Issues

• Report-quality and colour-vision audit

  verify
  NaNIsaac-Research/ReflectoLearn#55 · Filter by author NaNIsaac opened on Oct 26, 2026

• verify: baseline of checks, CI matrix, Ruff exceptions and known failures

  verify
  NaNIsaac-Research/ReflectoLearn#54 · Filter by author NaNIsaac opened on Oct 26, 2026

• fit_policy: A1v critical-edge estimate includes the inner validation block

  NaNIsaac-Research/ReflectoLearn#50 · Filter by author NaNIsaac opened on Oct 25, 2026

• Beam-width prior 5-15 um makes the footprint term far weaker than measured curves show

  NaNIsaac-Research/ReflectoLearn#48 · Filter by author NaNIsaac opened on Oct 24, 2026

• refine.mcmc: emcee DESnookerMove mix is not detailed-balance invariant; posterior widths 4-8 % too narrow

  NaNIsaac-Research/ReflectoLearn#40 · Filter by author NaNIsaac opened on Oct 23, 2026
  1

• CI runs ruff 0.17.1 but uv.lock pins 0.15.15, so local ruff misses CI findings

  NaNIsaac-Research/ReflectoLearn#28 · Filter by author NaNIsaac opened on Oct 22, 2026

• refine stops short of the best fit on curves from its own forward model

  NaNIsaac-Research/ReflectoLearn#25 · Filter by author NaNIsaac opened on Oct 22, 2026
  3

• refine: tth_offset trades off against film SLD; with no offset in the data the SLD comes out up to 30% low

  bug
  NaNIsaac-Research/ReflectoLearn#14 · Filter by author NaNIsaac opened on Oct 21, 2026
  1

• example: validate XRR fitting on a small approved measured dataset

  NaNIsaac-Research/ReflectoLearn#2 · Filter by author NaNIsaac opened on Oct 19, 2026
  1