Using `scikit-base` `BaseObject` for the base class?

[fkiraly]

Really nice package!

I was wondering, might I interest you in scikit-base? https://github.com/sktime/skbase

There are some features that you might like:

• support for parametric objects and their compositions, e.g., "first sums of sequence x" becomes Cumulative(Fibonacci()) or similar - with sklearn-like get_params, set_params, or added sequences like you want in https://github.com/VascoSch92/sequentium/issues/21
• things like __eq__ and __repr__ have default implementations compatible with sklearn
• support for tags, you can tag sequences with properties, e.g., is it monotonously increasing, is it integer, etc
• dependency handling localized to object, in case you need to import a package to compute a sequence
• templates for testing classes for interface conformance

[VascoSch92]

I'm not against the idea. There are features implemented in this classes which seems very interesting.

But first, I would like to understand how to use the scikit-base class properly :-)

I think the project discussed in https://github.com/VascoSch92/sequentium/issues/46 will be a good learning experience, i.e., I would like to spend time in implementing this module/package before porting this class here.

[fkiraly]

makes sense, given that you are interested in helping with that!