How baopng/nsga-ii works
This is a small Python library for multi-objective optimization. The user defines objectives and variable bounds in a Problem, then runs Evolution; its utilities create and evaluate candidates, select and vary them, rank populations, and return the final nondominated front. The graph focuses on this runtime workflow and its core data structures. README documentation establishes the public workflow; source excerpts show the internal calls. Example problems are documented but are not part of the library runtime.
Main entry points: evolution.py Evolution runner · utils.py NSGA-II operators · problem.py Problem definition · population.py Population
Components 6
Problem Model
- Problem definition
nsga2/problem.py
Evolution Runtime
- Evolution runner
nsga2/evolution.py - NSGA-II operators
nsga2/utils.py
Optimization Data
- Population
nsga2/population.py - Individual
nsga2/individual.py
Other components
- Library user
Connections 7
- Library user → Problem definition: defines objectives
README.md - Library user → Evolution runner: runs evolution
README.md - Evolution runner → NSGA-II operators: delegates evolution
nsga2/evolution.py - NSGA-II operators → Problem definition: generates and evaluates
nsga2/utils.py - NSGA-II operators → Population: builds and ranks
nsga2/utils.py - Problem definition → Individual: creates candidates
nsga2/problem.py - Evolution runner → Library user: returns final front
README.md