read_samples_class_svm — Read the training data of a support vector machine from a file.
read_samples_class_svm reads training samples from the file given by FileName and adds them to the training samples that have already been added to the support vector machine (SVM) given by SVMHandle. The SVM must be created with create_class_svm before calling read_samples_class_svm. As described with train_class_svm and write_samples_class_svm, the operators read_samples_class_svm, add_sample_class_svm, and write_samples_class_svm can be used to build up a extensive set of training samples, and hence to improve the performance of the SVM by retraining the SVM with extended data sets.
It should be noted that the training samples must have the correct dimensionality. The feature vectors and target vectors stored in FileName must have the lengths NumFeatures and NumClasses that were specified with create_class_svm. The target is stored in vector form for compatibility reason with the MLP (see read_samples_class_mlp). If the dimensions are incorrect an error message is returned.
This operator modifies the state of the following input parameter:
If the parameters are valid the operator read_samples_class_svm returns the value 2 (H_MSG_TRUE). If necessary, an exception is raised.