simulated-annealing

What problem does it solve?

Simulated Annealing provides a robust approach for finding near-optimal solutions in complex, multimodal optimization landscapes where gradient information is unavailable or unreliable.

Core Features & Use Cases

• Global optimization for continuous, non-convex functions.
• Flexible neighbor generation (Gaussian, Uniform, Cauchy) and multiple cooling schedules.
• Suitable for parameter tuning, design optimization, and black-box problem solving.

Quick Start

Create a SimulatedAnnealing instance with your objective function and variable bounds, then call optimize to obtain the best solution. For example, import SimulatedAnnealing, define an objective and bounds, and run: sa = SimulatedAnnealing(objective, bounds); result = sa.optimize()