The trainlogistic program is intended only as a small demo program. It would be better to use it as a model for your own code.
Also, the logistic regression implementation in Mahout is probably not going to be competitive with a full-scale production version such as the one in H2O. That may change in the future, but the current implementation is not very impressive performance wise. On Mon, Sep 1, 2014 at 5:52 AM, sara khodeir <[email protected]> wrote: > Just to make sure because I don't have access to the cluster at the moment > so I can't try it out the command should look like this: > mahout conf/[trainlogistic].props ? > > > On Mon, Sep 1, 2014 at 3:50 PM, sara khodeir <[email protected]> > wrote: > > > Yes that's what I was looking for, thanks! > > > > > > On Mon, Sep 1, 2014 at 3:43 PM, tuxdna <[email protected]> wrote: > > > >> > Yes that's the same command I'm using, and yes that is what I'm > >> referring > >> > to. > >> > > >> > > >> > >> Well, in that case you can use "trainlogistic.props" file to specify > >> command line parameters in a separate file. This file is located in > >> conf/ folder of Mahout distribution. > >> > >> In general you can specify CLI options in file conf/[X].props where > >> [X] is the command you are using with mahout ( such as trainlogistic, > >> runlogistic, viterbi etc. ) > >> > >> I hope that helps. > >> > >> /tuxdna > >> > > > > >
