Added warning of LLTSA and NPE are hardly applicable to very high-dim…

…ensional data
shogun-toolbox · Oct 23, 2011 · bf0bf48 · bf0bf48
1 parent e7b34f1
commit bf0bf48
Show file tree

Hide file tree

Showing 2 changed files with 7 additions and 1 deletion.
diff --git a/src/shogun/converter/LinearLocalTangentSpaceAlignment.h b/src/shogun/converter/LinearLocalTangentSpaceAlignment.h
@@ -31,6 +31,9 @@ class CDistance;
  * Neurocomputing, 70(7-9), 1547-1553. 
  * Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S0925231206004577
  * 
+ * This method is hardly applicable to very high-dimensional data due to 
+ * necessity to solve eigenproblem involving matrix of size (dim x dim).
+ *
  */
 class CLinearLocalTangentSpaceAlignment: public CLocalTangentSpaceAlignment
 {

diff --git a/src/shogun/converter/NeighborhoodPreservingEmbedding.h b/src/shogun/converter/NeighborhoodPreservingEmbedding.h
@@ -30,7 +30,10 @@ class CDistance;
  * Neighborhood preserving embedding. 
  * Tenth IEEE International Conference on Computer Vision ICCV05 Volume 1, 2, 1208-1213. Ieee.
  * Retrieved from http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=1544858
- * 
+ *
+ * This method is hardly applicable to very high-dimensional data due to 
+ * necessity to solve eigenproblem involving matrix of size (dim x dim).
+ *
  */
 class CNeighborhoodPreservingEmbedding: public CLocallyLinearEmbedding
 {