Merged
Conversation
Add Constraint ABC with LinearConstraint and FunctionalConstraint implementations. ParameterSpace now accepts an optional constraints tuple, validates constraint param references at construction time, and filters infeasible points from grid_points(). RandomSearchOptimizer uses rejection sampling with a 1000-retry limit. 22 new tests. Closes #113
feat(gds-psuu): declarative parameter space constraints
Add Objective ABC with SingleKPI and WeightedSum implementations, separating "what to optimize" from "how to optimize". SweepResults gains best_by_objective() for multi-KPI ranking. Sweep accepts an optional objective parameter. Migrate BayesianOptimizer from unmaintained scikit-optimize to optuna (TPE sampler, ask/tell API). Update optional dependency accordingly. 15 new tests (4 Bayesian skipped without optuna installed). Closes #114
feat(gds-psuu): composable objectives + migrate to optuna
Add sensitivity/ subpackage with pluggable Analyzer ABC and two implementations: - OATAnalyzer: one-at-a-time perturbation (mean_effect, relative_effect) - MorrisAnalyzer: elementary effects screening (mu_star, sigma) SensitivityResult provides ranking() and to_dataframe(). Both analyzers compose with existing Evaluator and ParameterSpace primitives — no new dependencies. 12 new tests. Closes #112
feat(gds-psuu): sensitivity analysis framework
Introduce composable Metric (per-run scalar) and Aggregation (across-run reducer) types that compose into KPIs, aligning with the simulation glossary hierarchy. KPI now accepts either legacy fn or metric+aggregation pair. Built-in metrics: final_value, trajectory_mean, max_value, min_value Built-in aggregations: mean, std, percentile, probability_above/below Evaluator populates per-run distributions for metric-based KPIs. Closes #118
Overview, getting started guide, concept hierarchy (Metric/Aggregation/KPI), parameter spaces reference, optimizer guide, and API reference pages. Adds psuu section to mkdocs nav and landing page package table.
feat: composable Metric/Aggregation primitives for KPI
New features: composable Metric/Aggregation/KPI, parameter constraints, sensitivity analysis (OAT + Morris), composable objectives, optuna migration.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Promotes
devtomainfor gds-psuu v0.2.0 release. All changes are scoped to thegds-psuupackage.New Features
Metricextracts per-run scalars,Aggregationreduces across runs,KPIcomposes both. Built-in factories:final_value,trajectory_mean,max_value,min_value,mean_agg,std_agg,percentile_agg,probability_above,probability_below. Per-run distributions tracked inEvaluationResult.LinearConstraintandFunctionalConstraintfor feasible region definition. Grid search filters infeasible points; random search uses rejection sampling.SingleKPIandWeightedSumobjective types.SweepResults.best_by_objective()for multi-KPI optimization.scikit-optimizewithoptuna>=4.0for Bayesian optimization.OATAnalyzer(one-at-a-time) andMorrisAnalyzer(elementary effects) for parameter importance screening.Stats
Test plan
uv run --package gds-psuu pytest packages/gds-psuu/tests -v— 127 passeduv run ruff check packages/gds-psuu/— all checks passeduv run ruff format --check packages/gds-psuu/— all formatted🤖 Generated with Claude Code