Changeset 465

Timestamp:

Dec 17, 2005, 12:19:24 AM (18 years ago)

Author:

Peter

Message:

implemented score for WilcoxonFoldChange?

Location:

trunk/lib/statistics

Files:

• ROC.h (modified) (2 diffs)
• WilcoxonFoldChange.cc (modified) (3 diffs)
• WilcoxonFoldChange.h (modified) (3 diffs)

trunk/lib/statistics/ROC.h

@@ -35 +35 @@
     /// belonges to class + otherwise sample belongs to class
     /// -. @return the area under the ROC curve. If the area is less
-    /// than 0.5 and absolute=true, 1-area is returned.
+    /// than 0.5 and absolute=true, 1-area is returned. Complexity is
+    /// \f$ N\log N \f$ where \f$ N \f$ is number of samples.
     ///
     double score(const gslapi::vector& target, const gslapi::vector& value, 
@@ -48 +49 @@
     /// -. @return wheighted version of area under the ROC curve. If
     /// the area is less than 0.5 and absolute=true, 1-area is
-    /// returned.
+    /// returned. Complexity is \f$ N^2 \f$ where \f$ N \f$ is number
+    /// of samples.
     ///
     double score(const gslapi::vector& target, const gslapi::vector& value,

trunk/lib/statistics/WilcoxonFoldChange.cc

@@ -1 +1 @@
 // $Id$
 
-#include "WilcoxonFoldChange.h"
+#include <c++_tools/statistics/WilcoxonFoldChange.h>
+#include <c++_tools/statistics/utility.h>
 
-#include "Score.h"
-#include "Averager.h"
-#include "AveragerWeighted.h"
+#include <cmath>
 
 namespace theplu {
@@ -28 +27 @@
       train_set_=train_set;
 
+    std::vector<double> distance;
+    //Peter, should reserve the vector to avoid reallocations   
     weighted_=false;
-
+    for (size_t i=0; i<train_set_.size(); i++) {
+      for (size_t j=0; i<j; j++) {
+        if (target(train_set_[i])==target(train_set_[j])) continue;
+        distance.push_back(value(train_set_[i])-value(train_set_[j]));
+      }
+    }
+    if (absolute_)
+      return fabs(median(distance));
+    return median(distance);
   }
 
@@ -37 +46 @@
                                    const std::vector<size_t>& train_set) 
   {
-    if (!train_set_.size())
-      for (size_t i=0; i<target_.size(); i++)
-        train_set_.push_back(i);  
-    else
-      train_set_=train_set;
-
-    weighted_=true;
+    return 0;
   }
 

trunk/lib/statistics/WilcoxonFoldChange.h

@@ -4 +4 @@
 #define _theplu_statistics_wilcoxonfoldchange_
 
-#include "Score.h"
+#include <c++_tools/statistics/Score.h>
 
 namespace theplu {
@@ -34 +34 @@
           const std::vector<size_t>& train_set = std::vector<size_t>()); 
  
-    ///
+    /// @todo
     /// @return difference of the weighted means of the two classes
     ///
@@ -42 +42 @@
     /// @train_set defining which values to use (number of values used
     /// in the calculation is equal to size of \a train_set)
-    /// 
+    ///
+    /// @note not implemented
+    ///
     double 
     score(const gslapi::vector& target,