v2023.1

What's Changed

• Fix definition of norm_cdf psychometric function by @aforren1 in #41
• Add Thurstone scaling via paired comparisons by @hoechenberger in #46
• Ensure compatibility with latest from NumPy and xarray versions by @hoechenberger in #44 and #49

Code health & maintenance

• Adopt Black coding style (except for tests) by @hoechenberger in #45
• Switch from versioneer to setuptools-scm, and use GH Actions for CI by @hoechenberger in #48
• Prepare release by @hoechenberger in #51

New Contributors

• @aforren1 made their first contribution in #41

Full Changelog: 2019.4...2023.1

v2019.4

• Allow JSON serialization of the random number generator (if min_n_entropy stimulus selection is used)

v2019.3

• Allow to pass a prior when instantiating QuestPlusWeibull

v2019.2

• Allow passing a random seed via stim_selection_options keyword argument
• Better handling of stim_selection_options defaults (now allows to supply only a subset of options)

v2019.1

• Allow to pass priors for only some parameters (the remaining parameters will be assigned an uninformative prior)
• Add more docstrings, fix typo in usage example
• Test on Python 3.8

0.0.5 – Fifth alpha release

What's new:

• Allow retrieval of marginal posterior PDFs via QuestPlus.marginal_posterior
• Allow nan values in JSON output (violating JSON std but useful)

0.0.4 – Fourth alpha release

Merge pull request #17 from hoechenberger/json

NF: Add JSON dump and load, and support "equals" operator

0.0.3 – Third alpha release

Merge pull request #16 from hoechenberger/subclasses

NF: Add QuestPlusWeibull convenience class

0.0.2 – Second alpha release

Merge pull request #15 from hoechenberger/readme

DOC/PKG: Render readme nicely on PyPI,  add directories to .gitignore

0.0.1 – First alpha release

Merge pull request #14 from hoechenberger/readme

DOC: Update readme