leonarski_f
100fe7b7e7
Build Packages / build:rpm (ubuntu2404_nocuda) (push) Successful in 25m0s
Build Packages / build:rpm (rocky8_nocuda) (push) Successful in 26m42s
Build Packages / build:rpm (rocky8_sls9) (push) Successful in 27m7s
Build Packages / build:rpm (ubuntu2204_nocuda) (push) Successful in 28m25s
Build Packages / build:rpm (rocky9_nocuda) (push) Successful in 29m44s
Build Packages / build:rpm (rocky9_sls9) (push) Successful in 32m14s
Build Packages / build:rpm (rocky8) (push) Successful in 24m39s
Build Packages / build:rpm (ubuntu2404) (push) Successful in 23m52s
Build Packages / build:rpm (ubuntu2204) (push) Successful in 25m19s
Build Packages / Generate python client (push) Successful in 23s
Build Packages / XDS test (durin plugin) (push) Successful in 20m42s
Build Packages / Create release (push) Skipped
Build Packages / build:rpm (rocky9) (push) Successful in 27m2s
Build Packages / Build documentation (push) Successful in 1m23s
Build Packages / DIALS test (push) Successful in 31m5s
Build Packages / XDS test (JFJoch plugin) (push) Successful in 14m55s
Build Packages / XDS test (neggia plugin) (push) Successful in 13m7s
Build Packages / Unit tests (push) Successful in 2h14m40s
PixelRefine is now an intensity-only operation: geometry is fixed (refined upstream by XtalOptimizer) and the only objective is the factored per-reflection likelihood (FACTORED_MODEL.md Terms 1+2) - measured per-resolution profile width R1 plus one Fisher-weighted intensity/scaling residual per reflection, fitting the per-image scale G and B. Validated on crystal 2 (fixed_master.h5 as stills, 1.7 A): CC1/2 84-92%, CCref 77-92%, flat - reproduces the env-flag prototype and matches the rotation path from the stills path. Removed: - the per-pixel ShoeboxResidual loss and PixelResidual cost functor; - all in-PixelRefine geometry refinement (orientation/cell/beam/distance/R), the regularised-orientation LSQ, signal-weighting, and the global sweep; - Term 3 (per-spot recentring) - a confirmed no-op on both crystals; - the diagnostic scaffolding (covariance, centroid, adaptive_R1) and the PR_* env knobs + stderr dumps in IndexAndRefine; - the PredictImage/ChiSquaredImage renderers and the entire viewer PixelRefine window/table/params + worker bindings + shoebox overlay. The sweep box-integrator background median became mean (consistency) by virtue of removing the sweep. METHODS.md rewritten for the current model; findings recorded in FINDINGS-2026-06.md. Net -2200 lines. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>