Logistic Regression with Many dimensions

[Devin Ekins]

Perhaps it is the fact that I've got a headache and am exhausted, but I'm suddenly confused by the mlpy documentation regarding Large Linear Classifiers.  I was under the impression that Logistic regression could be performed with a multidimensional data set (d > 2) but I see listed in the documentation that LibLearner.learn(x,y) expects two-dimensional input for the x values.  Am I missing something about logistic regression, or am I misunderstanding the documentation?  

I was hoping to use Logistic regression to solve a k=2 classification problem against 20,000 data points with maybe 15 or so dimensions.  Am I approaching this from the wrong direction given the data set size and the dimensions?

(Sorry if this seems like a brain-dead question; I've really go to get some sleep!)

Thanks