You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@mahout.apache.org by Koert Kuipers <ko...@tresata.com> on 2011/11/09 16:52:06 UTC

AdaptiveLogisticRegression

To train the AdaptiveLogisticRegression, do i need to feed in new training
data only once? Or is iteration over the training data here helpful as well?
Thanks! Koert

Re: AdaptiveLogisticRegression

Posted by Ted Dunning <te...@gmail.com>.

Depends on your data set size.  ALR learns primarily based on the number of
examples seen.  If that takes several passes on a small dataset, it has
more to do with the dataset being small.

On Wed, Nov 9, 2011 at 10:52 AM, Koert Kuipers <ko...@tresata.com> wrote:

> To train the AdaptiveLogisticRegression, do i need to feed in new training
> data only once? Or is iteration over the training data here helpful as
> well?
> Thanks! Koert
>