Changeset 1577
- Timestamp:
- Oct 14, 2008, 3:49:49 PM (14 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
trunk/yat/normalizer/QuantileNormalizer.h
r1575 r1577 33 33 \brief Perform quantile normalization 34 34 35 After a quantile normalization each column has the same 36 distribution of data (the quantiles are the same). Also, within 37 each column the rank of an element is not changed. The 38 distribution that each column follows is determined by taking the 39 average across columns, i.e., the largest element in each column 40 will be equal to the average of the largest elements. The 2nd 41 largest element in each column will be equal to the average of 42 the 2nd largest element in eacc column et cetera. 43 44 There is currently no weighted version of QuantileNormalizer 45 35 46 \since New in yat 0.5 36 47 */ … … 39 50 public: 40 51 /** 41 The \a matrix is normalized by sorting each column, replacing 42 each value with the average across rows, and undo the sorting 43 in step 1. This results in a Matrix \a result in which each 44 column has the same distribution of data, and the rank of an 45 element within a column is the same as in \a matrix. 52 \brief perform the quantile normalization. 53 54 It is possible to normalize "in place"; it is permissible for 55 \a matrix and \a result to reference to the same Matrix. 46 56 47 57 \note dimensions of \a matrix and \a result must match.
Note: See TracChangeset
for help on using the changeset viewer.