Skip to content

[T-052] Analytic benchmarks (branin_falsify, multi_mode_6d) and benchmark backend #103

Closed

Description

@CodewithJha

Task T-052 · Milestone M4 — Baselines + metrics + experiments · Size S · Priority P0 · Epic #4

Implements FR-SEL-002 · FR-SEL-003

Branch feat/t-052-analytic-benchmarks-and-benchmark-backend — one task = one branch = one PR (doc 32 §6)

Summary

Implement selection/benchmarks.py with the two functions of doc 07 §9 (known failure modes, ~0.7 % failure volume for multi_mode_6d) plus a BenchmarkBackend adapter so experiments can run analytic campaigns through the normal pipeline.

Context / motivation

Lets selection algorithms be tested statistically in milliseconds (doc 18 §4.4) and powers EXP-02.

Scope

In scope

  • Functions with GT mode labelling helper label_mode(u) -> {'A','B','C','D',None}.
  • backends/benchmark/backend.py producing a synthetic trace whose oracle evaluation yields the benchmark ρ (or a direct-ρ shortcut flagged in capabilities).
  • Benchmark spaces configs/scenario_spaces/branin_2d.yaml, multi_mode_6d.yaml.

Implementation guidance

  • Seeded noise from the run stream. Validate failure volume by QMC (≈ 0.7 % ± tolerance) in a test.
  • Standards: respect the layering contracts (doc 16 §2) and conventions in doc 34; no hard-coded tunables — thresholds/defaults come from typed settings (doc 27); no magic numbers (ruff PLR2004); public APIs typed (mypy strict) and documented.

Acceptance criteria

  • Volume check ≈ 0.7 % (documented tolerance).
  • Each mode's canonical point fails; nominal passes.
  • Tests listed below added and green in CI; coverage of new code meets the gate in doc 18 §3.
  • Docs, config reference and CHANGELOG Unreleased updated where behaviour, keys or interfaces changed.
  • Definition of Done (doc 32 §3) satisfied.

Tests required

  • tests/unit/selection/test_benchmarks.py.

Dependencies

Blocked by: #84 (T-050a), #48 (T-035a)

Blocks: #115 (T-057), #116 (T-058), #119 (T-059c)

Parallelization notes

Files / paths owned by this task — avoid editing other paths; shared files (pyproject.toml entry points, cli/main.py, api/routers/__init__.py, router.tsx, docker/compose.yaml, ci.yml) are append-only:

  • backend/src/roboprobe/selection/benchmarks.py
  • backend/src/roboprobe/backends/benchmark/backend.py
  • configs/scenario_spaces/branin_2d.yaml
  • configs/scenario_spaces/multi_mode_6d.yaml

Independent.

Definition of Done

Per doc 32 §3 and AGENTS.md §5: lint, format, mypy strict and import-linter green; tests added and passing; docs updated; PR template completed with Closes #<this issue>; Conventional Commit messages referencing T-052 (doc 16 §6); commits authored only by the repository owner identity (AGENTS.md §4); CI green; reviewer checklist (doc 32 §5.1) satisfied.

References

Design docs: doc 07 §9 · doc 18 §4.4 · doc 19 §4 · doc 10 §7.2

Requirements: FR-SEL-002 · FR-SEL-003 (traceability: doc 28)

Activity

  1. added
    area:selectionScenario selection strategies (BO, LSE, PROBE, baselines)
    area:evalEvaluation campaigns and success criteria
    priority:P0Critical path / blocks others
    milestone:M4M4 — Baselines + metrics + experiments
    taskChild task of a milestone epic
    on Nov 19, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    area:evalEvaluation campaigns and success criteriaarea:selectionScenario selection strategies (BO, LSE, PROBE, baselines)milestone:M4M4 — Baselines + metrics + experimentspriority:P0Critical path / blocks otherssize:S<= 1 daytaskChild task of a milestone epictype:featureNew functionality

    Projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions