glotaran/pyglotaran: v0.3.2

pyglotaran 0.3.2 Release Notes The first release of pyglotaran ready for a wider audience was v0.3.0, this is the first proper patch release to follow, fixing some small but important issues. Highlights Compatible with NumPy 1.20. Using less memory during optimization (during Result creation). Saved result datasets are smaller in size. Changelog Added compatibility for NumPy 1.20 and raised minimum required NumPy version to 1.20 (#555) Fixed excessive memory consumption in result creation due to full SVD computation (#574) Added feature parameter history (#557) Moved setup logic to setup.cfg (#560) Correction to setup.cfg to satisfy PyPI and version bump 0.3.1 -> 0.3.2 Version numbering A glitch in the matrix resulted v0.3.1 having a visually unappealing front page on PyPI. To fix this, a new release had to be drafted which is identical to v0.3.1, other than a configuration file used in rendering said front page. For all intents and purposes both versions are the same, released on the same date, and thus share the same Highlights and Changelog (w.r.t v0.3.0) Usage notice This software package is made available as an early access release, to the advantage of the (scientific) community who wishes to make use of it before it has fully matured, but without any warranties. Please refer to the usage notice in the README. For a modest collection of examples, see the pyglotaran_examples repository. For examples on how to do plotting of results see the pyglotaran_extras repository.Identical to the v0.3.1 release, other than a configuration file used in rendering the pypi project page