Script 'mail_helper' called by obssrc
Hello community,

here is the log from the commit of package python-arviz-stats for 
openSUSE:Factory checked in at 2026-10-26 10:39:58
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/python-arviz-stats (Old)
 and      /work/SRC/openSUSE:Factory/.python-arviz-stats.new.2252680 (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Package is "python-arviz-stats"

Mon Oct 26 10:39:58 2026 rev:5 rq:1392403 version:1.3.3

Changes:
--------
--- /work/SRC/openSUSE:Factory/python-arviz-stats/python-arviz-stats.changes    
2026-08-22 21:37:24.031915904 +0200
+++ 
/work/SRC/openSUSE:Factory/.python-arviz-stats.new.2252680/python-arviz-stats.changes
       2026-10-26 10:40:01.588188185 +0100
@@ -1,0 +2,45 @@
+Sun Oct 25 12:23:54 UTC 2026 - Dirk Müller <[email protected]>
+
+- update to 1.3.3:
+  * Add PSIS LFO-CV
+  * Accept coords in ci_in_rope, as its docstring documents
+  * Correct interval for pit evaluation
+  * Only build the summary's LaTeX repr when asked for
+  * Add bivariate histogram and hexbin statistics
+  * Add ELPD subclasses
+  * Add default `ci_kind` handling in `ci_in_rope`
+  * Miscelaneous docstrings fixes
+  * Update citation information
+  * More miscellaneous docstring fixes
+  * Fix ddof in `summary` std computation
+  * Hexbin and histogram2d: Use adaptive binning
+  * Fix histogram with array-like bins
+  * Allow hdi with prob=1
+- update to 1.3.2:
+  * ### Maintenance and bug fixes
+  * Respect ci_kind in summary for the stats and all kinds
+  * Remove deprecated inplace in sort_values
+  * Write the bandwidth into the kde gufunc output
+  * Ensure correct order for bounds in circular std
+  * Add PyMC and Bambi support for `loo_subsample`
+  * Fix `p_loo` calculation in sub-sampled PSIS-LOO-CV
+  * Fix rounding of `subsampling_dse` in `compare()`
+  * Honour circular in the Numba KDE
+  * Fix _mcse_sd to use ESS of squared deviations
+  * Fix custom limits for circular KDE
+  * Fix Kaplan–Meier estimates for integer event times
+  * Fix covariance and split steps in `loo_moment_match`
+  * Fix `num_samples` default in survival curves
+  * Fix bug in acc_balanced, and switch to ddof=1
+  * Fix `loo_subsample` pointwise output for multi-dim
+    observations
+  * Fix reloo SE, log weights, and input checks
+  * Fix jacobian term for p_loo
+  * ### Documentation
+  * Correct the BaseArray summary statistic docstrings
+  * Add weight_predictions to docs, add example
+- update to 1.3.1:
+  * Fix NaN in `loo_score` CRPS and simplify score calculations
+  * Fix truncated Cauchy combination: restore correct denominator
+
+-------------------------------------------------------------------

Old:
----
  arviz_stats-1.3.0.tar.gz

New:
----
  arviz_stats-1.3.3.tar.gz

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Other differences:
------------------
++++++ python-arviz-stats.spec ++++++
--- /var/tmp/diff_new_pack.2jvDRp/_old  2026-10-26 10:40:05.956370555 +0100
+++ /var/tmp/diff_new_pack.2jvDRp/_new  2026-10-26 10:40:05.957370597 +0100
@@ -17,7 +17,7 @@
 
 
 Name:           python-arviz-stats
-Version:        1.3.0
+Version:        1.3.3
 Release:        0
 Summary:        Statistical computation and diagnostics for ArviZ
 License:        Apache-2.0

++++++ arviz_stats-1.3.0.tar.gz -> arviz_stats-1.3.3.tar.gz ++++++
/work/SRC/openSUSE:Factory/python-arviz-stats/arviz_stats-1.3.0.tar.gz 
/work/SRC/openSUSE:Factory/.python-arviz-stats.new.2252680/arviz_stats-1.3.3.tar.gz
 differ: char 14, line 1

Reply via email to