Stuti, RF is an ensemble approach, generating 10 trees is too less to be of any value. Are you running in-mem version? Even in case of the distributed version, model building time depends on your compute cluster configuration and capacity.
Som On Fri, Jul 11, 2014 at 4:38 AM, Suneel Marthi <suneel.mar...@gmail.com> wrote: > Please work off of trunk, few fixes for RDF have gone in that should > address this issue. See release notes for details. > > Sent from my iPhone > > > On Jul 11, 2014, at 7:06 AM, Stuti Awasthi <stutiawas...@hcl.com> wrote: > > > > Mahout 0.7 > > > > -----Original Message----- > > From: Suneel Marthi [mailto:suneel.mar...@gmail.com] > > Sent: Friday, July 11, 2014 4:29 PM > > To: user@mahout.apache.org > > Subject: Re: Random Forest Implementation training is too slow for 2 GB > of data > > > > R u working off if trunk? Mahout version?? > > > > Sent from my iPhone > > > >> On Jul 11, 2014, at 6:53 AM, Stuti Awasthi <stutiawas...@hcl.com> > wrote: > >> > >> Hi all, > >> > >> I have some 2 GB of data and tried to execute RF with no of trees = 10 > and maxsplitsize as 90 MB. The execution takes too much time. > >> I have also tried with #of trees =2, then it takes less time but gives > >> accuracy <50% If I use less data with greater no of trees , then > >> output accuracy is >90% > >> > >> Is there any tuning to execute it quickly with optimal no of trees for > accuracy > 80%. > >> > >> > >> Please suggest > >> > >> Thanks > >> Stuti Awasthi > >> > >> > >> ::DISCLAIMER:: > >> ---------------------------------------------------------------------- > >> ---------------------------------------------------------------------- > >> -------- > >> > >> The contents of this e-mail and any attachment(s) are confidential and > intended for the named recipient(s) only. > >> E-mail transmission is not guaranteed to be secure or error-free as > >> information could be intercepted, corrupted, lost, destroyed, arrive > >> late or incomplete, or may contain viruses in transmission. The e mail > and its contents (with or without referred errors) shall therefore not > attach any liability on the originator or HCL or its affiliates. > >> Views or opinions, if any, presented in this email are solely those of > >> the author and may not necessarily reflect the views or opinions of > >> HCL or its affiliates. Any form of reproduction, dissemination, > >> copying, disclosure, modification, distribution and / or publication of > this message without the prior written consent of authorized representative > of HCL is strictly prohibited. If you have received this email in error > please delete it and notify the sender immediately. > >> Before opening any email and/or attachments, please check them for > viruses and other defects. > >> > >> ---------------------------------------------------------------------- > >> ---------------------------------------------------------------------- > >> -------- >
