Pyvis Network Visualization Library

0.3.2 · active · verified Sat Apr 11

Pyvis is a Python library designed for the quick generation of interactive network graphs with minimal Python code. It acts as a wrapper around the popular JavaScript Vis.js library, enabling users to build graphs in Python, customize nodes and edges with various properties, and export them as interactive HTML files. The library is currently at version 0.3.2 and maintains an active development status with regular updates, including a significant 0.3.0 release that merged an experimental branch and added features like node/edge filtering from the web interface.

Warnings

gotcha Node IDs must be unique within a network. Re-adding a node with an existing ID will overwrite its properties, not create a new node. Ensure uniqueness for proper graph structure.
Fix: Assign unique integer or string IDs to each node using `net.add_node(id, ...)` or ensure lists passed to `net.add_nodes()` contain unique identifiers.
gotcha Some attributes from the underlying VisJS library may not work as expected or at all when accessed via Pyvis. This is due to direct Python translation limitations.
Fix: Consult both the Pyvis documentation (specifically `network.Network.add_node()` and `network.Network.add_edge()` docs) and the original VisJS documentation for supported attributes and their Python equivalents. Experimentation may be required.
gotcha When using `net.show_buttons(filter_menu=True, select_menu=True, ...)` to interactively tweak layout or filtering settings in the browser, these changes are not automatically saved to your Python code. To persist interactive settings, you need to copy the generated options from the browser UI and apply them to your `Network` object using `net.set_options()`.
Fix: After interactively adjusting settings via `show_buttons()`, use the 'generate options' button in the UI to copy the JSON configuration. Then, pass this JSON string to `net.set_options(json_string)` in your Python script to save the desired layout and interaction configurations.
gotcha Pyvis is primarily a visualization tool and is not as robust as libraries like NetworkX for complex graph analysis or generating network metrics (e.g., centrality). For advanced graph algorithms, it's recommended to build and analyze your graph with NetworkX first, then pass the NetworkX graph to Pyvis for visualization using `net.from_nx(nx_graph)`.
Fix: Leverage NetworkX for graph creation, manipulation, and analysis. Integrate Pyvis at the visualization stage by converting your NetworkX graph into a Pyvis network using `net.from_nx()`.

Install

pip install pyvis Install Pyvis

Imports

Network
```
from pyvis.network import Network
```

Quickstart

This quickstart demonstrates how to create a basic interactive network graph using Pyvis. It initializes a `Network` object, adds three nodes with custom labels, titles, and colors, then connects them with edges. Finally, it generates an HTML file (`basic_network.html`) that can be opened in a browser to interact with the visualization. The `notebook=False` parameter ensures it generates a standalone HTML file.

from pyvis.network import Network

# Create a Network object
net = Network(height="750px", width="100%", bgcolor="#222222", font_color="white", notebook=False)

# Add nodes
net.add_node(1, label="Node 1", title="This is node 1")
net.add_node(2, label="Node 2", title="This is node 2", color="#00ff1e")
net.add_node(3, label="Node 3", title="This is node 3")

# Add edges
net.add_edge(1, 2, title="connects 1 and 2")
net.add_edge(2, 3, title="connects 2 and 3", value=5)

# Generate and save the network visualization to an HTML file
net.show("basic_network.html")

view raw JSON →