RANTAI MetNumLab 🧮

Interactive Numerical Methods Lab for teaching — CSV & JSON input → plots, iterations, reports.

---

What is MetNumLab?

RANTAI MetNumLab is an educational web tool for Metode Numerik (Numerical Methods) that allows students to upload CSV / JSON inputs and run standard numerical algorithms (e.g., Gaussian elimination, Newton-Raphson, RK4, Simpson). The platform focuses on concepts & interpretation (table of iterations, plots, downloadable report) rather than coding syntax.

---

Goals

• Make numerical experiments accessible to non-programmer students.
• Provide reproducible job runs with downloadable results (CSV/JSON/PNG/PDF).
• Provide instructor features: assignment creation, auto-grading (tolerance-based), and grade export.

---

Key Features

• Upload CSV (matrices, samples) or JSON (functions & parameters).
• Pre-implemented algorithms: Gauss Elimination, LU, Newton-Raphson, Secant, Simpson, RK4.
• Interactive iteration tables & plots (convergence, solution vs iteration).
• Job queue + persistent results (job_id → downloadable artifacts).
• Instructor mode: create assignment, expected output (with tolerance), auto-grade.
• OpenAPI endpoints for integration with RANTAI Diva / LMS.

---

Input Formats (examples)

1. Augmented matrix (CSV)

Single CSV file: each row a1,a2,...,an,b (n columns for A, last column is b). 3,2,-1,1 2,-2,4,-2 -1,0.5,-1,0

2. Root finding (JSON)

{
  "method": "newton",
  "function": "x**3 - 2*x - 5",
  "derivative": "3*x**2 - 2",
  "initial_guess": 2.0,
  "tolerance": 1e-6,
  "max_iter": 50
}

Output (example JSON)

{
  "job_id": "uuid-123",
  "method": "newton",
  "converged": true,
  "final_x": 1.4142135624,
  "iterations": [
    {"k": 0, "x": 1.0, "f": -1.0, "error": 1.0},
    {"k": 1, "x": 1.4167, "f": 0.0069, "error": 0.4167}
  ],
  "downloads": {
    "csv": "/api/metnum/jobs/uuid-123/result.csv",
    "pdf": "/api/metnum/jobs/uuid-123/report.pdf"
  }
}

---

Tech stack (suggested)

• Frontend: Next.js + Plotly / Chart.js

= Backend: FastAPI (Python)

• Numerics: NumPy, SciPy, SymPy (safe parsing)

• Queue: Celery / RQ or FastAPI BackgroundTasks for MVP

• Storage: PostgreSQL (metadata) + S3/MinIO (artifacts)

• Auth: SIWE / wallet-based (optional instructor API keys)

---

Getting started (MVP)

1. Clone repo

2. python -m venv venv && source venv/bin/activate

3. pip install -r requirements.txt

4. Run FastAPI backend (see example server files)

5. Push sample CSV/JSON → test with POST /metnum/jobs

---

Contributing & Roadmap

Week 0–1: core endpoints + Gauss & Newton

Week 2: plotting + downloads

Week 3: instructor assignment & autograde

Week 4: extra methods & polish

---

License

MIT © 2025 RANTAI / ELPEEF