train_sample_identifierT_train_sample_identifierTrainSampleIdentifierTrainSampleIdentifiertrain_sample_identifier (Operator)

Name

train_sample_identifierT_train_sample_identifierTrainSampleIdentifierTrainSampleIdentifiertrain_sample_identifier — Train a sample identifier.

Warning

train_sample_identifiertrain_sample_identifierTrainSampleIdentifierTrainSampleIdentifierTrainSampleIdentifiertrain_sample_identifier is obsolete and is only provided for reasons of backward compatibility. New applications should use the operators for Deep-Learning-based classification instead, for details see Deep Learning / Classification.

Signature

train_sample_identifier( : : SampleIdentifier, GenParamName, GenParamValue : )

Description

train_sample_identifiertrain_sample_identifierTrainSampleIdentifierTrainSampleIdentifierTrainSampleIdentifiertrain_sample_identifier trains the given SampleIdentifierSampleIdentifierSampleIdentifierSampleIdentifiersampleIdentifiersample_identifier using the training data, which has been set with add_sample_identifier_training_dataadd_sample_identifier_training_dataAddSampleIdentifierTrainingDataAddSampleIdentifierTrainingDataAddSampleIdentifierTrainingDataadd_sample_identifier_training_data.

For an explanation of the concept of sample-based identification see the introduction of chapter Legacy / Identification.

In order to be able to identify objects, the SampleIdentifierSampleIdentifierSampleIdentifierSampleIdentifiersampleIdentifiersample_identifier must be trained with a representative set of sample images of the objects to be identified (see add_sample_identifier_training_dataadd_sample_identifier_training_dataAddSampleIdentifierTrainingDataAddSampleIdentifierTrainingDataAddSampleIdentifierTrainingDataadd_sample_identifier_training_data).

To train the SampleIdentifierSampleIdentifierSampleIdentifierSampleIdentifiersampleIdentifiersample_identifier, first, all the relevant sample images of the objects to be identified must be added with add_sample_identifier_training_dataadd_sample_identifier_training_dataAddSampleIdentifierTrainingDataAddSampleIdentifierTrainingDataAddSampleIdentifierTrainingDataadd_sample_identifier_training_data. Alternatively, the preparation data can be reused. See prepare_sample_identifierprepare_sample_identifierPrepareSampleIdentifierPrepareSampleIdentifierPrepareSampleIdentifierprepare_sample_identifier for an explanation on how to reuse preparation data for the training of the sample identifier. Then, with the call of train_sample_identifiertrain_sample_identifierTrainSampleIdentifierTrainSampleIdentifierTrainSampleIdentifiertrain_sample_identifier, the SampleIdentifierSampleIdentifierSampleIdentifierSampleIdentifiersampleIdentifiersample_identifier is trained.

Note that the training itself is very fast and does not consume a lot of memory. Therefore, it is very easy to retrain the SampleIdentifierSampleIdentifierSampleIdentifierSampleIdentifiersampleIdentifiersample_identifier, e.g., after new objects were added or already trained objects were removed from the SampleIdentifierSampleIdentifierSampleIdentifierSampleIdentifiersampleIdentifiersample_identifier.

To retrain an existing sample identifier with new objects, just add the relevant sample images with add_sample_identifier_training_dataadd_sample_identifier_training_dataAddSampleIdentifierTrainingDataAddSampleIdentifierTrainingDataAddSampleIdentifierTrainingDataadd_sample_identifier_training_data and call train_sample_identifiertrain_sample_identifierTrainSampleIdentifierTrainSampleIdentifierTrainSampleIdentifiertrain_sample_identifier, again. To remove objects, use the operator remove_sample_identifier_training_dataremove_sample_identifier_training_dataRemoveSampleIdentifierTrainingDataRemoveSampleIdentifierTrainingDataRemoveSampleIdentifierTrainingDataremove_sample_identifier_training_data and retrain the sample identifier with train_sample_identifiertrain_sample_identifierTrainSampleIdentifierTrainSampleIdentifierTrainSampleIdentifiertrain_sample_identifier.

Note that the training of the sample identifier is only possible, if the sample identifier contains training data for at least two different training objects.

The generic parameters GenParamNameGenParamNameGenParamNameGenParamNamegenParamNamegen_param_name and GenParamValueGenParamValueGenParamValueGenParamValuegenParamValuegen_param_value are intended for future use, only.

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

SampleIdentifierSampleIdentifierSampleIdentifierSampleIdentifiersampleIdentifiersample_identifier (input_control) sample_identifier → (handle)

Handle of the sample identifier.

GenParamNameGenParamNameGenParamNameGenParamNamegenParamNamegen_param_name (input_control) attribute.name-array → (string)

Parameter name.

Default: []

List of values: []

GenParamValueGenParamValueGenParamValueGenParamValuegenParamValuegen_param_value (input_control) attribute.value-array → (real / string / integer)

Parameter value.

Default: []

List of values: []

Result

If the parameters are valid, the operator train_sample_identifiertrain_sample_identifierTrainSampleIdentifierTrainSampleIdentifierTrainSampleIdentifiertrain_sample_identifier returns the value 2 ( H_MSG_TRUE) . If necessary an exception is raised.