[us-commits] [ehb54/ultrascan3] d46047: extrap_c0 manual: explain expected negative per-q ...

[emre brookes]

  Branch: refs/heads/ehb54-issue-920
  Home:   https://github.com/ehb54/ultrascan3
  Commit: d460470454f4b259675608a096197802f9c3ae0d
      https://github.com/ehb54/ultrascan3/commit/d460470454f4b259675608a096197802f9c3ae0d
  Author: ehb54 <brookes at uthscsa.edu>
  Date:   2026-07-13 (Mon, 13 Jul 2026)

  Changed paths:
    M us_somo/somo/doc/manual/somo/somo_saxs_extrap_c0.html

  Log Message:
  -----------
  extrap_c0 manual: explain expected negative per-q R^2 on the regplot

Document that a negative R^2 in the per-q regression viewer is expected and
benign, not an error: R^2 is measured about the plotted line, which under GCV
/ fit-broadening uses a slope borrowed from neighbouring q, so at noise-
dominated low q it can fit worse than a flat mean (R^2 < 0) -- the regularization
working as intended; also the fit is inverse-variance weighted while R^2 is
unweighted. Advise judging on reduced chi^2 and the Guinier QC too.

Also fix a stale "Absolute-scale mode" reference in the same section: the
reference/intercept title note appears when the reference is spliced (merge
checkbox), after that single mode was decoupled into ref_scale + merge_ref.

Co-Authored-By: Claude Fable 5 <noreply at anthropic.com>



To unsubscribe from these emails, change your notification settings at https://github.com/ehb54/ultrascan3/settings/notifications