Open Erol444 opened 2 years ago
Hi @Erol444 , thanks for using distinctipy! I think I'd be happy for this to be added to distinctipy because I can see some of the benefits, but note that:
But PRs certainly welcome 🙂
Draft PR here: https://github.com/alan-turing-institute/distinctipy/pull/25
Head up: for our project (depthai-sdk) we might fork this to remove dependencies to panda/matplotlib and create a separate library, as we don't require such features (only create unique colors).
Draft PR here: #25
Great, I'll take a look when I have some time (hopefully later this week)
Head up: for our project (depthai-sdk) we might fork this to remove dependencies to panda/matplotlib and create a separate library, as we don't require such features (only create unique colors).
@Erol444 The latest versions of distinctipy
only have numpy as a core dependency actually, the pandas/matplotlib/etc. dependencies are optional extras now, which you should only get if you do pip install 'distinctipy[optional]'
or pip install 'distinctipy[all]'
. If you're still getting them from pip install distinctipy
that's a bug.
I have now noticed I was on the old branch (develop) and that main has had a few updates. Now I see that matplotlib/pandas aren't dependencies anymore, which is perfect! Here's the PR, you can link it to this feature request (github issue): https://github.com/alan-turing-institute/distinctipy/pull/26
I would add a main class
Color
(naming TBD), and the whole library would work with it.eg.
distinct_color()
returnsColor
.color_distance()
takes 2x Color as argument. Thewhy
is that we don't need to create our own helper classes for normalization/color order conversion.Let me know your thoughts, as I (or one of our team members) would like to start working on this. Thanks, Erik