Python rewrite of the Urb/Homemaker stack
Find a file
Bruno Postle 7d7994e7a3 Phase 7 §12.1: larger-than-house benchmark maple-court + baseline (leu.1)
26 programme entries / 52 rooms / 3 storeys (~1015 m2 internal). Mirrors
harbor's adjacency-to-c load + secondary adjacencies; room codes avoid the
generic c/o/s leading-letter trap. Staged adjacency-aware baseline (20000
evals, URB_NO_OCCLUSION=1): 145/158/152 fails, mean 151.7; all native
re-score OK. Best (145) saved as generated.dom. Recorded in DESIGN.md §12.1.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-19 14:00:21 +01:00
.beads Phase 7 §12.1: larger-than-house benchmark maple-court + baseline (leu.1) 2026-06-19 14:00:21 +01:00
.claude Scaffold homemaker-py with validated geometry port 2026-06-10 20:50:20 +01:00
examples Phase 7 §12.1: larger-than-house benchmark maple-court + baseline (leu.1) 2026-06-19 14:00:21 +01:00
experiments Phase 6 §11.7: adjacency-aware lift + secondary adjacencies (ld5) 2026-06-19 11:47:40 +01:00
src/homemaker_layout Phase 6 §11.7: adjacency-aware lift + secondary adjacencies (ld5) 2026-06-19 11:47:40 +01:00
tests Phase 6 §11.7: adjacency-aware lift + secondary adjacencies (ld5) 2026-06-19 11:47:40 +01:00
.gitignore bd init: initialize beads issue tracking 2026-06-11 23:27:11 +01:00
AGENTS.md bd init: initialize beads issue tracking 2026-06-11 23:27:11 +01:00
CLAUDE.md Fix parity gap: oracle.py must run with URB_NO_OCCLUSION=1 2026-06-17 18:40:56 +01:00
DESIGN.md Phase 7 §12.1: larger-than-house benchmark maple-court + baseline (leu.1) 2026-06-19 14:00:21 +01:00
pyproject.toml Add core_divide, core_undivide, level_fix operators; wire reqs to mutate() 2026-06-14 16:10:20 +01:00
README.md Fix parity gap: oracle.py must run with URB_NO_OCCLUSION=1 2026-06-17 18:40:56 +01:00

homemaker-layout

Programme-driven building-layout search over slicing trees. A clean-room Python successor to the Perl Urb project, intended to eventually be 100% Python.

Why a rewrite

Urb represents a building as a binary slicing tree where room sizes are derived top-down from division ratios. That makes room area an emergent property of every cut above it, which:

  • gives the genome low locality (a cut near the root rescales every descendant),
  • makes target room sizes nearly impossible to hit, so the gaussian size penalty dominates fitness, and
  • defeats crossover (transplanted subtrees lose their proportions).

homemaker inverts this: leaves carry target dimensions from the programme and division ratios are solved bottom-up for a fixed topology. The evolutionary search then only explores topology + types + adjacency.

Phase plan

  1. Solver experiment: port Urb's geometry, re-solve ratios from programme targets, score the result against the original via the Perl oracle.
  2. Native Python fitness (retire the Perl oracle).
  3. Memetic search (current): canonical slicing genome + high-locality operators + Nelder-Mead inner loop.

Layout

  • src/homemaker_layout/dom.py — read/write Urb .dom YAML into a Node tree.
  • src/homemaker_layout/geometry.py — faithful port of Urb's top-down geometry.
  • src/homemaker_layout/programme.py — parse patterns.config space requirements.
  • src/homemaker_layout/solver.py — bottom-up ratio solve (scipy).
  • src/homemaker_layout/fitness.py — native Python fitness evaluator.
  • src/homemaker_layout/fitness_cmd.pyhomemaker-fitness CLI (drop-in for urb-fitness.pl).
  • src/homemaker_layout/graph.py — leaf-adjacency graph for programme-driven checks.
  • src/homemaker_layout/genome.py — topology genome: base-floor tree + per-storey deltas.
  • src/homemaker_layout/operators.py — high-locality mutation and subtree crossover.
  • src/homemaker_layout/innerloop.py — ratio optimisation inner loop (Nelder-Mead / CMA-ES).
  • src/homemaker_layout/driver.py — memetic search outer loop.
  • src/homemaker_layout/evolve.pyhomemaker-evolve CLI entry point.
  • src/homemaker_layout/oracle.py — legacy Perl shim, kept for cross-validation only.