@author Hua Ouyang @file README Relevance Vector Machines (Multiclass classification; Regression) Files in this folder: README: this file # Source codes: build.py: build scripts rvm_main.cc: contains main function of RVM rvm.h: contains classification/regression main routines sbl_est.h: sequential minimal optimization # Unit test data: synthClassTrain.csv: a classification training dataset. synthClassTest.csv: a classification testing dataset for synthClassTrain. sincTrain.csv: a regression training dataset. sincTest.csv: a regression testing dataset for sincTrain. # Generated files: artificialdata.csv: artificial data if no training/testing data file provided rvm_model: RVM model file predicted_values: predicte dvalues Explanation of arguments: 0. learner_name (required): the name of the support vecotr learner, values: "rvm_c" for classification, "rvm_r" for regression 1. mode (required) : the mode of rvm_main, values: "cv", "train", "train_test", "test". 2. k_cv (optional): the number of folds for cross validation, only required under "cv" mode. 3. cv_data (optional): file name for cross validation data, only required under "cv" mode. 4. train_data (optional): file name for training data, only required under "train" or "train_test" mode. 5. test_data (optional): file name for testing data, only required under "test" or "train_test" mode. 6. kernel (required): kernel name, values:"linear", "gaussian". 7. sigma (optional): sigma in the gaussian kernel k(x1,x2)=exp(-(x1-x2)^2/(2sigma^2)), only required when using "guassian" kernel. 8. c (required): the weight that controls compromise between large margins and small margin violations. 9. normalize (optional): where need to do data normalization before training/testing, values: "0" for no normalize, "1" for normalize. Examples: --------------------------------------------------------------------------------------------------- b1.Relevance Vector Regression, Cross validation mode rvm_main --learner_name=rvm_r --mode=cv --k_cv=4 --cv_data=sincTrain.csv --kernel=gaussian --sigma=0.1 --c=10 --normalize=0 rvm_main --learner_name=rvm_r --mode=cv --k_cv=4 --cv_data=sincTrain.csv --kernel=linear --c=10 --normalize=0 b2.Relevance Vector Regression, Training mode (model will be saved as "rvm_model") rvm_main --learner_name=rvm_r --mode=train --train_data=sincTrain.csv --kernel=gaussian --sigma=0.1 --c=10 --normalize=0 rvm_main --learner_name=rvm_r --mode=train --train_data=sincTrain.csv --kernel=linear --c=10 --normalize=0 b3.Relevance Vector Regression, Training+testing mode rvm_main --learner_name=rvm_r --mode=train_test --train_data=sincTrain.csv --test_data=sincTest.csv --kernel=gaussian --sigma=0.1 --c=10 --normalize=0 rvm_main --learner_name=rvm_r --mode=train_test --train_data=sincTrain.csv --test_data=sincTest.csv --kernel=linear --c=10 --normalize=0 b4.Relevance Vector Regression, Testing mode (the model file "rvm_model" should exist) rvm_main --learner_name=rvm_r --mode=test --test_data=sincTest.csv --kernel=gaussian --sigma=0.1 --c=10 --normalize=0 rvm_main --learner_name=rvm_r --mode=test --test_data=sincTest.csv --kernel=linear --c=10 --normalize=0 --------------------------------------------------------------------------------------------------- TODO: a1.Relevance Vector Classification, Stratified cross validation mode rvm_main --learner_name=rvm_c --mode=cv --k_cv=4 --cv_data=synthClassTrain.csv --kernel=gaussian --sigma=0.1 --c=10 --normalize=0 rvm_main --learner_name=rvm_c --mode=cv --k_cv=4 --cv_data=synthClassTrain.csv --kernel=linear --c=10 --normalize=0 a2.Relevance Vector Classification, Training mode (model will be saved as "rvm_model") rvm_main --learner_name=rvm_c --mode=train --train_data=synthClassTrain.csv --kernel=gaussian --sigma=0.1 --c=10 --normalize=0 rvm_main --learner_name=rvm_c --mode=train --train_data=synthClassTrain.csv --kernel=linear --c=10 --normalize=0 a3.Relevance Vector Classification, Training+testing mode rvm_main --learner_name=rvm_c --mode=train_test --train_data=synthClassTrain.csv --test_data=synthClassTest.csv --kernel=gaussian --sigma=0.1 --c=10 --normalize=0 rvm_main --learner_name=rvm_c --mode=train_test --train_data=synthClassTrain.csv --test_data=synthClassTest.csv --kernel=linear --c=10 --normalize=0 a4.Relevance Vector Classification, Testing mode (the model file "rvm_model" should exist) rvm_main --learner_name=rvm_c --mode=test --test_data=synthClassTest.csv --kernel=gaussian --sigma=0.1 --c=10 --normalize=0 rvm_main --learner_name=rvm_c --mode=test --test_data=synthClassTest.csv --kernel=linear --c=10 --normalize=0