PLearn 0.1
PrecomputedKernel.cc
Go to the documentation of this file.
00001 // -*- C++ -*-
00002 
00003 // PLearn (A C++ Machine Learning Library)
00004 // Copyright (C) 1998 Pascal Vincent
00005 // Copyright (C) 1999-2002 Pascal Vincent, Yoshua Bengio, Rejean Ducharme and University of Montreal
00006 // Copyright (C) 2001-2002 Nicolas Chapados, Ichiro Takeuchi, Jean-Sebastien Senecal
00007 // Copyright (C) 2002 Xiangdong Wang, Christian Dorion
00008 
00009 // Redistribution and use in source and binary forms, with or without
00010 // modification, are permitted provided that the following conditions are met:
00011 // 
00012 //  1. Redistributions of source code must retain the above copyright
00013 //     notice, this list of conditions and the following disclaimer.
00014 // 
00015 //  2. Redistributions in binary form must reproduce the above copyright
00016 //     notice, this list of conditions and the following disclaimer in the
00017 //     documentation and/or other materials provided with the distribution.
00018 // 
00019 //  3. The name of the authors may not be used to endorse or promote
00020 //     products derived from this software without specific prior written
00021 //     permission.
00022 // 
00023 // THIS SOFTWARE IS PROVIDED BY THE AUTHORS ``AS IS'' AND ANY EXPRESS OR
00024 // IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
00025 // OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN
00026 // NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
00027 // SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED
00028 // TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
00029 // PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
00030 // LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
00031 // NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
00032 // SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
00033 // 
00034 // This file is part of the PLearn library. For more information on the PLearn
00035 // library, go to the PLearn Web site at www.plearn.org
00036 
00037 
00038 /* *******************************************************      
00039  * $Id: PrecomputedKernel.cc 6861 2007-04-09 19:04:15Z saintmlx $
00040  * This file is part of the PLearn library.
00041  ******************************************************* */
00042 
00043 #include "PrecomputedKernel.h"
00044 
00045 namespace PLearn {
00046 using namespace std;
00047 
00048 
00049 
00050 // ** PrecomputedKernel **
00051 
00052 PLEARN_IMPLEMENT_OBJECT(PrecomputedKernel, "ONE LINE DESCR", "NO HELP");
00053 
00054 // PrecomputedKernel::~PrecomputedKernel()
00055 // {
00056 //   if(precomputedK)
00057 //     delete[] precomputedK;
00058 // }
00059 
00060 void PrecomputedKernel::build_()
00061 {}
00062 
00063 
00064 void PrecomputedKernel::build()
00065 {
00066     inherited::build();
00067     build_();
00068 }
00069 
00070 
00071 
00072 void PrecomputedKernel::makeDeepCopyFromShallowCopy(CopiesMap& copies)
00073 {
00074     inherited::makeDeepCopyFromShallowCopy(copies);
00075     deepCopyField(ker, copies);
00076     deepCopyField(precomputedK, copies);
00077 }
00078 
00079 
00080 // Old
00081 // void PrecomputedKernel::setDataForKernelMatrix(VMat the_data)
00082 // { 
00083 //   Kernel::setDataForKernelMatrix(the_data);
00084 //   ker->setDataForKernelMatrix(the_data);
00085   
00086 //   if(precomputedK)
00087 //     delete[] precomputedK;
00088 //   int l = data.length();
00089 //   precomputedK = new float(l*l);
00090 //   float* Kdata = precomputedK;
00091 //   for(int i=0; i<l; i++)
00092 //     {
00093 //       cerr << "Precomputing Kernel Matrix Row " << i << " of " << l << " ..." << endl;
00094 //       for(int j=0; j<l; j++)
00095 //         Kdata[j] = (float)ker->evaluate_i_j(i,j);
00096 //       Kdata += l;
00097 //     }
00098 // }
00099 
00100 /*
00101   Given that the matrix is symetric, we
00102   reduce the computation from n^2 to (n^2)/2 + n/2 calls to evaluate_i_j
00103 */
00104 void PrecomputedKernel::setDataForKernelMatrix(VMat the_data)
00105 {   
00106     Kernel::setDataForKernelMatrix(the_data);
00107     ker->setDataForKernelMatrix(the_data);
00108   
00109     int len = data.length();
00110     precomputedK.resize(len); //TVec of lines!!!
00111     for(int i=0; i < len; i++)
00112     {
00113         precomputedK[i].resize(len);
00114     
00115         for(int j=0; j < len; j++)
00116         {
00117             if(is_symmetric && j<i)
00118                 precomputedK[i][j] = precomputedK[j][i];
00119             else
00120                 precomputedK[i][j] = ker->evaluate_i_j(i,j);
00121         }
00122     }
00123 }
00124 
00125 
00126 real PrecomputedKernel::evaluate(const Vec& x1, const Vec& x2) const
00127 { return ker->evaluate(x1,x2); }
00128 
00129 
00130 real PrecomputedKernel::evaluate_i_j(int i, int j) const
00131 { 
00132 #ifdef BOUNDCHECK
00133     if(precomputedK.isNull())
00134         PLERROR("In PrecomputedKernel::evaluate_i_j data must first be set with setDataForKernelMatrix");
00135     else if(i<0 || j<0 || i>=data.length() || j>=data.length())
00136         PLERROR("In PrecomputedKernel::evaluate_i_j i (%d) and j (%d) must be between 0 and data.length() (%d)",
00137                 i, j, data.length());
00138 #endif
00139     return precomputedK[i][j];//[i*data.length()+j];
00140 }
00141 
00142 
00143 real PrecomputedKernel::evaluate_i_x(int i, const Vec& x, real squared_norm_of_x) const 
00144 { return ker->evaluate_i_x(i,x,squared_norm_of_x); }
00145 
00146 
00147 real PrecomputedKernel::evaluate_x_i(const Vec& x, int i, real squared_norm_of_x) const
00148 { return ker->evaluate_x_i(x,i,squared_norm_of_x); }
00149 
00150 void PrecomputedKernel::declareOptions(OptionList &ol)
00151 {
00152     declareOption(ol, "ker", &PrecomputedKernel::ker, OptionBase::buildoption,
00153                   "The underlying kernel.");    
00154     inherited::declareOptions(ol);
00155 }
00156 
00157 
00158 
00159 } // end of namespace PLearn
00160 
00161 
00162 /*
00163   Local Variables:
00164   mode:c++
00165   c-basic-offset:4
00166   c-file-style:"stroustrup"
00167   c-file-offsets:((innamespace . 0)(inline-open . 0))
00168   indent-tabs-mode:nil
00169   fill-column:79
00170   End:
00171 */
00172 // vim: filetype=cpp:expandtab:shiftwidth=4:tabstop=8:softtabstop=4:encoding=utf-8:textwidth=79 :
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Defines