kaiwu-community 1.0.4

An SDK for CIM or QUBO.

Kaiwu Community

Language: 中文 | English

---

Kaiwu Community is a Python development toolkit designed specifically for QUBO (Quadratic Unconstrained Binary Optimization) problems, providing rapid modeling, flexible extension, and efficient solving capabilities.

Installation

pip install kaiwu_community

Requirements:

• Python 3.10.x
• Make (for development builds)
• Virtualenv (optional, for environment isolation)

For more installation information, please refer to the Installation Documentation.

---

Quick Start

Learning Resources

• Kaiwu SDK Help Center - View detailed documentation and learn QUBO modeling and solving
• Example Code - Browse complete code examples

Getting Help

• Developer Community - Participate in community activities and connect with other developers
• GitHub Issues - Submit issues or feature requests

Example: Traveling Salesman Problem (TSP)

This example demonstrates how to solve the Traveling Salesman Problem using Kaiwu SDK Community, including the following steps:

• Build a TSP problem model using a distance matrix
• Define QUBO objective function and constraints
• Solve the model using built-in solvers
• Validate the solution results

For the complete TSP example code, please see example/tsp_sdk_1_2_doc_example.py.

For more detailed explanations, please refer to the TSP Documentation.

---

Application Scenarios

Kaiwu SDK is suitable for:

• Combinatorial Optimization: Traveling Salesman Problem (TSP), Knapsack Problem, Graph Coloring, Max-Cut
• Machine Learning: Feature Selection, Cluster Analysis, Model Compression
• Quantum Computing Research: QUBO Model Research, Solver Development, Quantum Algorithm Verification

---

Contributing

Community contributions are welcome! 🎉

You can extend SDK functionality, such as new solver implementations, classical optimization algorithms, example programs, and documentation.

Contribution Example

Quickly extend functionality by inheriting base classes:

from kaiwu.core import OptimizerBase, SolverBase
import kaiwu

# Custom optimizer
class CustomOptimizer(OptimizerBase):
    def solve(self, ising_matrix=None):
        # Implement custom optimization logic
        pass

# Custom solver
class CustomSolver(SolverBase):
    def solve_qubo(self, model):
        # Implement custom solver logic
        self._optimizer.solve()
        pass

# Modeling
qubo_model = kaiwu.qubo.QuboModel()
# Calling the custom component
optimizer = CustomOptimizer()
solver = CustomSolver(optimizer)
result = solver.solve_qubo(qubo_model)

➡️ For detailed contribution guidelines, please refer to the Contributing Guide.

---

License

This project is open-sourced under the Apache License 2.0. For details, please see the LICENSE file.