trade-study¶
Multi-objective trade-study orchestration: scoring, Pareto optimization, and Bayesian stacking.
For installation and quick-start examples, see the README.
Overview¶
trade-study provides a structured workflow for multi-objective
design-of-experiments studies:
- Define observables and design factors via lightweight Protocols
- Build experimental grids with Design utilities (full factorial, Latin hypercube, Morris screening)
- Run simulations across the grid with Runner
- Score posterior predictive accuracy with proper scoring rules (Scoring)
- Filter the Pareto front (Pareto)
- Stack models via Bayesian stacking (Stacking)
- Orchestrate multi-phase studies with Study
API Reference¶
| Module | Description |
|---|---|
| Protocols | Core types: Observable, Direction, Scorer, Simulator, etc. |
| Design | Factor, build_grid, screen, reduce_factors |
| Runner | run_grid, run_adaptive |
| Study | Phase, Study, top_k_pareto_filter, weighted_sum_filter, feasibility_filter |
| Scoring | score, coverage_curve |
| Pareto | extract_front, pareto_rank, hypervolume, igd_plus |
| Stacking | stack_bayesian, stack_scores, ensemble_predict |
| Visualization | plot_front, plot_parallel, plot_scores, plot_calibration |
| I/O | load_results, save_results |