Weka[]Bagging源代码分析研究.docVIP

个人收集整理 仅供参考学习 个人收集整理 仅供参考学习 PAGE / NUMPAGES 个人收集整理 仅供参考学习 Weka[25] Bagging源代码分析 作者：Koala++/屈伟 先翻译一段 Bagging地介绍，Breiman地 bagging算法，是 bootstrap aggregating地缩写，b5E2RGbCAP 是最早地 Ensemble算法之一，它也是最直接容易实现，又有着另人惊讶地好效果地算法之 一.Bagging中地多样性是由有放回抽取训练样本来实现地，用这种方式随机产生多个训练 数据地子集，在每一个训练集地子集上训练一个同种分类器，最终分类结果是由多个分类器 地分类结果多数投票而产生地.Breiman’s bagging, short for bootstrap aggregating, is one ofp1EanqFDPw the earliest ensemble based algorithms. It is also one of the most intuitive and simplest toDXDiTa9E3d implement, with a surprisingly good performance . Diversity in bagging is obtained by usingRTCrpUDGiT bootstrapped replicas of the training data: different training data subsets are randomly5PCzVD7HxA drawn—with replacement—from the entire training data. Each training data subset is used tojLBHrnAILg train a different classifier of the same type. Individual classifiers are then combined by taking axHAQX74J0X majority vote of their decisions. For any given instance, the class chosen by most classifiers is theLDAYtRyKfE ensemble decision. Bagging 类在 weka.classifiers.meta 包下面. Bagging 继承 自 RandomizeableInteratedSingleClassifierEnhancer，而它又继承自 IteratedSingleClassifierEnhancer，Zzz6ZB2Ltk 它再继承自 SingleClassifierEnhancer，最后一个继承自 Classifier.我地 UML工具似乎过期了，dvzfvkwMI1 有空补上. 看一下构造函数： public Bagging() { m_Classifier = new weka.classifiers.trees.REPTree();rqyn14ZNXI } 可以看到默认地基分类器是 REPTree. 接下来看 buildClassifier函数： // can classifier handle the data? getCapabilities().testWithFail(data); // remove instances with missing class data = new Instances(data); data.deleteWithMissingClass(); super.buildClassifier(data); if (m_CalcOutOfBag (m_BagSizePercent != 100)) { throw new IllegalArgumentException(Bag size needs to be 100% if EmxvxOtOco + out-of-bag error is to be calculated!); } 只有一行代码值得看一下 super.buildClassifier： public void buildClassifier(Instances data) throws Exception {SixE2yXPq5 if (m_Classifier == null) { throw new Exception(A base classifier has not been specified!);6ewMyirQFL } m_Classifiers = Cl