mbls 0.0.3

Model-Based Local Search algorithm designer built on Routix

mbls

A Python toolkit for Model-Based Local Search algorithm design.

Built on top of Routix, it provides a modular framework for orchestrating subroutines, managing experimental runs, and integrating mathematical models into heuristic search routines.

Features

• Modular Architecture Compose and extend LNS-style strategies using reusable subroutine components.
• Seamless Routix Integration Take advantage of structured routine execution, logging, timers, and experiment summarization.
• Extensible Modeling Layer Easily add custom models, constraints, or solvers.

✅ Includes support for OR-Tools CP-SAT via mbls.cpsat.

Installation

pip install mbls

Requirements

• routix (experiment orchestration)
• ortools (for mbls.cpsat components)

🚀 Example

from mbls.cpsat import CustomCpModel

# Initialize model
model = CustomCpModel()

# Define variables, constraints, and objective
# ...

# Configure and solve
model.init_solver(computational_time=10.0, n_threads=4)
status, elapsed, ub, lb = model.solve_and_get_status(10.0, 4)

print(f"Status: {status}, UB: {ub}, LB: {lb}")

🧩 Extendability

mbls is designed for research and experimentation. You can:

• Subclass SubroutineController to define custom LNS or hybrid metaheuristics
• Extend CustomCpModel to support new problem domains
• Compose repeatable flows with structured routine names and modular inputs

License

MIT License