[us-commits] [ehb54/ultrascan3] b35b1e: extrap_c0 regplot: fix reciprocal display, add fit...

[emre brookes]

  Branch: refs/heads/ehb54-issue-920
  Home:   https://github.com/ehb54/ultrascan3
  Commit: b35b1eb1794552dce22906c5116170db2b3b6e31
      https://github.com/ehb54/ultrascan3/commit/b35b1eb1794552dce22906c5116170db2b3b6e31
  Author: ehb54 <brookes at uthscsa.edu>
  Date:   2026-07-13 (Mon, 13 Jul 2026)

  Changed paths:
    M us_somo/develop/include/us_hydrodyn_saxs_iqq_extrap_c0_regplot.h
    M us_somo/develop/src/us_hydrodyn_saxs_iqq_extrap_c0.cpp
    M us_somo/develop/src/us_hydrodyn_saxs_iqq_extrap_c0_regplot.cpp

  Log Message:
  -----------
  extrap_c0 regplot: fix reciprocal display, add fit stats, clearer intercept error bar

The per-q regression viewer previously plotted the reciprocal/virial fit in
the wrong axis: it drew the c/I data points but used the inverted output
intercept (I0 = 1/u) as the fit line, so line and points disagreed. Capture
the fit in its plotted axis instead -- additive in I/c, reciprocal/virial in
c/I -- and pass the model + quadratic coefficient to the viewer:

  - reciprocal/virial plots now show c/I vs c with the correct intercept u and,
    for the 2nd-order virial, the fitted parabola (sampled) rather than a line;
  - the results line reports u(c=0) and the derived I(0) = 1/u with propagated SE;
  - the y-axis is labelled c/I for the reciprocal models.

Also add a per-q goodness-of-fit readout (R^2, and reduced chi^2 when the data
carry error bars), and improve the extrapolated-point error bar visibility: the
c=0 reference line is now white (was red, hiding the red intercept error bar
drawn over it) and the intercept error bar is drawn a bit wider.

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