úÎ!!|&`      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_None GMÜHSvm"Managed type for struct svm_model.7 ! "#$%&'()*+,-./01234567! ,-.12/ 03456+*)('&%$#"None?HSvm&Extra parameters of SVM implementationDHSvmSVM Algorithm with parametersEHSvmc-SVC algorithmFHSvmnu-SVC algorithmGHSvmnu-SVR algorithmHHSvmeps-SVR algorithmIHSvm One class SVMMHSvm"Kernel function for SVM algorithm.NHSvm(Linear kernel function, i.e. dot productOHSvm.Gaussian radial basis function with parameter RPHSvmSigmoid kernel functionQHSvm!Inhomogeneous polynomial functionUHSvmU& is a wrapper over foreign pointer to VHSvmBSVM problem is a list of maps from training vectors to 1.0 or -1.0WHSvmdVector type provides a sparse implementation of vector. It uses IntMap as underlying implementation.`HSvm.Default extra parameters of SVM implamentationXHSvmLike Y but with extra parametersYHSvmThe Y function allows training a U starting from a V by specifying an D and a MZHSvmLike  crossvalidate but with extra parameters[HSvmStratified cross validation\HSvmSave model to the file]HSvmLoad model from the file^HSvmPredict a value for W by using U_HSvm7Wrapper to change the libsvm output reporting function.olibsvm by default writes some statistics to stdout. If you don't want any output from libsvm, you can do e.g.:@withPrintFn (\_ -> return ()) $ train (NuSvc 0.25) (RBF 1) feats"?@BCADEFGHIJKLMNOPQTRSUVWXYZ[\]^_"WVMNOPQTRSDEFGHIJKL?@BCAUYX[Z]\^_a      !"#$%&'()*+,-./0123456789:;<==>?@ABCD EFGHIJ  KLMNOPQRSTUVW&HSvm-0.1.1.3.22-DfDeE8htVzeIojVsRqkwOM Data.SVM.RawData.SVM CSvmPrintFn CSvmModel CSvmParametersvm_type kernel_typedegreegammacoef0 cache_sizeepsc nr_weight weight_labelweightnup shrinking probability CKernelType unCKernelTypeCSvmType unCSvmType CSvmProblemlyxCSvmNodeindexvaluecreateSvmPrintFnPtrc_svm_set_print_string_functionc_clone_model_support_vectorsc_svm_destroy_modelc_svm_check_parameterc_svm_load_modelc_svm_save_model c_svm_predictc_svm_cross_validation c_svm_traincSvcnuSvconeClass epsilonSvrnuSvrlinearpolyrbfsigmoid precomputed defaultCParam$fStorableCSvmNode$fStorableCSvmProblem$fStorableCSvmParameter$fStorableCSvmType$fShowCSvmType$fStorableCKernelType$fShowCKernelType$fShowCSvmParameter ExtraParam cacheSize AlgorithmCSvcNuSvcNuSvr EpsilonSvr OneClassSvmepsilon KernelTypeLinearRBFSigmoidPolyModelProblemVectortrain'traincrossValidate' crossValidate saveModel loadModelpredict withPrintFn defaultExtra