aboutsummaryrefslogtreecommitdiff
path: root/src/post-refinement.c
AgeCommit message (Collapse)Author
2015-07-17partialator: Add --no-prThomas White
2015-07-17Move post refinement multiprocessing stuff to post-refinement.cThomas White
2015-07-08Fix PR gradientsThomas White
D'oh! This must have crept in during a big tidyup before a commit
2015-06-19Remove old code/formattingThomas White
2015-05-19partialator: Show initial and final residuals (including free)Thomas White
2015-05-19Add free residual and convergence checkThomas White
2015-05-19partialator: Add cross-validation pobs/pcalc graphThomas White
2015-05-19Fix R and div gradientsThomas White
2015-05-19Refuse to refine any crystal with fewer data points than parametersThomas White
2015-05-19Refuse all nan shiftsThomas White
2015-05-19p gradientsThomas White
2015-05-19More debugging / logarithm stuffThomas White
2015-05-19Gradients for logarithmic scalingThomas White
2015-05-19Remove debugging stuffThomas White
2015-05-19partialator: Honour --no-scaleThomas White
2015-05-19Don't choke on negative scaling factorsThomas White
2015-05-19Refine logarithmicallyThomas White
2015-05-19Pass verbosity through to pr_iterate()Thomas White
2015-05-19Debugging stuffThomas White
2015-05-19Refine B factor as wellThomas White
2015-05-19Minimise Ipart-GpIfull/L instead of Ipart-pIfull/GLThomas White
It's about a million times easier
2015-05-13Debugging stuffThomas White
2015-05-13Fix guide_dev()Thomas White
2015-05-13Show shifts when being verboseThomas White
2015-05-13Refuse to make OSFs negativeThomas White
2015-05-13Change order of gradient calculationThomas White
2015-05-13Fix totally bogus convergence criterionThomas White
2015-05-13Show number of reflections used when being verboseThomas White
2015-05-13partialator: Move scaling calculation into PR properThomas White
2015-05-06Remove crystal backup stuffThomas White
2015-05-06Remove unused revert_crystal()Thomas White
2015-05-06partialator: Make the gradients indirect, for runtime choice of parameters ↵Thomas White
to refine
2015-04-20Factorise dr/da part of gradient calculation for PR and predictionThomas White
2015-04-20Make panel assignments invariant during prediction- and post-refinementThomas White
Reflections appearing and disappearing are problematic when trying to do a least-squares refinement. Therefore, assume that reflections stay on panel and keep them under consideration even if their partialities go to zero (i.e. they drift off Bragg). This should stabilise both refinements, and simplifies quite a lot of code. Collateral "damage": the old "select_intersection()" is now gone.
2015-03-13Move solve_svd() to utilsThomas White
2014-10-14Final gradients for SCGaussianThomas White
2014-10-06WIP on SCGaussian gradientsThomas White
2014-10-02div and R gradients for SCSphereThomas White
2014-09-30Cell vector gradients for SCSphere, plus general rationalisationThomas White
2014-09-25Remove "sphere", "thin" and "gaussian" partiality models, add "scgaussian"Thomas White
2014-09-25WIP on gradients for scsphereThomas White
2014-06-27Add Thin Ewald Sphere modelThomas White
2014-06-26Rationalise counting of filtered eigenvaluesThomas White
2014-06-05partialator: Use reciprocal scale factors insteadThomas White
2014-05-26partialator: Restore minimum number of observations for PRThomas White
2014-05-23Fix flagsThomas White
2014-05-23Remove old errorThomas White
2014-05-23Simplify criteria for scaling, merging and PRThomas White
2014-05-23Outlier rejection and new convergence criterion for scalingThomas White
2014-05-22partialator: Recalculate partialities after reverting crystalThomas White
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-02 14:59:12 +0000