通过源码学算法--AdaBoost (CART) -- Example_1.m

 Ionosphere.txt中的数据是351x35，每一行代表一个采样数据，总共351个采样。
 数据空间是34维（最后一列是该采样所属类别），所以下面要转置一下

% Step1: reading Data from the file
file_data = load('Ionosphere.txt');
Data = file_data(:,1:end-1)';
Labels = file_data(:, end)';
Labels = Labels*2 - 1;

MaxIter = 100; % boosting iterations

把这些数据平分成两份：训练(176)和测试(175)

% Step2: splitting data to training and control set
TrainData   = Data(:,1:2:end);
TrainLabels = Labels(1:2:end);

ControlData   = Data(:,2:2:end);
ControlLabels = Labels(2:2:end);

每个弱分类器都是一棵最大深度为3的树节点

% Step3: constructing weak learner
weak_learner = tree_node_w(3); % pass the number of tree splits to the constructor

计时，开始训练

tic
% Step4: training with Gentle AdaBoost
[RLearners RWeights] = RealAdaBoost(weak_learner, TrainData, TrainLabels, MaxIter);
realElapsed = toc

tic
% Step5: training with Modest AdaBoost
[MLearners MWeights] = ModestAdaBoost(weak_learner, TrainData, TrainLabels, MaxIter);
modestElapsed = toc

用测试数据验证训练好的分类器，输出分类错误率

% Step6: evaluating on control set
ResultR = sign(Classify(RLearners, RWeights, ControlData));

ResultM = sign(Classify(MLearners, MWeights, ControlData));

% Step7: calculating error
ErrorR  = sum(ControlLabels ~= ResultR) / length(ControlLabels)

ErrorM  = sum(ControlLabels ~= ResultM) / length(ControlLabels)