stats op• Data kinds: matrix → matrix
• Call: import mathops; mathops.stat_correlation(x) (or opsmath.get("stat_correlation"))
Pearson correlation matrix of `(N, D) observations → (D, D)`.
Same orientation as :func:stat_covariance (rows = observations).
Entries are clipped to `[-1, 1]` (floating-point can overshoot by an
ulp), the diagonal is exactly `1` and the matrix exactly symmetric by
construction.
**A constant column raises `ValueError`** (naming the column) instead of
yielding NaN: correlation with a zero-variance variable is mathematically
undefined (0/0), and a NaN that surfaces three ops downstream is the
classic zero-division bug family this module fails closed against. Drop or
perturb the constant column deliberately if that is what you mean.
HALCON: no public tuple operator (see :func:stat_covariance).
Every mathops op validates its input before computing (nothing slips through silently):
• **complex input raises ValueError** — coercing to float64 silently discards the imaginary part (numpy only emits a ComplexWarning and returns a plausible-looking wrong real number). State .real/.imag/abs() explicitly, or use complexops, which handles complex data.
• **masked arrays with masked elements raise ValueError** — the implicit conversion that peels off the mask and uses the raw values underneath is refused. Say explicitly whether to fill or to drop.
• **NaN/Inf raises ValueError on every input** (refused with the count stated — it propagates through the whole result).
• Shapes are strict: 1-D and 2-D are never implicitly promoted or broadcast (a matrix in a vector slot, or a vector in a matrix slot, raises ValueError; reshape explicitly).
• Size cap: ops that take a matrix, and the stat_histogram bins, raise ValueError beyond mathops.MAX_ELEMENTS (2^26 ≈ 67 million elements).
• measurement_uncertainty — 計測の不確かさと校正の知識 — 「測れている」を主張するために
• Sample-data catalog (download URLs / licences) — 2-D uses skimage.data (BSD/public domain) plus synthetic images; 3-D lists download URLs for real data sources (Stanford, PDS, …).
• Operator provenance and references — the sources of the research/methods this op family came from.
• The canonical algorithm (author, year) and its uses are named in the family usage guide above.
• math_metrology — py -3.11 examples/math_metrology.py
• poc_ct_fidelity — py -3.11 examples/poc_ct_fidelity.py
matrix as input)mat_solve · mat_lstsq · mat_svd · mat_eigh · mat_pinv · mat_cond · stat_covariance
stats)stat_describe · stat_histogram · stat_covariance · stat_zscore
*Provenance: mathops.py — MATH operator registry. This per-op note is generated by tools/opdocs.py md (do not hand-edit).*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.