Changeset 202

Timestamp:

Nov 1, 2004, 4:00:41 PM (19 years ago)

Author:

Peter

Message:

adding classes for Regression

Location:

trunk/src

Files:

• Makefile.am (modified) (2 diffs)
• Regression.h (added)
• RegressionLinear.cc (modified) (2 diffs)
• RegressionLinear.h (modified) (5 diffs)
• RegressionLocal.cc (added)
• RegressionLocal.h (added)
• RegressionNaive.cc (added)
• RegressionNaive.h (added)

trunk/src/Makefile.am

@@ -14 +14 @@
   Kernel.cc kNNI.cc matrix.cc Merge.cc NNI.cc PCA.cc \
   PolynomialKernelFunction.cc random_singleton.cc \
+  RegressionLinear.cc RegressionLocal.cc RegressionNaive.cc ROC.cc \
+  Score.cc Statistics.cc \
+  PolynomialKernelFunction.cc random_singleton.cc \
   RegressionLinear.cc ROC.cc Score.cc Statistics.cc \
   SVD.cc SVM.cc tScore.cc vector.cc WeightedAverager.cc WeNNI.cc
@@ -28 +31 @@
   KernelFunction.h kNNI.h matrix.h Merge.h NNI.h PCA.h Pearson.h \
   PolynomialKernelFunction.h random_singleton.h \
+  Regression.h RegressionLinear.h RegressionLocal.h RegressionNaive.h \
+  ROC.h Score.h \
+  PolynomialKernelFunction.h random_singleton.h \
   RegressionLinear.h ROC.h Score.h \
   Statistics.h stl_utility.h SVD.h SVM.h tScore.h vector.h WeNNI.h \

trunk/src/RegressionLinear.cc

@@ -1 +1 @@
 // $Id$
 
+// Header
 #include "RegressionLinear.h"
 
+// C++_tools
+#include "Regression.h"
 #include "vector.h"
 
@@ -13 +16 @@
   
   RegressionLinear::RegressionLinear(void)
-    : cov00_(0), cov01_(0), cov11_(0), k_(0), m_(0), sumsq_(0)
+    : Regression(), cov00_(0), cov01_(0), cov11_(0), k_(0), m_(0), sumsq_(0)
   {
   }

trunk/src/RegressionLinear.h

@@ -6 +6 @@
 // C++ tools include
 /////////////////////
+#include "Regression.h"
 #include "vector.h"
 
@@ -20 +21 @@
 
   
-  class RegressionLinear 
+  class RegressionLinear : public Regression 
   {
   
@@ -30 +31 @@
 
     ///
+    /// Copy Constructor. (not implemented)
+    ///
+    RegressionLinear(const RegressionLinear&);
+
+    ///
     /// Destructor 
     ///
@@ -36 +42 @@
     ///
     /// This function computes the best-fit linear regression
-    /// coefficients (k,m) of the model  \f$ y = kx + m \f$ from from
+    /// coefficients (k,m) of the model  \f$ y = kx + m \f$ from
     /// vectors \a x and \a y, by minimizing \f$ \sum{(kx_i+m-x_i)^2}
     /// \f$.
@@ -48 +54 @@
     /// This function computes the best-fit linear regression
     /// coefficients (k,m) of the model \f$ y = kx + m \f$ from from
-    /// vectors \a x and \a y, by minimizing \f$ \sum w_i(kx_i+m-x_i)^2
-    /// \f$. The weight \f$ w_i \f$ is the inverse of the variance for
-    /// \f$ y_i \f$
+    /// vectors \a x and \a y, by minimizing \f$ \sum
+    /// w_i(kx_i+m-x_i)^2 \f$.
     ///
-    inline int fit_weigted(gslapi::vector x, gslapi::vector y, gslapi::vector w)
+    inline int fit_weighted(const gslapi::vector x, const gslapi::vector y, 
+                           const gslapi::vector w)
       { return gsl_fit_wlinear_vector(x.gsl_vector_pointer(),
                                       w.gsl_vector_pointer(),