v3.5.5

3.5.5 – 2025-02-05

General

• Option for stricter run-time input type checking (#388, thanks @ggalloni)
• Detect and fix incomplete last lines when resuming or minimizing from existing runs (#306, #378)
• Added functions module and refactored some numerical functions into it
• Tidier numpy2 outputs
• Other minor fixes (thanks @mgerbino, @ark93-cosmo, @MariaLuisaGallavotti, @ewanchamberlain)

Cosmology

CLASS

• Regression: no multiple local installs after v3.3 (see #404)

v3.5.4

• Allow classes to have both yaml and class attributes as long as no duplicate keys
• Added get_modified_defaults() class method to cobaya components to dynamically set/modify defaults

3.5.3

• Added cobaya-run --allow-changes to resume or minimize after changing input settings (turns off checking)
• Updates for numpy 2 and other compatibility/deprecation fixes
• Fixes #357, #358, #360, #361, #362, #368
• Added _fast_chi_squared method to base class InstallableLikelihood

v3.5.1

• Added DESI 1yr BAO data and SN from Pantheon Plus, DESY5 and Union3 (thanks DESI team, @adematti, @rubind, @WillMatt4 and @rodri981)
• Use of vector parameters now documented (PR #191; inspired by @lukashergt, thanks!)

v3.5

General

• Updated UGE sample job submission template (for cobaya-job-run and cobaya-grid-run)
• Clarify log feedback when using oversample_thin
• Fixed #345, #346, #347, #348

Grid scripts

• Support for running grids of models, including grid getdist, PDF tables, importance sampling, minimization (almost all features of CosmoMC grid now available in Cobaya). See the new doc pages.

Minimization

• Support for iminuit minimizer and getting best-fits for all mpi runs (#332, thanks @ggalloni)
• Support for minimzation with an importance-sampled input yaml config

v3.4.1

• Fixed a critical packaging bug (thanks Biagio De Simone!)
• Updated CLASS to 3.2.1 (no more Cython version restriction)

v3.4

• A couple of major bugfixes (parallel MCMC chains getting stuck; issues when resuming).
• Easier loading and MPI sharing of SampleCollections and sampler products, including native GetDist conversion.
• Python 3.12 support.
• Added 1-d grid LSS likelihood and BAO-only ELG and QSO (@msyriac)
• Some CLASS-related minor updates.

v3.3.2

• Class instance methods can now be used as external likelihoods.
• Added planck_2018_lowl.EE_sroll2 low-E Planck likelihood
• CAMB now supports using sigma8 as an input parameter (thanks @tilmantroester); note this is inefficient compared to A_s-based sampling)
• cobaya-install cosmo now installs set of Planck NPIPE (PR4) python likelihoods
• CAMB minimum version now 1.5 (with important bug fix in CAMB 1.5)
• Various bug fixes/additional warnings

v3.3.1

typo

v3.3

Some highlights:

• Min Python version is now 3.8!
• Components can now use a package_install settings to specify if they should be installed from pip, github or URL when cobaya-install is run.
• MCMC: added tempered sampling.
• Planck:
  • Replaced default planck_2018_lowl.EE and planck_2018_low.TT with native versions, and using GitHub-hosted clik.
  • Added latest PR4 NPIPE fully-Python likelihoods