PLearn 0.1
|
00001 // -*- C++ -*- 00002 00003 // BaseRegressorWrapper.h 00004 // Copyright (c) 1998-2002 Pascal Vincent 00005 // Copyright (C) 1999-2002 Yoshua Bengio and University of Montreal 00006 // Copyright (c) 2002 Jean-Sebastien Senecal, Xavier Saint-Mleux, Rejean Ducharme 00007 // 00008 // Redistribution and use in source and binary forms, with or without 00009 // modification, are permitted provided that the following conditions are met: 00010 // 00011 // 1. Redistributions of source code must retain the above copyright 00012 // notice, this list of conditions and the following disclaimer. 00013 // 00014 // 2. Redistributions in binary form must reproduce the above copyright 00015 // notice, this list of conditions and the following disclaimer in the 00016 // documentation and/or other materials provided with the distribution. 00017 // 00018 // 3. The name of the authors may not be used to endorse or promote 00019 // products derived from this software without specific prior written 00020 // permission. 00021 // 00022 // THIS SOFTWARE IS PROVIDED BY THE AUTHORS ``AS IS'' AND ANY EXPRESS OR 00023 // IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES 00024 // OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN 00025 // NO EVENT SHALL THE AUTHORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, 00026 // SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED 00027 // TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR 00028 // PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 00029 // LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING 00030 // NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS 00031 // SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 00032 // 00033 // This file is part of the PLearn library. For more information on the PLearn 00034 // library, go to the PLearn Web site at www.plearn.org 00035 00036 00037 /* ******************************************************************************** 00038 * $Id: BaseRegressorWrapper.h, v 1.0 2004/07/19 10:00:00 Bengio/Kegl/Godbout * 00039 * This file is part of the PLearn library. * 00040 ******************************************************************************** */ 00041 00044 #ifndef BaseRegressorWrapper_INC 00045 #define BaseRegressorWrapper_INC 00046 00047 #include <plearn_learners/generic/PLearner.h> 00048 00049 namespace PLearn { 00050 using namespace std; 00051 class RegressionTree; 00052 class BaseRegressorConfidence; 00053 class RegressionTreeRegisters; 00054 00055 class BaseRegressorWrapper: public PLearner 00056 { 00057 typedef PLearner inherited; 00058 00059 private: 00060 00061 /* 00062 Build options: they have to be set before training 00063 */ 00064 00065 real loss_function_weight; // the hyper parameter to balance the error and the confidence factor 00066 int mean_imputation; // if set to 1, the algorithm will perform mean imputation based on the training set 00067 int regression_tree; // indicator to use the tree_regressor_template is set to 1, and the base_regressor_template otherwise 00068 int use_confidence_function; // indicator to use the confidence_function_template to build a confidence estimate 00069 int use_base_regressor_confidence; // indicator to use the confidence computed as the second output of the base regressor 00070 PP<PLearner> base_regressor_template; // template for a generic regressor as the base learner to be boosted 00071 PP<RegressionTree> tree_regressor_template; // template for a tree regressor to be boosted as the base regressor 00072 PP<BaseRegressorConfidence> confidence_function_template; // template for the confidence function to be learnt from the train set 00073 PP<RegressionTreeRegisters> sorted_train_set; // a sorted train set when using a regression tree 00074 00075 /* 00076 Learnt options: they are sized and initialized if need be, at stage 0 00077 */ 00078 00079 00080 PP<PLearner> base_regressor; // base regressors built at each boosting stage 00081 PP<RegressionTree> tree_regressor; // base regressors built at each boosting stage 00082 PP<BaseRegressorConfidence> confidence_function; // confidence function learnt from the train set 00083 VMat base_regressor_train_set; // a train set with the chosen missing value strategies applied for the base regressor 00084 TVec<real> variable_means; // the vector of variable means when doing mean imputation 00085 00086 /* 00087 Work fields: they are sized and initialized if need be, at buid time 00088 */ 00089 00090 00091 00092 public: 00093 BaseRegressorWrapper(); 00094 virtual ~BaseRegressorWrapper(); 00095 00096 PLEARN_DECLARE_OBJECT(BaseRegressorWrapper); 00097 00098 static void declareOptions(OptionList& ol); 00099 virtual void makeDeepCopyFromShallowCopy(CopiesMap &copies); 00100 virtual void build(); 00101 virtual void train(); 00102 virtual void forget(); 00103 virtual int outputsize() const; 00104 virtual TVec<string> getTrainCostNames() const; 00105 virtual TVec<string> getTestCostNames() const; 00106 virtual void computeOutput(const Vec& input, Vec& output) const; 00107 virtual void computeOutputAndCosts(const Vec& input, const Vec& target, Vec& output, Vec& costs) const; 00108 virtual void computeCostsFromOutputs(const Vec& input, const Vec& output, const Vec& target, Vec& costs) const; 00109 void setSortedTrainSet(PP<RegressionTreeRegisters> the_sorted_train_set); 00110 00111 private: 00112 void build_(); 00113 void verbose(string msg, int level); 00114 }; 00115 00116 DECLARE_OBJECT_PTR(BaseRegressorWrapper); 00117 00118 } // end of namespace PLearn 00119 00120 #endif 00121 00122 00123 /* 00124 Local Variables: 00125 mode:c++ 00126 c-basic-offset:4 00127 c-file-style:"stroustrup" 00128 c-file-offsets:((innamespace . 0)(inline-open . 0)) 00129 indent-tabs-mode:nil 00130 fill-column:79 00131 End: 00132 */ 00133 // vim: filetype=cpp:expandtab:shiftwidth=4:tabstop=8:softtabstop=4:encoding=utf-8:textwidth=79 :