I'm more than willing to start porting the SVM to mahout once I finish the customer POC request.
At the same time, Michael Yang, who is my colleague at Alpine Data Labs will start to work on "single-pass algorithm for penalized linear regression with cross validation" first. We are looking forward to working with community closely. Thanks. Sincerely, DB Tsai ----------------------------------- Web: http://www.dbtsai.com Phone : +1-650-383-8392 On Sun, Jun 30, 2013 at 10:48 AM, Ted Dunning <[email protected]> wrote: > +1 to what Grant says. > > If you have code and tests, porting to Mahout is not that terribly hard. > > > > On Sun, Jun 30, 2013 at 4:32 AM, Grant Ingersoll <[email protected]>wrote: > >> Just coming back to this... >> >> On Jun 12, 2013, at 5:38 PM, DB Tsai <[email protected]> wrote: >> >> > Hi, >> > >> > For scalable SVM, since our codebase is quite different from mahout, >> > it may take some time to refactorize it to work in mahout. >> >> Note, the community may be able to help, here, if you put up a patch, then >> others likely will jump on and help. Your call, of course. >> >> Food for thought, >> Grant
