- Timestamp:
- Oct 14, 2008, 9:01:02 PM (13 years ago)
- Location:
- trunk/yat/normalizer
- Files:
-
- 2 edited
Legend:
- Unmodified
- Added
- Removed
-
trunk/yat/normalizer/ColumnNormalizer.h
r1575 r1578 33 33 34 34 /** 35 \brief Normalize each column35 \brief Using a functor T to normalize each column. 36 36 37 Using a functor T to normalize each column. 37 If the underlying normalizer_type allows normalization "in 38 place", i.e., it allows input range and result range to be the 39 same range, then it is safe to have same container for input as 40 output. 41 42 In the case of views and lookups it is more complicated. The 43 assignment is done sequentially, so if the input container and 44 result container have the same underlying data, the normalization 45 is typically not safe. 38 46 39 47 \since New in yat 0.5 48 49 \see RowNormalizer 40 50 */ 41 51 template<class T> -
trunk/yat/normalizer/QuantileNormalizer.h
r1577 r1578 40 40 will be equal to the average of the largest elements. The 2nd 41 41 largest element in each column will be equal to the average of 42 the 2nd largest element in eac ccolumn et cetera.42 the 2nd largest element in each column et cetera. 43 43 44 44 There is currently no weighted version of QuantileNormalizer
Note: See TracChangeset
for help on using the changeset viewer.