Great to hear that you use Mahout in production! If you want to start working on it, you can either browse our jira issues or propose some issue to work on yourself.
If you need some input, it would be awesome to enhance our ALS recommenders with cross-validation and tooling for finding a good regularization parameter. On 05.04.2013 01:56, Andrew Musselman wrote: > In case this thread is still a good place to reply with an offer to help, > I'd love to pitch in. I have built a few production recommenders, most > recently using Mahout at a large retailer along with my partner where we > used ALS, with a pipeline of transforming transactions in XML into vectors > using Pig, running them through Mahout, and writing to Cassandra. I've > also used Hadoop extensively and am experienced writing non-trivial UDF > libraries for Pig, so I am comfortable in that world. > > Where to start? Do you guys have bugs ranked in priority? > > Thanks > Andrew > > > On Wed, Mar 27, 2013 at 5:01 PM, Daniel Longest <dlong...@gmail.com> wrote: > >>> Have you used Mahout in some context before? Did you already checkout the >>> code base and build it? It usually helps if you are looking into >> solving a >>> specific problem with the project to find an area that interests you >>> personally. >> >> I worked through the Coursera ML course, which is how I heard about >> Mahout. I then read through some of Mahout in Action. I have checked >> out the codebase, had difficulty building it on Windows, but I've done >> it successfully on Ubuntu tonight so full steam ahead. Not sure I have >> a specific interest, seen a lot of good ideas thrown around on this >> thread this week. Ted had listed some possible GSOC ideas that I'd be >> open to if no students took them on. I see the JIRA has an "intro" >> tag, was going to explore some of those a bit to get my feet wet. >> >> Daniel >> >
