Helper function for embeddings

[Optimox]

Feature request

As requested by some, and as @ekamioka started on this PR #244. It might be interesting to get some helper functions to use embeddings as it's not the simplest concept in deep learning.

What is the expected behavior?
Calling a few helper function to get all the correct parameters before using TabNet.

What is motivation or use case for adding/changing the behavior?
I think this would make things easier and help a lot of people.

How should this be implemented in your opinion?
Clean and clear, might be available in a helper.py file, completely independent of the rest of the code.

Are you willing to work on this yourself?
Sure why not

[askhan18]

Hey @Optimox, what kind of helper functions are we looking to implement?

[Optimox]

Hello @askhan18,

Basically the helper functions would facilitate #244.

The very basic would just ask for the categorical column id and would generate the input for tabnet : cat_dims, cat_emb_dims etc...

[Rachit-Gandhi]

Is this issue still open to work on? Would love to work on this just wanted to understand the progress done

[Optimox]

Hello @Rachit-Gandhi,

Nothing has been done at the moment, the scope is not well defined.
Basically there are small processing steps needed to correctly use tabnet like defining which columns are categorical and what embedding size should be used for each of those.

The idea would be to propose very simple functions that would make these things easier. Feel free to open a draft PR so that we can discuss potential changes on concrete basis.

[Rachit-Gandhi]

@Optimox sure, will do it in a week, this project is awesome used in my own project would love to work on this, my current responsibility ends at 5 th August, then I can contribute to this.