bruno/homemaker-layout
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
24 commits 1 branch 0 tags 5.4 MiB
Commit graph

5 commits

Author SHA1 Message Date
Bruno Postle
bc61f8cb73 Bake-off: CMA-ES confirmed as inner-loop optimiser (homemaker-py-d0s) 
4-way comparison (NM / CMA-ES / compass / compass-ms) over 3 corpus files ×
3 seeds at budget 200, cold-start, URB_NO_OCCLUSION=1. CMA-ES wins on
batch-efficiency (18 oracle calls vs 200 for NM, 12x speedup on Perl startup
amortisation per §4.6) with acceptable quality (x1.41 @200 vs NM's x1.56).
Compass stalls on narrow-valley landscapes and introduces fail regressions.
NM flagged as Phase 3+ candidate once native fitness removes oracle call overhead.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-06-13 09:47:15 +01:00
Bruno Postle
5f0c159112 Re-baseline under URB_NO_OCCLUSION: new reference gains, DESIGN §4.7 
Corpus: all 35 scores shift (x1.0-1.24, daylight pinned), one expected
failure-set change (458aa8b8 +2 crinkliness), oracle ~8% faster batched.
New deterministic-seed reference gains become the accept_innerloop bars:
x1.63 / x1.70 / x1.68 at budget 400, ~35 oracle calls per topology.
urb-evolve respects the flag by construction (in-process fitness reads
ENV at call time). Old flag-off numbers kept in DESIGN as historical.

Closes homemaker-py-gp2 (Urb-side patch lives in /home/bruno/src/urb,
uncommitted there pending review).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
 2026-06-12 10:31:38 +01:00
Bruno Postle
d4266f46dc Descope occlusion/daylight: disable in Urb, port simple crinkliness only 
Strategy decision (Bruno): occlusion is orthogonal to building a scalable
optimiser. Instead of porting Sun/Occlusion/CIESky, disable it in Urb
behind an env flag (daylight -> 1, illumination factor -> 1 so crinkliness
is unweighted external wall area / floor area). Python occlusion rebuild
deferred until optimisation is fully native.

Tracker: new homemaker-py-gp2 (flag + re-baseline) gates gnw/way/uxz;
homemaker-py-2g5 re-scoped to the post-Phase-5 Python rebuild (P4).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
 2026-06-12 08:31:20 +01:00
Bruno Postle
8efe25601f Update DESIGN.md with findings from source review of Urb 
- §4.5: slide() re-randomises cuts (no fine-tuning operator exists) —
  strengthens the geometry-headroom explanation
- §4.6: throughput arithmetic shows native fitness effectively gates
  topology search at scale; oracle suffices for Phase 1 + small Phase 2
- §5: new decision 6 — Lamarckian warm-start of the inner loop; penalty
  reshaping must preserve inner-loop cliff protection
- §6: native-fitness port scope expanded (occlusion/daylight subsystem,
  cost denominator, structural failures, missing-space failure stacking,
  two-phase graph build, has_vertical_connection stub)
- §7: Phase 2 re-scoped as small-scale proof with budgets in oracle
  evaluations; Phase 1 gains warm-vs-cold + optimiser bake-off experiments
- §8: risks updated (reshaping tension, height DOF, confirmed t3 bug)
 2026-06-12 00:34:51 +01:00
Bruno Postle
7fccc05c0d Add comprehensive DESIGN.md capturing this session's findings 
Self-contained design + plan structured for breaking into bd (beads) tasks:
domain constraints that fix the slicing representation; what was built;
full empirical record (geometry port validated 35/35; area-proxy solver
falsified; perpendicular artifact resolved via equal-offset cuts; full-fitness
frozen-topology optimisation validated with 24-67% headroom; 0.5^n cliff);
validated memetic architecture; component plan; phased roadmap; risks/open
questions; repro steps; gotchas.

Oracle throughput measured: ~0.99s/dom batched vs 1.65s single (assessment-
dominated). urb-fitness.pl batches many doms per call, so native fitness is a
later speed/scale optimisation, not a gate; favour population/batch optimisers
and prototype the search on the batched oracle first.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
 2026-06-10 22:45:44 +01:00