Benchmark Functions API

API reference for benchmark functions available in opt.benchmark.functions.

Overview

All benchmark functions share the same interface:

from opt.benchmark.functions import sphere, rosenbrock, ackley
import numpy as np

def benchmark_function(x: np.ndarray) -> float:
    """
    Args:
        x: Input vector of shape (n,)

    Returns:
        Scalar fitness value (lower is better)
    """
    pass

Available Functions

For detailed mathematical definitions and properties, see Benchmark Functions.

Unimodal Functions

• sphere(x) - Simple sum of squares
• rosenbrock(x) - Rosenbrock valley function
• schwefel_2_22(x) - Schwefel 2.22 function
• schwefel_1_2(x) - Schwefel 1.2 function
• step(x) - Step function

Multi-Modal Functions

• ackley(x) - Ackley function
• shifted_ackley(x) - Shifted Ackley function
• rastrigin(x) - Rastrigin function
• griewank(x) - Griewank function
• levy(x) - Levy function
• schwefel(x) - Schwefel function

Usage

from opt.swarm_intelligence import ParticleSwarm
from opt.benchmark.functions import sphere, ackley

# Test with sphere function
optimizer = ParticleSwarm(
    func=sphere,
    lower_bound=-5.12,
    upper_bound=5.12,
    dim=10,
    max_iter=100
)
best_solution, best_fitness = optimizer.search()

# Test with Ackley function
optimizer = ParticleSwarm(
    func=ackley,
    lower_bound=-32.768,
    upper_bound=32.768,
    dim=10,
    max_iter=100
)
best_solution, best_fitness = optimizer.search()

See Also

• Benchmark Functions Details - Mathematical definitions and properties
• Benchmark Methodology - How benchmarks are conducted
• Benchmark Results - Performance comparison results