Notebook examples¶
These examples are rendered from the Jupyter notebooks in
examples/notebooks. They are published as regular documentation pages so
they can be searched, linked, and read without opening Jupyter.
Choose a notebook¶
Infomap quickstart — first end-to-end Infomap workflow in Python.
Infomap options guide — what every Infomap option is for, with runnable examples and a complete reference table.
Compare Infomap and Louvain with NetworkX — compare Infomap and Louvain with NetworkX when your graph already lives in NetworkX.
Compare Infomap, Louvain, and Leiden with igraph — compare Infomap, Louvain, and Leiden when you want igraph-native clustering objects.
Compare Infomap and Leiden in a Scanpy workflow — compare Infomap and Leiden in an AnnData and Scanpy-style workflow.
Run Infomap on GraphRAG-style tables — run Infomap on GraphRAG-style entity and relationship tables and compare with Leiden.
Run Infomap on HPC — native CLI recipes for scheduler-aware HPC runs and Python shard merging.
Performance and run planning — estimate run time and memory from network size, threads, trials, and hierarchy depth, with empirical scaling laws.
Additional tutorial notebooks¶
The notebook source tree also includes companion material for Community Detection with the Map Equation and Infomap: Theory and Applications:
the two-level map equation;
the two-level search phase and solution landscapes;
memory, multilayer, temporal, and multi-body network models;
networks with metadata, bipartite structure, and incomplete data;
map equation centrality, similarity, bioregions, and model selection with correlational data.
Those source notebooks are available in examples/notebooks. Some require external research code or additional data-processing packages and are not rendered in this first docs set.
Run locally¶
From an Infomap source checkout:
python -m pip install -e '.[notebooks]'
cd examples/notebooks
jupyter lab