[mlpack-svn] [MLPACK] #345: Sparse Autoencoder Module

[MLPACK Trac]

#345: Sparse Autoencoder Module
----------------------------+-----------------------------------------------
  Reporter:  siddharth.950  |        Owner:     
      Type:  enhancement    |       Status:  new
  Priority:  major          |    Milestone:     
 Component:  mlpack         |   Resolution:     
  Keywords:                 |     Blocking:     
Blocked By:                 |  
----------------------------+-----------------------------------------------

Comment (by rcurtin):

 > 13:13 < oldbeardo> you said we should work directly with parameters, but
 I'm using w1 and w2 for matrix operations
 > 13:13 < oldbeardo> how can that be done using parameters directly?
 > 13:15 -!- sumedhghaisas [~sumedhgha at 115.248.45.78] has joined #mlpack
 > 13:16 < oldbeardo> naywhayare: if you can reply right now, I can get a
 lot of work done tomorrow :)
 > 13:17 < naywhayare> parameters.subvec()
 > 13:18 < naywhayare> or parameters.submat(), or a function like that
 > 13:21 < oldbeardo> subvec() will give me a vector, I won't be able to
 use that for matrix operations without reshaping i

 I think you can use the standalone `reshape()` for this:

 `reshape(parameters.subvec(...), new_n_rows, new_n_cols)`

-- 
Ticket URL: <http://trac.research.cc.gatech.edu/fastlab/ticket/345#comment:4>
MLPACK <www.fast-lab.org>
MLPACK is an intuitive, fast, and scalable C++ machine learning library developed at Georgia Tech.