[mlpack-svn] [MLPACK] #181: Documentation or tutorial for each machine learning method

[MLPACK Trac]

#181: Documentation or tutorial for each machine learning method
---------------------+------------------------------------------------------
 Reporter:  rcurtin  |        Owner:                        
     Type:  defect   |       Status:  new                   
 Priority:  major    |    Milestone:  MLPACK 1.0.1          
Component:  MLPACK   |     Keywords:  documentation tutorial
 Blocking:           |   Blocked By:                        
---------------------+------------------------------------------------------
 I talked to a scikit-learn contributor at NIPS (actually, the guy who
 wrote their ball trees), and we discussed documentation and I noted that
 scikit-learn has far better tutorials and documentation than any other
 machine learning library I've seen.

 For example, take this tutorial on classification:

 http://scikit-
 learn.org/stable/auto_examples/plot_classification_probability.html
 #example-plot-classification-probability-py

 Certainly, MLPACK is far different and needs different handling, but we
 should have complete examples for usage like this.  I envision one
 tutorial for each method we have, with the following general sections:

  * Introduction to task / method (what are we trying to do?)
  * Command-line executable documentation and usage [if it exists]
  * Simple, barebones C++ example (and documentation)
  * More complex C++ example
  * Full documentation of generalizability of class (through template
 parameters) and link to Doxygen documentation (or, Doxygen documentation
 inlined?)

 This shouldn't be too incredibly difficult because we don't have too many
 methods already.  CCing the usual suspects for comments on the style /
 format / my own poor life choices / etcetera...

-- 
Ticket URL: <https://trac.research.cc.gatech.edu/fastlab/ticket/181>
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.