fullseye

spc_mt_unit_space — SPC mt op

• データ種: matrix → table
• 呼び出し: import fullseye as fs; fs.ledger.spc_mt_unit_space(data) (実装を直接呼ぶなら import spc; spc.spc_mt_unit_space(data)、台帳から引くなら opsspc.get("spc_mt_unit_space"))

使い方

Build an MT-method unit space from known-good observations.

data is a 2-D array of shape (n, p) — n observations of p features, all of which must be normal product. The unit space is the triple (mean, std, inv_corr): per-feature mean and ddof=1 standard deviation, and the inverse of the correlation matrix of the standardised data. Feed it to :func:spc_mt_distance to score new observations.

Returns a dict with mean, std, corr and inv_corr arrays, the unit space’s own md per row, md_sq_mean, the exact expected value md_sq_mean_exact = (n-1)/n, the counts n and p, how many features had no spread (flat_features), and whether the correlation matrix was singular enough to need a pseudo-inverse (pseudo_inverse).

The mean of md**2 over the unit space equals (n-1)/n exactly — the quantitative form of “a unit space averages a distance of one”. It reaches 1 only as n grows; the deficit is exactly 1/n.

Raises ValueError: a non-2-D / empty data, fewer than two observations (no spread to standardise by), zero features, or non-finite input.

参考(サンプルデータ・文献)

• 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。

実行できる例(この op を実際に呼ぶ検証済みサンプル)

• poc_mt_hidden_fault — py -3.11 examples/poc_mt_hidden_fault.py
• poc_spc — py -3.11 examples/poc_spc.py

型が繋がる次の op(table を入力に取れる)

msa_anova_table · msa_gauge_rr · msa_bias_linearity · msa_attribute_agreement · gum_standard_uncertainty · gum_propagate · gum_expanded · gum_monte_carlo

同カテゴリ(mt)

spc_mt_distance · spc_mt_sn_ratio

---

Provenance: spc.py — SPC 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.

This site is open source. Improve this page.