Operator Reference

get_support_vector_num_class_svmT_get_support_vector_num_class_svmGetSupportVectorNumClassSvmGetSupportVectorNumClassSvmget_support_vector_num_class_svm (Operator)

get_support_vector_num_class_svmT_get_support_vector_num_class_svmGetSupportVectorNumClassSvmGetSupportVectorNumClassSvmget_support_vector_num_class_svm — Return the number of support vectors of a support vector machine.

Signature

get_support_vector_num_class_svm( : : SVMHandle : NumSupportVectors, NumSVPerSVM)

Herror T_get_support_vector_num_class_svm(const Htuple SVMHandle, Htuple* NumSupportVectors, Htuple* NumSVPerSVM)

void GetSupportVectorNumClassSvm(const HTuple& SVMHandle, HTuple* NumSupportVectors, HTuple* NumSVPerSVM)

Hlong HClassSvm::GetSupportVectorNumClassSvm(HTuple* NumSVPerSVM) const

static void HOperatorSet.GetSupportVectorNumClassSvm(HTuple SVMHandle, out HTuple numSupportVectors, out HTuple numSVPerSVM)

int HClassSvm.GetSupportVectorNumClassSvm(out HTuple numSVPerSVM)

def get_support_vector_num_class_svm(svmhandle: HHandle) -> Tuple[int, Sequence[int]]

Description

get_support_vector_num_class_svmget_support_vector_num_class_svmGetSupportVectorNumClassSvmGetSupportVectorNumClassSvmget_support_vector_num_class_svm returns in NumSupportVectorsNumSupportVectorsNumSupportVectorsnumSupportVectorsnum_support_vectors the number of support vectors that are stored in the support vector machine (SVM) given by SVMHandleSVMHandleSVMHandleSVMHandlesvmhandle. get_support_vector_num_class_svmget_support_vector_num_class_svmGetSupportVectorNumClassSvmGetSupportVectorNumClassSvmget_support_vector_num_class_svm should be called before the labels of individual support vectors are read out with get_support_vector_class_svmget_support_vector_class_svmGetSupportVectorClassSvmGetSupportVectorClassSvmget_support_vector_class_svm, e.g., for visualizing which the training data become a SV (see get_support_vector_class_svmget_support_vector_class_svmGetSupportVectorClassSvmGetSupportVectorClassSvmget_support_vector_class_svm). The number of SVs in each classifier is listed in NumSVPerSVMNumSVPerSVMNumSVPerSVMnumSVPerSVMnum_svper_svm. The reason that its sum differs from the Number obtained in NumSupportVectorsNumSupportVectorsNumSupportVectorsnumSupportVectorsnum_support_vectors is that SV evaluations are reused throughout different sub-classifiers. NumSVPerSVMNumSVPerSVMNumSVPerSVMnumSVPerSVMnum_svper_svm provides the possibility for controlling the process of speeding up SVM classification time with the operator reduce_class_svmreduce_class_svmReduceClassSvmReduceClassSvmreduce_class_svm.

Execution Information

  • Multithreading type: reentrant (runs in parallel with non-exclusive operators).
  • Multithreading scope: global (may be called from any thread).
  • Processed without parallelization.

Parameters

SVMHandleSVMHandleSVMHandleSVMHandlesvmhandle (input_control)  class_svm HClassSvm, HTupleHHandleHTupleHtuple (handle) (IntPtr) (HHandle) (handle)

SVM handle.

NumSupportVectorsNumSupportVectorsNumSupportVectorsnumSupportVectorsnum_support_vectors (output_control)  integer HTupleintHTupleHtuple (integer) (int / long) (Hlong) (Hlong)

Total number of support vectors.

NumSVPerSVMNumSVPerSVMNumSVPerSVMnumSVPerSVMnum_svper_svm (output_control)  integer-array HTupleSequence[int]HTupleHtuple (integer) (int / long) (Hlong) (Hlong)

Number of SV of each sub-SVM.

Result

If SVMHandleSVMHandleSVMHandleSVMHandlesvmhandle is valid the operator get_sample_num_class_svmget_sample_num_class_svmGetSampleNumClassSvmGetSampleNumClassSvmget_sample_num_class_svm returns the value 2 ( H_MSG_TRUE) . If necessary, an exception is raised.

Possible Predecessors

train_class_svmtrain_class_svmTrainClassSvmTrainClassSvmtrain_class_svm

Possible Successors

get_sample_class_svmget_sample_class_svmGetSampleClassSvmGetSampleClassSvmget_sample_class_svm

See also

create_class_svmcreate_class_svmCreateClassSvmCreateClassSvmcreate_class_svm

Module

Foundation