Make nn algorithm configurable

[msluszniak]

Closes #280

[krstopro]

Just a comment regarding allowed algorithms.

[krstopro]

I would add at least :brute. Maybe custom k-NN graph construction algorithm to be passed as a module as well.

[josevalim]

Is there any benefit in using brute other than nndescent or large_vis?

[krstopro]

I'd say it's the best one to use for smaller datasets.

[krstopro]

Perhaps making the selection automatic depending on dataset size (sample size and number of features) would be ideal.

[msluszniak]

I will force use brute for n < 100 and otherwise one of these two approximated algorithms, is it ok or do we want to add :brute anyway?

[krstopro]

Maybe use brute-force for $N \times D^2 \leq T$ for some constant $T$ (e.g. $10^5$ or $10^6$ or so).
I would add :brute anyway; it might be useful to see how much the quality of embeddings differ when approximate k-NN search algorithms are used.

Looking at it now, we might wanna change predict to transform in these algorithms as well.

[msluszniak]

for $N = 20000$ and $D = 2$, we rather won't use brute, generally, it's more effective to have this condition only on N I guess, I can increase this to 500-1000

[krstopro]

Sounds good to me.

[krstopro]

Function that computes the Trimap is called embed (not fit or transform). Does this deviate from the rest?

[msluszniak]

Function that computes the Trimap is called embed (not fit or transform). Does this deviate from the rest?

Yeah, I will change the name to transform.

[krstopro]

Two small comments.

Perhaps renaming algorithm to knn_algorithm would make it more understandable? Up to you.

[krstopro]

Looks good to me! If you want, I can have a more detailed look tomorrow.

Otherwise, feel free to merge it. I hope I am not annoying with these comments. ^_^

[msluszniak]

Looks good to me! If you want, I can have a more detailed look tomorrow.

Otherwise, feel free to merge it. I hope I am not annoying with these comments. ^_^

Your comments is a valueable feedback for me, so don't stop 😀