get_support_vector_class_svmT_get_support_vector_class_svmGetSupportVectorClassSvmGetSupportVectorClassSvmget_support_vector_class_svm (Operator)
Name
get_support_vector_class_svmT_get_support_vector_class_svmGetSupportVectorClassSvmGetSupportVectorClassSvmget_support_vector_class_svm — Return the index of a support vector from a trained support vector
machine.
Signature
Description
The operator get_support_vector_class_svmget_support_vector_class_svmGetSupportVectorClassSvmGetSupportVectorClassSvmget_support_vector_class_svm maps a support
vector of a trained SVM (given in SVMHandleSVMHandleSVMHandleSVMHandlesvmhandle) to the
original training data set. The index of the SV is specified with
IndexSupportVectorIndexSupportVectorIndexSupportVectorindexSupportVectorindex_support_vector. The index is counted from 0, i.e.,
IndexSupportVectorIndexSupportVectorIndexSupportVectorindexSupportVectorindex_support_vector must be a number between 0 and
NumSupportVectorsNumSupportVectorsNumSupportVectorsnumSupportVectorsnum_support_vectors - 1, where
NumSupportVectorsNumSupportVectorsNumSupportVectorsnumSupportVectorsnum_support_vectors can be determined with
get_support_vector_num_class_svmget_support_vector_num_class_svmGetSupportVectorNumClassSvmGetSupportVectorNumClassSvmget_support_vector_num_class_svm. The index of this SV in
the training data is returned in IndexIndexIndexindexindex. This
IndexIndexIndexindexindex can be used for a query with
get_sample_class_svmget_sample_class_svmGetSampleClassSvmGetSampleClassSvmget_sample_class_svm to obtain the feature vectors that
become support vectors. get_sample_class_svmget_sample_class_svmGetSampleClassSvmGetSampleClassSvmget_sample_class_svm can, for
example, be used to visualize the support vectors.
Note that when using train_class_svmtrain_class_svmTrainClassSvmTrainClassSvmtrain_class_svm with a mode different from
'default'"default""default""default""default" or reducing the SVM with reduce_class_svmreduce_class_svmReduceClassSvmReduceClassSvmreduce_class_svm, the
returned IndexIndexIndexindexindex will always be -1, i.e., it will be invalid.
The reason for this is that a consistent mapping between SV and
training data becomes impossible.
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.
IndexSupportVectorIndexSupportVectorIndexSupportVectorindexSupportVectorindex_support_vector (input_control) integer → HTupleintHTupleHtuple (integer) (int / long) (Hlong) (Hlong)
Index of the stored support vector.
IndexIndexIndexindexindex (output_control) real → HTuplefloatHTupleHtuple (real) (double) (double) (double)
Index of the support vector in the training set.
Result
If the parameters are valid the operator get_sample_class_svmget_sample_class_svmGetSampleClassSvmGetSampleClassSvmget_sample_class_svm
returns the value 2 (
H_MSG_TRUE)
. If necessary, an exception is raised.
Possible Predecessors
train_class_svmtrain_class_svmTrainClassSvmTrainClassSvmtrain_class_svm,
get_support_vector_num_class_svmget_support_vector_num_class_svmGetSupportVectorNumClassSvmGetSupportVectorNumClassSvmget_support_vector_num_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