plan op• データ種: なし → design(引数だけで決まる op —— 画像やデータの入力を取らない)
• 呼び出し: import fullseye as fs; fs.ledger.doe_lhs(n=10, factors=3, seed=0, maximin=0, centered=False) -> 'dict' (実装を直接呼ぶなら import doe; doe.doe_lhs(n=10, factors=3, seed=0, maximin=0, centered=False) -> 'dict'、台帳から引くなら opsmultivar.get("doe_lhs"))
ラテン超方格計画(McKay, Beckman & Conover 1979)-> design。
> 以下の詳細説明は原文のままです —— 要約と見出しは訳出済み。
Each factor's range is cut into n equal strata and every stratum holds exactly one run, so each factor is sampled evenly however many factors there are (a random design of the same size leaves strata empty). Construction (McKay et al. 1979, section 2): for factor j draw an independent random permutation pi_j of 0..n-1 and put run i at u_ij = (pi_j(i) + U_ij) / n with U_ij ~ Uniform[0, 1) (centered=True uses the stratum mid-point, U = 1/2). Coded value = 2u - 1.
maximin=m (m >= 2) draws m independent hypercubes and keeps the one with the largest smallest inter-point distance (each candidate is still a Latin hypercube, so the stratum property is never traded away); 0 or 1 = one plain draw.
Parameters: n runs (2..1e6), factors (int or factor list, see module docstring), seed (int, reproducible), maximin (0..1000 candidates), centered.
Returns a design whose meta holds strata_ok (every factor has one point per stratum — the defining property, checked on the returned matrix) and min_distance (in the unit cube).
• design_of_experiments ファミリ ガイド
• サンプルデータ カタログ(DL URL / ライセンス) — 2-D は skimage.data(BSD/public)+ 合成、3-D は実データ源(Stanford/PDS 等)の DL URL。
• 演算子の来歴・参考文献 — この op 族の元になった研究/手法の出典。
• アルゴリズムの正典(著者・年)と用途は上記ファミリ使い方ガイドに記載。
• doe_plan_tour — py -3.11 examples/doe_plan_tour.py
• multivar_inspection_tour — py -3.11 examples/multivar_inspection_tour.py
• sensitivity_tour — py -3.11 examples/sensitivity_tour.py
• simcal_checks_tour — py -3.11 examples/simcal_checks_tour.py
design を入力に取れる)doe_decode · rsm_fit · doe_main_effects · sens_morris · sens_sobol
plan)doe_orthogonal_array · doe_fractional_factorial · doe_central_composite · doe_box_behnken
*Provenance: doe.py — MULTIVAR operator registry. この per-op ノートは tools/opdocs.py md が自動生成(手編集しない)。*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.