[mlpack-svn] r16801 - in mlpack/trunk/src/mlpack: methods/decision_stump tests

[fastlab-svn at coffeetalk-1.cc.gatech.edu]

Author: saxena.udit
Date: Wed Jul  9 16:15:45 2014
New Revision: 16801

Log:
IsDistinct() improved.

Modified:
   mlpack/trunk/src/mlpack/methods/decision_stump/decision_stump.hpp
   mlpack/trunk/src/mlpack/methods/decision_stump/decision_stump_impl.hpp
   mlpack/trunk/src/mlpack/tests/decision_stump_test.cpp

Modified: mlpack/trunk/src/mlpack/methods/decision_stump/decision_stump.hpp
==============================================================================
--- mlpack/trunk/src/mlpack/methods/decision_stump/decision_stump.hpp	(original)
+++ mlpack/trunk/src/mlpack/methods/decision_stump/decision_stump.hpp	Wed Jul  9 16:15:45 2014
@@ -123,7 +123,7 @@
    *
    * @param featureRow The attribute which is checked for identical values.
    */
-  template <typename rType> int isDistinct(const arma::Row<rType>& featureRow);
+  template <typename rType> int IsDistinct(const arma::Row<rType>& featureRow);
 
   /**
    * Calculate the entropy of the given attribute.

Modified: mlpack/trunk/src/mlpack/methods/decision_stump/decision_stump_impl.hpp
==============================================================================
--- mlpack/trunk/src/mlpack/methods/decision_stump/decision_stump_impl.hpp	(original)
+++ mlpack/trunk/src/mlpack/methods/decision_stump/decision_stump_impl.hpp	Wed Jul  9 16:15:45 2014
@@ -44,13 +44,13 @@
   for (int i = 0; i < data.n_rows; i++)
   {
     // Go through each attribute of the data.
-    if (isDistinct<double>(data.row(i)))
+    if (IsDistinct<double>(data.row(i)))
     {
       // For each attribute with non-identical values, treat it as a potential
       // splitting attribute and calculate entropy if split on it.
       entropy = SetupSplitAttribute(data.row(i), labels);
 
-      Log::Debug << "Entropy for attribute " << i << " is " << entropy << ".\n";
+      // Log::Debug << "Entropy for attribute " << i << " is " << entropy << ".\n";
       gain = rootEntropy - entropy;
       // Find the attribute with the best entropy so that the gain is
       // maximized.
@@ -351,12 +351,13 @@
  */
 template <typename MatType>
 template <typename rType>
-int DecisionStump<MatType>::isDistinct(const arma::Row<rType>& featureRow)
+int DecisionStump<MatType>::IsDistinct(const arma::Row<rType>& featureRow)
 {
-  if (featureRow.max() - featureRow.min() > 0)
-    return 1;
-  else
-    return 0;
+  rType val = featureRow(0);
+  for (size_t i = 1; i < featureRow.n_elem; ++i)
+    if (val != featureRow(i))
+      return 1;
+  return 0;
 }
 
 /**

Modified: mlpack/trunk/src/mlpack/tests/decision_stump_test.cpp
==============================================================================
--- mlpack/trunk/src/mlpack/tests/decision_stump_test.cpp	(original)
+++ mlpack/trunk/src/mlpack/tests/decision_stump_test.cpp	Wed Jul  9 16:15:45 2014
@@ -307,22 +307,4 @@
   }
 }
 
-BOOST_AUTO_TEST_CASE(TempAttributeSplit)
-{
-  const size_t numClasses = 2;
-  const size_t inpBucketSize = 3;
-
-  mat trainingData;
-  trainingData << 1 << 1 << 1 << 2 << 2 << 2 << endr
-               << 0.5  << 0.6  << 0.7  << 0.4  << 0.3  << 0.5 << endr;
-
-  Mat<size_t> labelsIn;
-  labelsIn << 0 << 0 << 0 << 0 << 1 << 1 << 1;
-
-  DecisionStump<> ds(trainingData, labelsIn.row(0), numClasses, inpBucketSize);
-
-  // Row<size_t> predictedLabels(testingData.n_cols);
-  // ds.Classify(testingData, predictedLabels);
-  BOOST_CHECK_EQUAL(ds.SplitAttribute(), 0);
-}
 BOOST_AUTO_TEST_SUITE_END();