jacscanomaly.CandidateQuality

class jacscanomaly.CandidateQuality(n_window, n_contrib, n_eff, peak_frac, rho1, longest_run)[source]

Per-candidate quality diagnostics derived from per-point chi-square improvement.

Parameters:

n_window

Number of points inside the local chi2 evaluation window.

n_contrib

Number of points with improvement above the configured threshold.

n_eff

Effective number of contributing points (participation-ratio style).

peak_frac

Fraction of total positive improvement carried by the strongest point.

rho1

Lag-1 autocorrelation of signed per-point improvements within the window.

longest_run

Longest consecutive run length among above-threshold contributing points.

__init__(n_window, n_contrib, n_eff, peak_frac, rho1, longest_run)

Parameters:

Return type:

None

Methods

__init__(n_window, n_contrib, n_eff, ...)

Attributes