PometryDocs
GitHub

Raphtory + Jupyter

Interactive graph intelligence and prototyping.

Explore networks, test algorithms, and prototype full graph intelligence workflows in the most popular environment for data science.

Why Use Jupyter for Raphtory?

  • Immediate Feedback: See the results of complex temporal queries instantly.
  • Rich Visualization: Integrate Matplotlib, Plotly, and network-specific visualization libraries.
  • Reproducible Research: Document your analysis steps alongside your code.

Setup

Example Analysis Workflow

1. Ingest Data

Load your interaction data and build the temporal graph.

2. Exploratory Statistics

Visualize the degree distribution to understand your network's connectivity.

3. Run Temporal Analysis

Analyze the network at different points in time to see how it evolves.

Best Practices

  1. Memory Management: Use del g and gc.collect() if you are iterating over many large graph versions in a single session.
  2. Modularize Early: Once you are happy with a prototype, move the logic into a separate .py module to keep your notebooks clean.
  3. Save State: Use g.save_to_file() to checkpoint your graph so you don't have to rebuild it every time you restart the kernel.