[mlpack-git] [mlpack/mlpack] Random projection trees (#726)

Ryan Curtin notifications at github.com
Fri Aug 5 10:49:44 EDT 2016


> @@ -135,6 +135,20 @@ using MeanSplitBallTree = BinarySpaceTree<MetricType,
>                                            bound::BallBound,
>                                            MeanSplit>;
>  
> +template<typename MetricType, typename StatisticType, typename MatType>
> +using RPTreeMax = BinarySpaceTree<MetricType,
> +                                  StatisticType,
> +                                  MatType,
> +                                  bound::HRectBound,
> +                                  RPTreeMaxSplit>;
> +
> +template<typename MetricType, typename StatisticType, typename MatType>
> +using RPTreeMean = BinarySpaceTree<MetricType,
> +                                  StatisticType,
> +                                  MatType,
> +                                  bound::HRectBound,
> +                                  RPTreeMeanSplit>;

I think for these, `MaxSplitRPTree` and `MeanSplitRPTree` might be more natural names.  Which do you think is the "default"?  If the max split is the more common one, then I would just call that `RPTree`.  This is kind of what I've done with the other types of trees---only specify the split type if it's not the "usual" split type.

---
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/mlpack/mlpack/pull/726/files/e14a26f3e209af0608cb61ecaa5e30f253e72e9b#r73704636
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack-git/attachments/20160805/263e3c59/attachment.html>


More information about the mlpack-git mailing list