Now, we use a custom implementation for a data class (not a true data class but close enough) for PyCytoData objects. I'm wondering whether AnnData will make more sense. Maybe this is a PyCytoData 2.0, which is far far away, but I'm just throwing this out there.
Is your feature request related to a current problem? Please describe.
Not really. I'm just ~bored~ curious.
How do you envision this feature to be implemented?
We will need to basically swap out the engine of PyCytoData objects if we were to do it. I'm almost sure that there'll be tons of breaking changes. So, maybe an interface is enough? I'm just curious.
What feature would you like to propose?
Now, we use a custom implementation for a data class (not a true data class but close enough) for PyCytoData objects. I'm wondering whether
AnnData
will make more sense. Maybe this is a PyCytoData 2.0, which is far far away, but I'm just throwing this out there.Is your feature request related to a current problem? Please describe.
Not really. I'm just ~bored~ curious.
How do you envision this feature to be implemented? We will need to basically swap out the engine of
PyCytoData
objects if we were to do it. I'm almost sure that there'll be tons of breaking changes. So, maybe an interface is enough? I'm just curious.