Multiple people have expressed confusion about the way version numbering is done in TensorFlow Quantum. In particular, the version in the main branch on GitHub is our active development version, but the form of the version number is not well-enough differentiated from the form of version numbers used in releases. For example, the current released version on PyPI is 0.7.6 and the current version in main is 0.7.7. Many other projects (including most of our other QAI projects) follow a practice of adding a suffix like .dev0 to make it clear it is a development version. We should adopt a similar practice here.
The Python packaging guide page on versioning provides some more examples and recommendations.
Multiple people have expressed confusion about the way version numbering is done in TensorFlow Quantum. In particular, the version in the main branch on GitHub is our active development version, but the form of the version number is not well-enough differentiated from the form of version numbers used in releases. For example, the current released version on PyPI is
0.7.6and the current version in main is0.7.7. Many other projects (including most of our other QAI projects) follow a practice of adding a suffix like.dev0to make it clear it is a development version. We should adopt a similar practice here.The Python packaging guide page on versioning provides some more examples and recommendations.