Re: Training a neural net (or other classifier) across StarCluster

From: Alessandro Gagliardi <no email>
Date: Wed, 11 Dec 2013 21:14:47 +0000

Update: I had a stupid error in my input. (My target value was coming across as a factor, not a float, so every value was a different class. Whoops!)

That said, I’d still be interested if there is an answer to this question.

From: Alessandro <alessandro.gagliardi_at_glassdoor.com<mailto:alessandro.gagliardi_at_glassdoor.com>>
Date: Wednesday, December 11, 2013 at 11:59 AM
To: "starcluster_at_mit.edu<mailto:starcluster_at_mit.edu>" <starcluster_at_mit.edu<mailto:starcluster_at_mit.edu>>
Subject: Training a neural net (or other classifier) across StarCluster

I would like to train a neural net (or similar classifier) to predict one probabilistic value from 9 principal components. When I do it in R (using nnet) it caps at a few hundred observations, but that seems too small a sample when I have over 40k cases. I know that the space of machine learning algorithms (distributed and otherwise) is vast and so was wondering if there was something that the StarCluster community might recommend. (Ideally it would be something that I could set up on StarCluster with minimal difficulty.)

Thanks in advance,
-Alessandro Gagliardi
Received on Wed Dec 11 2013 - 16:15:20 EST