Repository files navigation

• Download the code:git clone https://github.com/qiyiping/gbdt.git
• Runmake to compile
• Run the demo script in `test`:./demo.sh

[InitalGuess] Label Weight Index0:Value0 Index1:Value1 ..

Each line contains an instance and is ended by a ‘\n’ character. Inital guess is optional. For two-class classification, Label is -1 or 1. For regression, Label is the target value, which can be any real number. Feature Index starts from 0. Feature Value can be any real number.

classConfigure {public:size_tnumber_of_feature;// number of featuressize_tmax_depth;// max depth for each treesize_titerations;// number of trees in gbdtdoubleshrinkage;// shrinkage parameterdoublefeature_sample_ratio;// portion of features to be spliteddoubledata_sample_ratio;// portion of data to be fitted in each iterationsize_tmin_leaf_size;// min number of nodes in leafLossloss;// loss typebooldebug;// show debug info?double*feature_costs;// mannually set feature costs in order to tune the modelboolenable_feature_tunning;// when set true, `feature_costs' is used to tune the modelboolenable_initial_guess;...};

• Friedman, J. H. “Greedy Function Approximation: A Gradient Boosting Machine.” (February 1999)
• Friedman, J. H. “Stochastic Gradient Boosting.” (March 1999)
• Jerry Ye, et al. (2009). Stochastic gradient boosted distributed decision trees. (Distributed implementation)