How skakim/braessparadox works
Project type and purpose
• This is a scientific/experimental Python project (“BraessParadox”) for running simulations and experiments on the Braess Paradox in traffic networks.
• It ingests network definitions and agent-based cost/choice files, runs path-finding and cost-evaluation algorithms, and outputs results for analysis.
File structure
Top-level directories and key files:
• network-files-master/
– Contains raw network definitions (.net) under subfolders (Bar-Gera TNTP, Braess-graphs, etc.).
– tools/: scripts to convert or validate network files (convert_bargera.py, validator.py, python_example.py).
• kspmaster/
– A small library for K-Shortest Paths (KSP.py, __init__.py), with an example caller (ksp_calling_example.py).
• py_expression_eval/
– A utility package for parsing/evaluating cost expressions.
• Agent.py
– Core agent logic for computing routes/costs.
• main.py
– Orchestrator script that ties together network loading, agent logic, KSP calls, and output generation.
• greedy_agents/
– Preset experiment results and inputs for a greedy-agent variant.
• Various text files at root (choices0.txt, cost0.txt, edges_usage8.txt, g_costs.txt)
– Example inputs or baseline outputs.
README
• Confirms the project is “Experiments on Braess Paradox.”