13:09 < haritha1313> Hi, I had a doubt regarding the SVDWrapper implemented in cf. I don't see that any SVD method being used by CF is using the wrapper.

13:09 < haritha1313> I am trying to enable using quic_svd as a MF method in CF and have a PR open for it. I am not sure whether I should use the wrapper or rather stick to writing function for the same in cf_impl.cpp since no other SVD method seems to be using it.

17:05 < rf_sust2018> hello

17:07 < zoq> rf_sust2018: Hello there!

17:59 < s1998> rcurtin: Sorry for noob question but I don't have much knowledge about DTBs (and trees). w.r.t HDBSCAN pr, I refactored the code to pass custom metric to the tree (for binary space tree) but then I think I need to refactor the code to include the metric in bound too, right? If that is correct then I think something needs to be modified for default bound policy, the default bound policy says :

17:59 < s1998> * Hyper-rectangle bound for an L-metric. This should be used in conjunction * with the LMetric class. Be sure to use the same template parameters for * LMetric as you do for HRectBound -- otherwise odd results may occur.

17:59 < s1998> So how should I refactor the policy? Or is there any paper (or if possible any other resource that you could point me to) that I could read and helps me with going ahead.

18:22 < Prabhat-IIT> zoq: can you take a look at my open pr related to Glorot Initialization as I've handled almost all the suggested changes. Moreover, plz also take a peek at SAGA optimizer pr. I've made sure that tests are also added in it.

18:50 < rajeshdm9> @zoq did you get time to look the draft of my proposal ?

21:22 < rf_sust2018> hi

21:23 < rf_sust2018> i want to work on Java bindings for mlpack

21:23 < rf_sust2018> let me know if anyone is assigned to it

21:23 < rf_sust2018> afaik someone it working on C# bindings