[mlpack-svn] r16697 - mlpack/trunk/src/mlpack/methods/amf/update_rules
fastlab-svn at coffeetalk-1.cc.gatech.edu
fastlab-svn at coffeetalk-1.cc.gatech.edu
Mon Jun 23 17:39:03 EDT 2014
Author: rcurtin
Date: Mon Jun 23 17:39:03 2014
New Revision: 16697
Log:
Make a note that this set of update rules often creates lots of NaNs when sparse
matrices are used.
Modified:
mlpack/trunk/src/mlpack/methods/amf/update_rules/nmf_mult_div.hpp
Modified: mlpack/trunk/src/mlpack/methods/amf/update_rules/nmf_mult_div.hpp
==============================================================================
--- mlpack/trunk/src/mlpack/methods/amf/update_rules/nmf_mult_div.hpp (original)
+++ mlpack/trunk/src/mlpack/methods/amf/update_rules/nmf_mult_div.hpp Mon Jun 23 17:39:03 2014
@@ -9,6 +9,10 @@
* \f$ \sum_i \sum_j (V_{ij} log\frac{V_{ij}}{(WH)_{ij}}-V_{ij}+(WH)_{ij}) \f$is
* non-increasing between subsequent iterations. Both of the update rules
* for W and H are defined in this file.
+ *
+ * This set of update rules is not meant to work with sparse matrices. Using
+ * sparse matrices often causes NaNs in the output, so other choices of update
+ * rules are better in that situation.
*/
#ifndef __MLPACK_METHODS_LMF_UPDATE_RULES_NMF_MULT_DIV_HPP
#define __MLPACK_METHODS_LMF_UPDATE_RULES_NMF_MULT_DIV_HPP
@@ -18,7 +22,6 @@
namespace mlpack {
namespace amf {
-
class NMFMultiplicativeDivergenceUpdate
{
public:
More information about the mlpack-svn
mailing list