Changeset 1717


Ignore:
Timestamp:
Jan 13, 2009, 11:32:00 PM (12 years ago)
Author:
Peter
Message:

improved(?) docs. Moved some docs from operator() to class docs.

Location:
trunk/yat/normalizer
Files:
4 edited

Legend:

Unmodified
Added
Removed
  • trunk/yat/normalizer/Centralizer.h

    r1576 r1717  
    55
    66/*
    7   Copyright (C) 2008 Peter Johansson
     7  Copyright (C) 2008, 2009 Peter Johansson
    88
    99  This file is part of the yat library, http://dev.thep.lu.se/yat
     
    5252       \brief default constructor
    5353
    54        Creates a UnaryFunction using default constructor.
     54       Internal UnaryFunction is created using its default
     55       constructor.
    5556     */
    5657    Centralizer(void){}
     
    7071       It is possible to centralize a range "in place"; it is
    7172       permissible for the iterators \a first and \a result to be the
    72        same. \see std::transform
     73       same.
     74
     75       \see std::transform
    7376
    7477       \return result + (last-first)
  • trunk/yat/normalizer/Gauss.h

    r1575 r1717  
    55
    66/*
    7   Copyright (C) 2008 Peter Johansson
     7  Copyright (C) 2008, 2009 Peter Johansson
    88
    99  This file is part of the yat library, http://dev.thep.lu.se/yat
     
    3535     \brief Gaussian Normalizer
    3636
     37     After normalization the range will follow a standard Gaussian
     38     distribution (mean zero and unity variance).
     39     
     40     The range is first rank normalized using Spearman, after which
     41     each element is between 0 and unity. Second each element is
     42     replaced by inverse cumulative standard Gaussian distribution.
     43     
    3744     \since New in yat 0.5
    3845   */
     
    4451       permissible for the iterators \a first and \a result to be the
    4552       same.
    46 
    47        The range is first rank normalized using Spearman, after which
    48        each element is between 0 and unity. Second each element is
    49        replaced by inverse cumulative standard Gaussian distribution.
    50        
    51        After normalization the range will follow a standard Gaussian
    52        distribution (mean zero and unity variance).
    5353
    5454       \see gsl_cdf_ugaussian_Pinv
  • trunk/yat/normalizer/Spearman.h

    r1575 r1717  
    55
    66/*
    7   Copyright (C) 2008 Peter Johansson
     7  Copyright (C) 2008, 2009 Peter Johansson
    88
    99  This file is part of the yat library, http://dev.thep.lu.se/yat
     
    3939     \brief Replace elements with normalized rank
    4040
     41     Each element x is replaced by \f$ \frac{\sum I(x_i-x) w_i}{\sum
     42     w_i} \f$ where \f$ I(x) = 1 \f$ for \f$ x>0 \f$, \f$I(x) = 0.5
     43     \f$ for \f$ x=0 \f$, and \f$ I(x) = 0 \f$ for \f$ x<0 \f$.
     44
    4145     \since New in yat 0.5
    4246   */
     
    5357       permissible for the iterators \a first and \a result to be the
    5458       same.
    55 
    56        Each element x is replaced by \f$ \frac{\sum I(x_i-x)
    57        w_i}{\sum w_i} \f$ where I(x) = 1 for x>0, I(x) = 0.5 for x=0,
    58        and I(x) = 0 for x<0.
    5959
    6060       \return result + (last-first)
  • trunk/yat/normalizer/Zscore.h

    r1575 r1717  
    55
    66/*
    7   Copyright (C) 2008 Peter Johansson
     7  Copyright (C) 2008, 2009 Peter Johansson
    88
    99  This file is part of the yat library, http://dev.thep.lu.se/yat
     
    3535     \brief Zero mean and unity variance
    3636
    37      Create a range that has zero mean and unity variance
     37     Shift and scale the values in a range as: \f$ y_i =
     38     \frac{x_i-m}{s} \f$ where \a m is the mean and \a s is the
     39     standard deviation. After normalization, the range will have zero
     40     mean and unity variance.
    3841
    3942     \since New in yat 0.5
     
    4447    /**
    4548       The element in range [result, result + (last-first)) is
    46        calculated as result[i] = (first[i] - m) / std where m and std
     49       calculated as result[i] = (first[i] - m) / s where m and std
    4750       are the mean and standard deviation, respectively, of the range
    4851       [first, last).
Note: See TracChangeset for help on using the changeset viewer.