Benchmarks for Boost Graph Library algorithms against reference implementations.
cd louvain && ./run_benchmark.sh # full suite
cd louvain && ./run_benchmark.sh --quick # smoke-testSee louvain/README.md for details and results.
boost-graph-benchmarks/
├── requirements-common.txt # Shared Python deps
└── louvain/ # Louvain community detection
├── run_benchmark.sh # Entry point (venv, build, run, plot)
├── src/ # C++ sources + CMake
├── scripts/ # benchmark.py + visualize.py
├── vendor/ # gen-louvain (auto-downloaded)
└── results/ # PNGs (committed), CSVs (gitignored)
- Python 3.8+, CMake 3.15+, C++17 compiler, Boost headers
- Python deps installed automatically into per-benchmark venvs