[mlpack-svn] [MLPACK] #177: Load(), Save(), and Validate() for each class

[MLPACK Trac]

#177: Load(), Save(), and Validate() for each class
-------------------------+--------------------------------------------------
 Reporter:  rcurtin      |        Owner:                                               
     Type:  enhancement  |       Status:  new                                          
 Priority:  major        |    Milestone:  MLPACK 1.1.0                                 
Component:  MLPACK       |     Keywords:  load save validate saverestoremodel class api
 Blocking:               |   Blocked By:                                               
-------------------------+--------------------------------------------------
 In accordance with a discussion that happened not so long ago, each method
 should implement the following:

  * `bool Load(NodeType& n)`
  * `bool Save(NodeType& n)`
  * `bool Validate()`

 The idea is that a main method would be able to call Save() or Load() with
 some NodeType class (the SaveRestoreModel works, but it needs hierarchical
 support [or maybe it already has it and I don't know how to use it]).
 `Load()` could call `Validate()` to make sure the loaded data is okay.
 But maybe that should have a boolean to indicate if the data should be
 validated...?

 Let's discuss the exact implementation of this one since there is no huge
 release pressure at the moment (relevant parties cc'ed).

-- 
Ticket URL: <http://trac.research.cc.gatech.edu/fastlab/ticket/177>
MLPACK <www.fast-lab.org>
MLPACK is an intuitive, fast, and scalable C++ machine learning library developed by the FASTLAB at Georgia Tech under Dr. Alex Gray.