fullseye

anomaly_pixel_auroc — SPC anomaly_eval op

使い方

Pixel-level AUROC of anomaly score maps against ground-truth defect masks (table).

scores and masks are (n, H, W) (or lists / single 2-D maps); higher score = more anomalous. Exact Mann–Whitney statistic with mid-ranks for ties: the probability that a random defect pixel outscores a random good pixel, ties counting 1/2. Returns auroc, n_pos, n_neg.

Raises ValueError on empty / mismatched / non-finite input, non-binary masks, or when the masks contain no defect pixel or no good pixel (AUROC undefined).

参考(サンプルデータ・文献)

実行できる例(この op を実際に呼ぶ検証済みサンプル)

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(anomaly_eval)

anomaly_image_auroc · anomaly_pro · anomaly_region_overlap · anomaly_pixel_f1_max · anomaly_image_f1_max · anomaly_operating_point


Provenance: anomeval.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.