Update to newer versions of used python libraries possible?

[juliusaka]

Hey,

I'd like to use this package in one of my projects. Unfortunately, a lot of dependencies of this project generate version conflicts with my current configuration. I'm asking myself if the dependencies, e.g.

dependencies = [
    "matplotlib >=3.6.0",
    "derivative ~= 0.6.0",
    "scikit-learn >= 1.1.3, <= 1.1.3",
    "numpy >=1.20, <= 1.26",
    "scipy >1.6.0, <= 1.11.2",
    "pydmd >0.4, <= 0.4.1",
    "optht ~= 0.2.0",
    "prettytable >3.0.0, <= 3.9.0",
    "torch ~= 2.1.0",
    "torchvision ~= 0.16.0",
    "torchaudio ~= 2.1.0",
    "lightning ~= 2.0.9",
]

were made because there were known issues with newer versions, or just because of safety?
Best,
Julius

[acmiyaguchi]

Yeah, I think these requirements are a bit too strict and are causing issues with other installed libraries. The general practice should be to leave these are globs where needed, and then to have the compiled dependencies only when you're installing to a specific system. Not sure how open the maintainers for pull requests relaxing this though, probably easiest to fork and git+https:// from your fork.