- Timestamp:
- Jan 15, 2009, 10:15:26 PM (12 years ago)
- Location:
- trunk/yat/utility
- Files:
-
- 3 edited
Legend:
- Unmodified
- Added
- Removed
-
trunk/yat/utility/NNI.h
r1487 r1726 90 90 virtual ~NNI(void) {}; 91 91 92 /// 93 /// Function doing the imputation. 94 /// 95 /// @return number of rows not imputed 96 /// 92 /** 93 \brief Function doing the imputation. 94 95 The return value can be used as an indication of how well the 96 imputation worked. The return value should be zero if proper 97 pre-processing of data is done. An example of bad data is a 98 matrix with a column of zero weights, another is a 99 corresponding situation with a row with all weights zero. 100 101 \return The number of rows that have at least one value not 102 imputed. 103 */ 97 104 virtual unsigned int estimate(void)=0; 98 105 -
trunk/yat/utility/WeNNI.h
r1725 r1726 62 62 const unsigned int neighbours); 63 63 64 /// 65 /// Perform WeNNI on data in \a matrix with continuous uncertainty 66 /// weights in \a weight using \a neighbours for the new impute 67 /// value. 68 /// 64 /** 65 \brief Function doing WeNNI imputation. 66 67 Perform WeNNI on data in \a matrix with continuous uncertainty 68 weights in \a weight using \a neighbours for the new impute 69 value. 70 71 The return value can be used as an indication of how well the 72 imputation worked. The return value should be zero if proper 73 pre-processing of data is done. An example of bad data is a 74 matrix with a column of zero weights, another is a 75 corresponding situation with a row with all weights zero. 76 77 \return The number of rows that have at least one value not 78 imputed. 79 */ 69 80 unsigned int estimate(void); 70 81 -
trunk/yat/utility/kNNI.h
r1487 r1726 57 57 const unsigned int neighbours); 58 58 59 /// 60 /// Perform kNNI on data in \a matrix with binary uncertainty 61 /// weights in \a weight using \a neighbours for the new impute 62 /// value. 63 /// 59 /** 60 \brief Function doing kNNI imputation. 61 62 Perform kNNI on data in \a matrix with binary uncertainty 63 weights in \a weight using \a neighbours for the new impute 64 value. 65 66 The return value can be used as an indication of how well the 67 imputation worked. The return value should be zero if proper 68 pre-processing of data is done. An example of bad data is a 69 matrix with a column of zero weights, another is a 70 corresponding situation with a row with all weights zero. 71 72 \return The number of rows that have at least one value not 73 imputed. 74 */ 64 75 unsigned int estimate(void); 65 76
Note: See TracChangeset
for help on using the changeset viewer.