portfolio_optimizer.py v1.2 — Disruptive Quantum-Inspired Portfolio Optimizer Features • Zero extra setup — single file (only NumPy/SciPy) • Real-world Markowitz optimization (50–200+ assets) with non-convex constraints • Quantum-inspired simulated annealing with proper risk-adjusted utility • Efficient Dirichlet + swap moves for fast global convergence • Full support for cardinality, L1 turnover, target return (penalty), weight bounds • Classical SLSQP baseline (convex cases only) • High-quality efficient frontier via risk_aversion sweep (classic bullet shape) • Professional visualization with highlighted optimal portfolio • Exports weights, statistics, and results to JSON/CSV Dependencies • Requires numpy (>=1.21.0), scipy (>=1.7.0) — standard in scientific Python • matplotlib (>=3.5.0) optional for --plot Intended for portfolio managers, quantitative analysts, and asset allocators requiring globally optimal solutions under complex real-world constraints — delivering superior performance where classical convex solvers fail, with a clear pathway to true quantum annealing/QAOA advantage. Real usage: python portfolio_optimizer.py --returns mu.csv --cov sigma.csv --risk_aversion 4 --cardinality 25 --turnover_limit 0.3 --quantum_annealing --plot python portfolio_optimizer.py --returns mu.csv --cov sigma.csv --target_return 0.12 --quantum_annealing --output results.json Made by Britt (2025) — MIT License