[mlpack-svn] r11628 - mlpack/trunk/src/mlpack/methods/kernel_pca

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

Author: rcurtin
Date: 2012-02-28 12:05:19 -0500 (Tue, 28 Feb 2012)
New Revision: 11628

Modified:
   mlpack/trunk/src/mlpack/methods/kernel_pca/kernel_pca_main.cpp
Log:
I had the options backwards; also clarify documentation.


Modified: mlpack/trunk/src/mlpack/methods/kernel_pca/kernel_pca_main.cpp
===================================================================
--- mlpack/trunk/src/mlpack/methods/kernel_pca/kernel_pca_main.cpp	2012-02-28 17:00:58 UTC (rev 11627)
+++ mlpack/trunk/src/mlpack/methods/kernel_pca/kernel_pca_main.cpp	2012-02-28 17:05:19 UTC (rev 11628)
@@ -63,8 +63,8 @@
     "the output dataset by ignoring the dimensions with the smallest "
     "eigenvalues.", "d", 0);
 
-PARAM_FLAG("scale", "If set, the data will be scaled such that the variance "
-    "of each feature is 1.", "s");
+PARAM_FLAG("scale", "If set, the data will be scaled before performing KPCA "
+    "such that the variance of each feature is 1.", "s");
 PARAM_FLAG("nocenter", "If set, the data will NOT be centered before performing"
     " KPCA.", "N");
 
@@ -108,7 +108,7 @@
 
   if (kernelType == "linear")
   {
-    KernelPCA<LinearKernel> kpca(LinearKernel(), scaleData, centerData);
+    KernelPCA<LinearKernel> kpca(LinearKernel(), centerData, scaleData);
     kpca.Apply(dataset, newDim);
   }
   else if (kernelType == "gaussian")
@@ -116,7 +116,7 @@
     const double bandwidth = CLI::GetParam<double>("bandwidth");
 
     GaussianKernel kernel(bandwidth);
-    KernelPCA<GaussianKernel> kpca(kernel, scaleData, centerData);
+    KernelPCA<GaussianKernel> kpca(kernel, centerData, scaleData);
     kpca.Apply(dataset, newDim);
   }
   else if (kernelType == "polynomial")
@@ -125,7 +125,7 @@
     const double offset = CLI::GetParam<double>("offset");
 
     PolynomialKernel kernel(offset, degree);
-    KernelPCA<PolynomialKernel> kpca(kernel, scaleData, centerData);
+    KernelPCA<PolynomialKernel> kpca(kernel, centerData, scaleData);
     kpca.Apply(dataset, newDim);
   }
   else if (kernelType == "hyptan")
@@ -134,7 +134,7 @@
     const double offset = CLI::GetParam<double>("offset");
 
     HyperbolicTangentKernel kernel(scale, offset);
-    KernelPCA<HyperbolicTangentKernel> kpca(kernel, scaleData, centerData);
+    KernelPCA<HyperbolicTangentKernel> kpca(kernel, centerData, scaleData);
     kpca.Apply(dataset, newDim);
   }
   else if (kernelType == "laplacian")
@@ -142,12 +142,12 @@
     const double bandwidth = CLI::GetParam<double>("bandwidth");
 
     LaplacianKernel kernel(bandwidth);
-    KernelPCA<LaplacianKernel> kpca(kernel, scaleData, centerData);
+    KernelPCA<LaplacianKernel> kpca(kernel, centerData, scaleData);
     kpca.Apply(dataset, newDim);
   }
   else if (kernelType == "cosine")
   {
-    KernelPCA<CosineDistance> kpca(CosineDistance(), scaleData, centerData);
+    KernelPCA<CosineDistance> kpca(CosineDistance(), centerData, scaleData);
     kpca.Apply(dataset, newDim);
   }
   else