Colour 0.4.3 is available!

The colour-science Developers are pleased to announce the release of Colour 0.4.3!

We worked on many optimisations around colour quality metrics that should result in overall performance increase, thanks again to Tucker Downs (@tjdcs) for the heavy lifting!

@charliermarsh's ruff is now used as a replacement for:

• flake8

• pydocstyle

• pyupgrade

We also replaced mypy with pyright for performance reasons. We took that opportunity to simplify our type annotations.

The colour.XYZ_to_RGB and colour.RGB_to_XYZ definition signatures have been changed to be easier to use, please update any code using them. See #1127 for more information.

With this release, the following scientific packages minimum versions are required:

• numpy >= 1.22

• scipy >= 1.8

• matplotlib >= 3.5

• networkx >= 2.7

• pandas >= 1.4

The highlights of this release are as follows:

• The Colour Fidelity Index (2017), i.e., colour.quality.colour_fidelity_index_CIE2017 definition, and TM 30-18, i.e., colour.quality.colour_fidelity_index_ANSIIESTM3018 definition, metrics performance has been improved by x100 thanks to overall optimizations, e.g., colour.temperature.uv_to_CCT_Ohno2013 definition is x5 faster.

• TE226 V2 colour checker reference values.

• New colour models:

  • Kirk (2019) Yrg colourspace.

• New RGB colourspaces and encodings:

  • Fujifilm F-Log2 encodings.

  • PLASA ANSI E1.54 colourspace.

• CSS Color 3 keywords

• New plotting definition for the Daylight Locus along with Mireds plotting support.

Please take a look at the releases page more information.