[mlpack-git] [mlpack] Maximum Inner-Product Search using Tree Data-structures in MLPACK ? (#427)

[Ryan Curtin]

Hmm, I should also add, I don't know how fast it will actually perform in practice, or how useful it will be.  Trees on high-dimensional sparse data don't generally work very well, although in this case we are using the cover tree, whose runtime depends instead on a measure of intrinsic dimensionality, so, maybe you can get away with using sparse data here (if it's intrinsically low-dimensional).  Either way, another consideration is that `arma::sp_mat`, in practice, can be slower than `arma::mat` unless the dimension of the sparse data is sufficiently high, and the sparsity level is high enough.  Exactly what "sufficiently high" and "high enough" mean are a bit hard to pin down... :)

---
Reply to this email directly or view it on GitHub:
https://github.com/mlpack/mlpack/issues/427#issuecomment-89867943
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack-git/attachments/20150405/8bb69de5/attachment.html>