How guofei9987/scikit-opt works
scikit-opt is a Python library for swarm-intelligence and evolutionary optimization. A user supplies an objective function and configures an algorithm, runs its iterative search, and receives a best solution and objective value. The principal workflow is clearest for genetic algorithms and differential evolution: GA evaluates populations through transformed objective functions and uses selectable operators; DE specializes the shared genetic-algorithm base. The README also documents PSO, annealing, ant-colony, fish-swarm, and immune algorithms, plus custom operators, continued runs, and acceleration modes. Several algorithm implementations were not excerpted, so their internal links are omitted.
Main entry points: GA.py Genetic algorithm · DE.py Differential evolution · tools.py Function modes · PSO.py Particle swarm
Components 17
Search algorithms
- Differential evolution
sko/DE.py - Genetic algorithm
sko/GA.py - Particle swarm
sko/PSO.py - Simulated annealing
sko/SA.py - Ant colony
sko/ACA.py - Artificial fish swarm
sko/AFSA.py - Immune algorithm
sko/IA.py
GA operators
- Crossover operators
sko/operators/crossover.py - Mutation operators
sko/operators/mutation.py - Selection operators
sko/operators/selection.py - Fitness ranking
sko/operators/ranking.py - GPU operators
sko/operators_gpu
Evaluation runtime
- Function modes
sko/tools.py
Extension layer
- Algorithm base
sko/base.py
Other components
- Optimization user
- Objective function
- Best solution
Connections 26
- Optimization user → Differential evolution: configures and runs
README.md - Optimization user → Genetic algorithm: configures and runs
README.md - Optimization user → Particle swarm: selects algorithm
README.md - Optimization user → Simulated annealing: selects algorithm
README.md - Optimization user → Ant colony: selects algorithm
README.md - Optimization user → Artificial fish swarm: selects algorithm
README.md - Optimization user → Immune algorithm: selects algorithm
README.md - Objective function → Differential evolution: evaluated by
README.md - Objective function → Genetic algorithm: evaluated by
README.md - Differential evolution → Genetic algorithm: inherits search base
sko/DE.py - Differential evolution → Best solution: returns best
sko/DE.py - Genetic algorithm → Best solution: returns best
sko/GA.py - Genetic algorithm → Function modes: transforms objective
sko/GA.py - Particle swarm → Function modes: transforms objective
sko/PSO.py - Genetic algorithm → Algorithm base: inherits base
sko/GA.py - Differential evolution → Crossover operators: imports operators
sko/DE.py - Differential evolution → Mutation operators: imports operators
sko/DE.py - Differential evolution → Selection operators: imports operators
sko/DE.py - Genetic algorithm → Crossover operators: uses operator
sko/GA.py - Genetic algorithm → Mutation operators: uses operator
sko/GA.py - Genetic algorithm → Fitness ranking: uses operator
sko/GA.py - Genetic algorithm → Selection operators: uses operator
sko/GA.py - Genetic algorithm → GPU operators: registers GPU variants
sko/GA.py - Immune algorithm → Genetic algorithm: imports algorithm base
sko/IA.py - Optimization user → Algorithm base: registers custom operators
README.md - Optimization user → Function modes: selects run mode
README.md