MATLAB随机森林回归模型：
调用matlab自带的TreeBagger.m
T=textread(\'E:\datasets-orreview\discretized-regression\10bins\abalone10\matlab\test_abalone10.2\'); X=textread(\'E:\datasets-orreview\discretized-regression\10bins\abalone10\matlab\train_abalone10.2\'); %nTree = round(sqrt(size(X,2)-1)); nTree = 50; train_data = X(:,1:end-1);train_label = X(:,end); test_data = T(:,1:end-1); Factor = TreeBagger(nTree, train_data, train_label,\'Method\',\'regression\'); [Predict_label,Scores] = predict(Factor, test_data); %Predict_label=cellfun(@str2num,Predict_label(1:end)); MZE = mean(round(Predict_label) ~= T(:,end)) MAE = mean(abs(round(Predict_label) - T(:,end)))
调用外部函数forestTrain.m来自https://github.com/karpathy/Random-Forest-Matlab
T=textread(\'E:\datasets-orreview\ordinal-regression\ERA\matlab\test_ERA.1\'); X=textread(\'E:\datasets-orreview\ordinal-regression\ERA\matlab\train_ERA.1\'); opts= struct; opts.depth= 9; opts.numTrees= 60; opts.numSplits= 5; opts.verbose= true; opts.classifierID= 2; % weak learners to use. Can be an array for mix of weak learners too train_data = X(:,1:end-1);train_label = X(:,end); test_data = T(:,1:end-1); tic; m= forestTrain(train_data, train_label, opts); timetrain= toc; tic; yhatTrain = forestTest(m, test_data); timetest= toc; MZE = mean(round(yhatTrain) ~= T(:,end)) MAE = mean(abs(round(yhatTrain) - T(:,end)))
