metaforge 1.0.1

MetaForge: A modular optimization framework for solving Job Shop Scheduling problems using metaheuristics and learning-based approaches.

🔧 MetaForge

MetaForge is a modular optimization framework for solving Job Shop Scheduling Problems (JSSP) using advanced metaheuristics and learning-based approaches.

🚀 From classic solvers like Tabu Search and Genetic Algorithms to modern DQN-based reinforcement learning and Neuroevolution — MetaForge brings it all together with clarity, structure, and fun.

---

---

🎯 Key Features

• ✅ Support for classic OR-Library and custom JSON benchmark formats
• 🧠 Modular solvers:
  • Tabu Search
  • Simulated Annealing
  • Genetic Algorithm
  • Ant Colony Optimization
  • Q-Learning
  • DQN (naive & replay-based)
  • Neuroevolution
• 📊 Beautiful convergence plots, runtime comparisons, and Gantt chart visualizations
• 📦 Easy packaging, CLI usage, and extension with new solvers
• 🔬 Designed for researchers, students, and practitioners alike

---

🚀 Quick Start

1. Install MetaForge

pip install metaforge

Or clone locally for development:

git clone https://github.com/Mageed-Ghaleb/MetaForge.git
cd MetaForge
pip install -e .

---

2.Run a Solver (Local or Remote Benchmark)

import os
import requests
from metaforge.problems.benchmark_loader import load_job_shop_instance
from metaforge.metaforge_runner import run_solver

# You can pass a local path or a URL to load the problem!
problem = load_job_shop_instance("https://raw.githubusercontent.com/Mageed-Ghaleb/MetaForge/main/data/benchmarks/ft06.txt")

# Run a solver (sa, ts, ga, aco, pso, ...)
result = run_solver("ga", problem, track_schedule=True)

print("Best Makespan:", result["makespan"])

---

3. Run All Solvers on All Benchmarks

python -m src.metaforge.utils.compare_solvers

Generates CSV, plots, and Gantt charts from the data/benchmarks/ folder.

---

4. Interactive Notebook

Explore MetaForge hands-on with our example notebook:

📓 MetaForge_Example.ipynb

It walks you through:

• Loading benchmark problems
• Running various solvers (TS, GA, DQN, etc.)
• Plotting convergence + Gantt charts
• Comparing performance across solvers

✅ Coming soon: One-click Colab version for browser-based use — no setup needed.

---

📚 Documentation

📘 Full documentation site coming soon! In the meantime, browse core modules below:

• 📖 Usage Guide (source)
• 🧠 Solvers Overview (source)
• 📂 Benchmark Datasets (source)

---

🧠 Why MetaForge?

Most libraries focus only on one type of solver. MetaForge unifies traditional, bio-inspired, and learning-based approaches in one clean, extensible Python package — built for experimentation, benchmarking, and educational use.

Whether you're doing a thesis, publishing research, or solving real-world factory problems — MetaForge is your launchpad. 🚀

---

📈 Contributing

We're just getting started! Feel free to:

• Suggest solvers, features, or dataset formats
• Fork and extend
• Submit pull requests with improvements

---

📄 License

MIT License — free for academic and commercial use.

---

👨‍💻 Author

Mageed Ghaleb
📧 [e-mail]
🔗 LinkedIn
🔗 GitHub

---

Built with ❤️ for solvers, schedules, and scientific curiosity.