Data cleaning and outlier detection using the visual data table and filtering tools to identify and annotate inconsistencies.
• Sharing analysis results by saving chart configurations and exporting visualizations as static images or interactive HTML.
• Building Streamlit dashboards that embed PyGWalker for web-based interactive data exploration without Jupyter.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Turns pandas DataFrames into interactive drag-and-drop visual exploration interfaces within Jupyter Notebooks, supporting chart creation, data cleaning, and filtering without coding.

Yes. PyGWalker is actively maintained, has no known vulnerabilities, installs with low friction, and fills a clear need for code-free interactive data exploration in Jupyter. The permissive Apache license and strong community adoption make it a safe choice. Install it if you want to reduce boilerplate visualization code and enable non-technical stakeholders to explore data interactively; skip it only if you need offline-only operation or have strict dependency minimalism requirements.

Install

pygwalker on PyPI

pip install pygwalker

uv add pygwalker

poetry add pygwalker

Installing pygwalker

Before you install

Low friction install with a pure-Python wheel. Active maintenance (last commit 2026-08-10) and strong popularity. Requires 25 runtime dependencies including pandas, duckdb, and ipywidgets, which may add setup complexity in constrained environments.

License in practice

Licensed under Apache Software License (permissive), allowing free use, modification, and distribution with minimal restrictions—suitable for commercial and open-source projects.

Quickstart

pip install pygwalker

import pandas as pd
import pygwalker as pyg

df = pd.read_csv('data.csv')
walker = pyg.walk(df)

Requires Jupyter Notebook or compatible environment (ipywidgets, ipylab); kernel_computation=True requires duckdb for larger datasets.

Verify before relying

• Maximum dataset size supported with kernel_computation=True (description mentions '<=100GB' but this is not in the fact sheet)
• Performance characteristics and latency for interactive operations on typical dataset sizes
• Compatibility with JupyterLab vs. classic Notebook and other notebook environments

Package facts

Evidence: pygwalker-0.5.0.1-py3-none-any.whl

Keywords: data-analysis, data-exploration, dataframe, jupyter, pandas, tableau, tableau-alternative, visualization

Tags

More Information Analysis packages