- 02 Feb, 2023 5 commits
-
-
Philipp Frank authored
User Experience: Fix `optimize_kl` minisanity plotting for runs with many parameters See merge request !821
-
Philipp Frank authored
-
-
Martin Reinecke authored
change from last to latest See merge request !839
-
Vincent Eberle authored
-
- 01 Feb, 2023 1 commit
-
-
Philipp Frank authored
don't open new fig in every iteration See merge request !838
-
- 27 Jan, 2023 1 commit
-
-
Jakob Roth authored
-
- 25 Jan, 2023 3 commits
-
-
Martin Reinecke authored
remove note mpi tasks in sample_list See merge request !836
-
Jakob Roth authored
-
Martin Reinecke authored
add missing "not" in changelog See merge request !837
-
- 24 Jan, 2023 1 commit
-
-
Jakob Roth authored
-
- 17 Jan, 2023 1 commit
-
-
Jakob Roth authored
-
- 13 Jan, 2023 8 commits
-
-
Gordian Edenhofer authored
Sampling: Do not raise on not converged samples See merge request !835
-
Gordian Edenhofer authored
Better Lanczos interface See merge request !832
-
Gordian Edenhofer authored
re: cg: Relax energy error criterium within Newton See merge request !833
-
Gordian Edenhofer authored
HarmonicSKI: Make pad symmetric See merge request !834
-
Philipp Frank authored
Operator: activate partial_insert by default See merge request !790
-
Philipp Frank authored
FIX: add missing sqrt, save real std See merge request !830
-
Philipp Frank authored
Implement refinement of HEALPix See merge request !809
-
Philipp Frank authored
-
- 12 Jan, 2023 20 commits
-
-
Gordian Edenhofer authored
Do not enforce a specific dimension where unnecessary (e.g. in `HEALPixChart`). However, do so where the computation either would fail or would produce bogus results if invoked with fewer or more dimensions.
-
Gordian Edenhofer authored
AbstractModel: Dynamically initialize attr See merge request !831
-
Gordian Edenhofer authored
-
Vincent Eberle authored
-
Vincent Eberle authored
-
Gordian Edenhofer authored
-
Gordian Edenhofer authored
-
Gordian Edenhofer authored
SciPy does not compute the energy, so at least we should not do worse than SciPy...
-
Gordian Edenhofer authored
-
Vincent Eberle authored
-
Vincent Eberle authored
-
Vincent Eberle authored
-
Gordian Edenhofer authored
-
Gordian Edenhofer authored
-
Gordian Edenhofer authored
Approximates tr(log(M)) using tr(log(T)) where T is the projection of M into the krylov subspace K(M, v) where v is a sample from the metric M (i.E. v = v_lh + v_pr where v_lh/v_pr are samples from the likelihood/prior metric, respectively. In addition, the projected sample is constructed by taking v_pr projecting out the subspace K(M,v) using its eigen-basis. This ensures that both, the prior dominated part of v and the part already covered by tr(log(T)) is projected out. Original author: Philipp Frank
-
Gordian Edenhofer authored
Drop unused code blocks and refactor the geomap energy computation.
-
Gordian Edenhofer authored
-
Gordian Edenhofer authored
Furthermore, add a remark on the necessity of sampling antithetically.
-
Gordian Edenhofer authored
-
Gordian Edenhofer authored
-