[mlpack-svn] [MLPACK] #298: Enhancement of linear regression library

[MLPACK Trac]

#298: Enhancement of linear regression library
--------------------------------+-------------------------------------------
  Reporter:  sumedhghaisas      |        Owner:              
      Type:  enhancement        |       Status:  new         
  Priority:  major              |    Milestone:  mlpack 1.0.7
 Component:  mlpack             |   Resolution:              
  Keywords:  linear_regression  |     Blocking:              
Blocked By:                     |  
--------------------------------+-------------------------------------------

Comment (by sumedhghaisas):

 I have changed my implementation so that to adjust with current column
 matrix convention. Column matrix will boost the speed calculations as the
 storing is column major, thats why in implementation I have multiplied
 trans(trans(parameters) * points) and not trans(points) * points so that
 major multiplication will be boost up my column major. And i have used
 arma::solve in place of inverse, does that solve the problem?? cause i
 dont know the exact implementation of arma::solve.

 and I have returned predictions as a column matrix but looking at your
 code I am confused about the conventional return. If your code wants to
 return predictions as row matrix then just remove the trans function in
 the implementation.

 and about the test, these files pass the test, and i am confused what to
 implement in the test function.

-- 
Ticket URL: <http://trac.research.cc.gatech.edu/fastlab/ticket/298#comment:3>
MLPACK <www.fast-lab.org>
MLPACK is an intuitive, fast, and scalable C++ machine learning library developed at Georgia Tech.