Changeset 1176
- Timestamp:
- Feb 27, 2008, 5:58:15 PM (14 years ago)
- File:
-
- 1 edited
Legend:
- Unmodified
- Added
- Removed
-
trunk/yat/classifier/SupervisedClassifier.h
r1162 r1176 43 43 44 44 /// 45 /// @brief Interface class for supervised classifiers 45 /// @brief Interface class for supervised classifiers that use data 46 /// in a matrix format with data points as columns and each row 47 /// corresponding to a variable for the data points. Supervised 48 /// classifiers that do not use data in this format include 49 /// kernel-based classifiers such as SVM. 46 50 /// 47 48 51 class SupervisedClassifier 49 52 { … … 63 66 64 67 /// 65 /// An interface for making new classifier objects. This function 66 /// allows for specification at run-time of which classifier to 67 /// instatiate (see 'Prototype' in Design Patterns). 68 /// An interface for making new %classifier objects. This function 69 /// allows for specification at run-time of which %classifier type 70 /// to instatiate (see 'Prototype' in Design Patterns). Derived 71 /// classes should implement this function with DerivedClass* as 72 /// the return type and not SupervisedClassifier*, and a 73 /// dynamically allocated %classifier should be returned. The 74 /// implementation of this function should correspond to a copy 75 /// constructor with the exception that the returned %classifier 76 /// is not trained. 68 77 /// 69 /// @note Returns a dynamically allocated SupervisedClassifier, which has78 /// @note Returns a dynamically allocated %classifier, which has 70 79 /// to be deleted by the caller to avoid memory leaks. 71 80 /// … … 87 96 88 97 /// 89 /// Train the classifier. 98 /// Train the %classifier using unweighted training data and 99 /// targets. The training data \a data should have one column per 100 /// training sample and one row for each variable measured for the 101 /// training samples. The size of \a target should be the number 102 /// of samples in \a data and \a target should contain the class 103 /// for each sample ordered in the same order as columns in \a data. 90 104 /// 91 virtual void train(const MatrixLookup& , const Target&)=0;105 virtual void train(const MatrixLookup& data, const Target& targets)=0; 92 106 93 107 /// 94 /// Train the classifier. 108 /// Train the %classifier using weighted training data and 109 /// targets. Both \a data and \a targets should follow the 110 /// description for train(const MatrixLookup& data, const Target& targets) 95 111 /// 96 virtual void train(const MatrixLookupWeighted& , const Target&)=0;112 virtual void train(const MatrixLookupWeighted& data, const Target& targets)=0; 97 113 98 114 };
Note: See TracChangeset
for help on using the changeset viewer.