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

[MLPACK Trac]

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

Comment (by sumedhghaisas):

 [[0 0]  [[0]
  [1 1]   [2]
  [2 2]   [4]
  [1 2]   [3]
  [6 2]   [8]
  [2 6]]  [8]]

 This dataset gives cost as (1.88099 * 10 ^ -31).
 So this dataset surely can be used for checking ComputeCost. Designing a
 matrix such that X * X' is non-invertible makes sense. But this way we
 need to have different datasets to check each different functionality.
 After verifying the cost function we can just apply lambda value of 10 to
 this dataset and check the cost is high enough. With lambda value of 10
 this dataset gives cost as 0.3278. Can this be used as the check for ridge
 regression?
 If the cost function is verified and its giving high results with high
 lambda can we assume ridge regression is correct??

-- 
Ticket URL: <http://trac.research.cc.gatech.edu/fastlab/ticket/298#comment:10>
MLPACK <www.fast-lab.org>
MLPACK is an intuitive, fast, and scalable C++ machine learning library developed at Georgia Tech.