Skip to content

SKILL.md

Latest commit

 

History

History
34 lines (24 loc) · 1.01 KB

SKILL.md

File metadata and controls

34 lines (24 loc) · 1.01 KB
Raw
name project-context
description Project context for Baswana-Sen spanner experiments — algorithm goals, structure, parameters, and workflow. Use when implementing or modifying spanner code, running experiments, or analyzing results.

Baswana-Sen Spanner Experiments

Goals

  1. Implement Baswana-Sen and Greedy spanner algorithms correctly
  2. Evaluate on Erdős–Rényi G(n,p) graphs across n, p, k
  3. Measure spanner size, stretch, and runtime
  4. Produce reproducible CSV results and notebook visualizations

Key decisions

  • Custom adjacency lists (not NetworkX) for performance
  • Unweighted graphs; BFS for distances
  • Largest connected component when G is disconnected
  • Deterministic seeds via set_seed()

Default experiment parameters

See scripts/run_all_experiments.py — typical n ∈ {500,1000,2000}, k ∈ {2,3,4,5}, 5 repetitions.

Commands

make install && make test
python scripts/run_all_experiments.py
jupyter lab

Policy and constraints: root AGENTS.md.