Skip to content

Releases: jcm-sci/trade-study

v0.1.0 — Initial PyPI release

15 Apr 21:26
@jc-macdonald jc-macdonald
v0.1.0
96fc4ce

Choose a tag to compare

View all tags
v0.1.0 — Initial PyPI release Latest
Latest

Highlights

First public release of trade-study: multi-objective trade-study orchestration for Python.

Features

  • Core protocols: Observable, Direction, Simulator, Scorer, TrialResult, Annotation, ResultsTable
  • Design: Factor, FactorType, build_grid (full factorial, LHS, Sobol, Halton), screen (Morris), reduce_factors
  • Runner: run_grid (joblib parallel), run_adaptive (optuna NSGA-II)
  • Multi-phase orchestration: Phase, Study, top_k_pareto_filter
  • Scoring: score (CRPS, WIS, interval, energy, RMSE, MAE, coverage, Brier), coverage_curve
  • Pareto analysis: extract_front, pareto_rank, hypervolume, igd_plus
  • Bayesian stacking: stack_bayesian, stack_scores, ensemble_predict
  • I/O: save_results, load_results (NumPy .npz + JSON metadata)
  • Input validation for Factor
  • PEP 561 py.typed marker

Install

pip install trade-study[all]

Full changelog: https://github.com/jcm-sci/trade-study/blob/main/CHANGELOG.md

Assets 2
Loading