s is a big decision that will effect a lot of code. Not sure if
> anybody else has some work in progress. I could potentially try to get some
> performance numbers for comparison and point at some code I've used in the
> past, though it's all very rough.
>
> thanks,
> Imran
llib...if you have a pull
> request or jira link could you please point to it ? Jblas does not implememt
> sparse formats the last time I looked at it but colt had sparse formats
> which could be reused...
>
> Thanks.
> Deb
>
> On Jan 31, 2014 11:15 AM, "Xiangrui Meng&q
Hi Jason,
Sorry, I didn't see this message before I replied in another thread.
So the following is copy-and-paste:
We are currently working on the sparse data support, one of the
highest priority features for MLlib. All existing algorithms will
support sparse input. We will open a JIRA ticket for
It happens when there are empty columns. Adding a very small smoothing
factor should help. Btw, I notice that the computation of variance
there is not stable, which should use the stable method implemented in
RDD[Double]. -Xiangrui
On Tue, Jan 28, 2014 at 5:22 AM, yinxusen wrote:
> Hi all,
>
> Th