- update method details (e1001ed31995e136d86022ece9579b25e8f8d6c9)
- add the lost flux cases for adam/lm (c6b3b8138f9954aa43ce4f3fe439aab97db51e5d)
- split gboed into ies and gboed based on the eig. upload all new runs (672c8cd2235411e7805d2145e694ed1fbfce4203)
- list methods toggles by family, and remove the scoring toggle (5cea6693a85412240451a7e04810d40b4ec9b359)
- split EKP methods into separate data storage, to enable consistent data handling (7965d977689d099687a4b495f4157645ac3aead2)
- treat NaNs as attempted and failed all the time (63643b4a09b6b76b5bdd33080c4581bac48e97da)
- remove 100% dead column artifacts (f7b2bbfb75889853e419183dd5811234b90a6460)
- add abc into UQ leaderboard, and remove the older HM/ABC runs from Opt leaderboard (f94e1f7eb8c210f383efa89b8e26f8e0ea04037d)
- added note on high-dim failures (042f1afbbe33ef5a1639625994890c52fa9269fb)
- add score-based adam with scaled alphas (c87009b0eb70a81ce26d895b3fe9d44dcc79c564)
- add link to CalibrationLeaderboard.jl (ac78a839206919563f8f3348f4f78393620b1a54)
- infrastructure for score-based methods (eb020957f8d8899857b6cb51085b22704f1b8bd6)
- data for score-based lm (a31db405c36ff7881d95823161d2f1ce8fc09baa)
- correct ordering/ranking to include failures (979ed02fb634278b6596b2f65d46eda0fd7597c4)
- change methods taxonomy (dc2991f092e8de5ffa9b85e39e87d5cd93d40939)
- change the taxonomy of the methods, from a list to a tag-system (280655e6828ac2d238d4992eb3236f0875edebf2)
- update ordering (c2c7b05d95b31655ff360d87d27a7249bd6e93f6)
- add a suitability table, (green = <3 x best effort and <20% failures) and remove "All" summary information (6aead36b149e5da6607fcf688fc057ba96f050a6)
- reorder suitability table (663f796aef4d9112b03264182f0ec6cea8d801f4)
- add readme installation steps (fd26f53a06158b68376061dcc49fc5e5c9ec14a5)
- add claude skills to manage the leaderboard, then apply them to the repository, divide a new UQ leaderboard (8e1be253e9811aff14b8ce03f4ab69b6ecbcf855)
- remove temp file (cb2395d8df49f06ab6f84e71cf503a2f40f8770f)
- add more ignore and readme (206a08333a25a955183e731e6e05a4549ffa486d)
- new skill updated (97aaef38ff26dd95a14199ffb475866fd9d72e41)
- Initial updates for new CES leaderboard for UQ results with quantiles 0.15,0.5,0.85 (245eff6db96adf6cb30941bccff15b123613bc55)
- improve plotting (27293d19cfc2b29b4eaef5916fb06b8b36df84ba)
- try minimal data files (f0772b8574668c28c9c55c1fa9ae46b2f514d226)
- smal notation (24c91df22a268b253367331de3aa1d7f1a351505)
- update minimal files with actually whitened metric (537b4b2fec1fbb4014c33630e676725f8ae2d050)
- add pngs via lfs (74637ab109c13c216e042b565290d06b82c97b35)
- pipeline for figures (82e4e6f6e0fe59fe4a075af549ba83310230be0c)
- figures for homepage (3220356d6…
