diff options
| author | Gertjan van den Burg <burg@ese.eur.nl> | 2016-10-17 15:40:33 +0200 |
|---|---|---|
| committer | Gertjan van den Burg <burg@ese.eur.nl> | 2016-10-17 15:40:33 +0200 |
| commit | 6e3b22ed8ece70539b6966107342ab8122320a24 (patch) | |
| tree | 73a91a8598ba36c2c70f9ad598cdda30129c1db2 /include/gensvm_sparse.h | |
| parent | refactor gensvm_pred to gensvm_predict (diff) | |
| download | gensvm-6e3b22ed8ece70539b6966107342ab8122320a24.tar.gz gensvm-6e3b22ed8ece70539b6966107342ab8122320a24.zip | |
minor documentation fixes
Diffstat (limited to 'include/gensvm_sparse.h')
| -rw-r--r-- | include/gensvm_sparse.h | 8 |
1 files changed, 2 insertions, 6 deletions
diff --git a/include/gensvm_sparse.h b/include/gensvm_sparse.h index bac576a..80945b2 100644 --- a/include/gensvm_sparse.h +++ b/include/gensvm_sparse.h @@ -41,11 +41,7 @@ * <a href="https://en.wikipedia.org/wiki/Sparse_matrix">Wikipedia</a> for * more details. The total storage requirement for this format is * 2*nnz+n_row+1, so it only makes sense to use this format if the number of - * nonzeros is smaller than @f$(n(m - 1) - 1)/2@f$. - * - * @note - * We use @f$n@f$ for rows, and @f$m@f$ for columns, to conform to GenSVM - * notation. This is in exact contrast with the Wikipedia page. + * nonzeros is smaller than @f$(n_{row}(n_{col} - 1) - 1)/2@f$. * * @param nnz number of nonzero elements * @param n_row rows of the matrix @@ -65,7 +61,7 @@ struct GenSparse { double *values; ///< actual nonzero values, should be of length nnz int *ia; - ///< cumulative row lengths, should be of length n+1 + ///< cumulative row lengths, should be of length n_row+1 int *ja; ///< column indices, should be of length nnz }; |