learn_sampset_boxT_learn_sampset_boxLearnSampsetBoxLearnSampsetBox (Operator)
Name
learn_sampset_boxT_learn_sampset_boxLearnSampsetBoxLearnSampsetBox — Train the classifier with one data set.
Warning
learn_sampset_boxlearn_sampset_boxLearnSampsetBoxLearnSampsetBoxLearnSampsetBox is obsolete and is only provided for
reasons of backward compatibility. New applications should use the
MLP, SVM, KNN or GMM operators instead.
Signature
void LearnSampsetBox(const HTuple& ClassifHandle, const HTuple& SampKey, const HTuple& Outfile, const HTuple& NSamples, const HTuple& StopError, const HTuple& ErrorN)
void HFeatureSet::LearnSampsetBox(const HClassBox& ClassifHandle, const HString& Outfile, Hlong NSamples, double StopError, Hlong ErrorN) const
void HFeatureSet::LearnSampsetBox(const HClassBox& ClassifHandle, const char* Outfile, Hlong NSamples, double StopError, Hlong ErrorN) const
void HFeatureSet::LearnSampsetBox(const HClassBox& ClassifHandle, const wchar_t* Outfile, Hlong NSamples, double StopError, Hlong ErrorN) const
(Windows only)
void HClassBox::LearnSampsetBox(const HFeatureSet& SampKey, const HString& Outfile, Hlong NSamples, double StopError, Hlong ErrorN) const
void HClassBox::LearnSampsetBox(const HFeatureSet& SampKey, const char* Outfile, Hlong NSamples, double StopError, Hlong ErrorN) const
void HClassBox::LearnSampsetBox(const HFeatureSet& SampKey, const wchar_t* Outfile, Hlong NSamples, double StopError, Hlong ErrorN) const
(Windows only)
static void HOperatorSet.LearnSampsetBox(HTuple classifHandle, HTuple sampKey, HTuple outfile, HTuple NSamples, HTuple stopError, HTuple errorN)
void HFeatureSet.LearnSampsetBox(HClassBox classifHandle, string outfile, int NSamples, double stopError, int errorN)
void HClassBox.LearnSampsetBox(HFeatureSet sampKey, string outfile, int NSamples, double stopError, int errorN)
Description
learn_sampset_boxlearn_sampset_boxLearnSampsetBoxLearnSampsetBoxLearnSampsetBox trains the classifier with data for the
key SampKeySampKeySampKeySampKeysampKey (see read_sampsetread_sampsetReadSampsetReadSampsetReadSampset). The training sequence
is terminated at least after NSamplesNSamplesNSamplesNSamplesNSamples examples.
If NSamplesNSamplesNSamplesNSamplesNSamples is bigger than the number of examples in
SampKeySampKeySampKeySampKeysampKey, then a cyclic start at the beginning occurs.
If the error underpasses the value StopErrorStopErrorStopErrorStopErrorstopError, then the training
sequence is prematurely terminated. StopErrorStopErrorStopErrorStopErrorstopError is calculated with
N / ErrorN. Whereby N means the number of examples which were wrong
classified during the last ErrorNErrorNErrorNErrorNerrorN training examples.
Typically ErrorNErrorNErrorNErrorNerrorN is the number of examples in SampKeySampKeySampKeySampKeysampKey
and NSamplesNSamplesNSamplesNSamplesNSamples is a multiple of it.
If you want a data set with 100 examples to run 5 times at most and if you
want it to terminate with an error lower than 5%, then the corresponding
values are NSamplesNSamplesNSamplesNSamplesNSamples = 500, ErrorNErrorNErrorNErrorNerrorN = 100 and
StopErrorStopErrorStopErrorStopErrorstopError = 0.05.
A protocol of the training activity is going to be written in file
OutfileOutfileOutfileOutfileoutfile.
Execution Information
- Multithreading type: reentrant (runs in parallel with non-exclusive operators).
- Multithreading scope: global (may be called from any thread).
- Processed without parallelization.
This operator modifies the state of the following input parameter:
During execution of this operator, access to the value of this parameter must be synchronized if it is used across multiple threads.
Parameters
ClassifHandleClassifHandleClassifHandleClassifHandleclassifHandle (input_control, state is modified) class_box → HClassBox, HTupleHTupleHtuple (handle) (IntPtr) (HHandle) (handle)
Handle of the classifier.
SampKeySampKeySampKeySampKeysampKey (input_control) feature_set → HFeatureSet, HTupleHTupleHtuple (handle) (IntPtr) (HHandle) (handle)
Number of the data set to train.
OutfileOutfileOutfileOutfileoutfile (input_control) filename.write → HTupleHTupleHtuple (string) (string) (HString) (char*)
Name of the protocol file.
Default value:
'training_prot'
"training_prot"
"training_prot"
"training_prot"
"training_prot"
NSamplesNSamplesNSamplesNSamplesNSamples (input_control) integer → HTupleHTupleHtuple (integer) (int / long) (Hlong) (Hlong)
Number of arrays of attributes to learn.
Default value: 500
StopErrorStopErrorStopErrorStopErrorstopError (input_control) real → HTupleHTupleHtuple (real) (double) (double) (double)
Classification error for termination.
Default value: 0.05
ErrorNErrorNErrorNErrorNerrorN (input_control) integer → HTupleHTupleHtuple (integer) (int / long) (Hlong) (Hlong)
Error during the assignment.
Default value: 100
Result
learn_sampset_boxlearn_sampset_boxLearnSampsetBoxLearnSampsetBoxLearnSampsetBox returns 2 (H_MSG_TRUE).
An exception is raised if key SampKeySampKeySampKeySampKeysampKey does not exist
or there are problems while opening the file.
Possible Predecessors
create_class_boxcreate_class_boxCreateClassBoxCreateClassBoxCreateClassBox
Possible Successors
test_sampset_boxtest_sampset_boxTestSampsetBoxTestSampsetBoxTestSampsetBox,
enquire_class_boxenquire_class_boxEnquireClassBoxEnquireClassBoxEnquireClassBox,
write_class_boxwrite_class_boxWriteClassBoxWriteClassBoxWriteClassBox,
close_class_boxclose_class_boxCloseClassBoxCloseClassBoxCloseClassBox,
clear_sampsetclear_sampsetClearSampsetClearSampsetClearSampset
See also
test_sampset_boxtest_sampset_boxTestSampsetBoxTestSampsetBoxTestSampsetBox,
enquire_class_boxenquire_class_boxEnquireClassBoxEnquireClassBoxEnquireClassBox,
learn_class_boxlearn_class_boxLearnClassBoxLearnClassBoxLearnClassBox,
read_sampsetread_sampsetReadSampsetReadSampsetReadSampset
Module
Foundation