Calculate the class of a feature vector by a multilayer perceptron.
classify_class_mlp computes the best Num classes of the feature vector Features with the multilayer perceptron (MLP) MLPHandle and returns the classes in Class and the corresponding confidences (probabilities) of the classes in Confidence. Before calling classify_class_mlp, the MLP must be trained with train_class_mlp.
classify_class_mlp can only be called if the MLP is used as a classifier with OutputFunction = 'softmax' (see create_class_mlp). Otherwise, an error message is returned. classify_class_mlp corresponds to a call to evaluate_class_mlp and an additional step that extracts the best Num classes. As described with evaluate_class_mlp, the output values of the MLP can be interpreted as probabilities of the occurrence of the respective classes. Therefore, in most cases it should be sufficient to use Num = 1 in order to decide whether the probability of the best class is high enough. In some applications it may be interesting to also take the second best class into account (Num = 2), particularly if it can be expected that the classes show a significant degree of overlap.
|
MLPHandle (input_control) |
class_mlp -> integer |
| MLP handle. | |
|
Features (input_control) |
real-array -> real |
| Feature vector. | |
|
Num (input_control) |
real-array -> real |
| Number of best classes to determine. | |
| Default value: 1 | |
| Suggested values: 1, 2, 3, 4, 5 | |
|
Class (output_control) |
integer(-array) -> integer |
| Result of classifying the feature vector with the MLP. | |
|
Confidence (output_control) |
real(-array) -> real |
| Confidence(s) of the class(es) of the feature vector. | |
If the parameters are valid, the operator classify_class_mlp returns the value 2 (H_MSG_TRUE). If necessary an exception handling is raised.
classify_class_mlp is reentrant and processed without parallelization.
train_class_mlp, read_class_mlp
Christopher M. Bishop: ``Neural Networks for Pattern Recognition''; Oxford University Press, Oxford; 1995.
Andrew Webb: ``Statistical Pattern Recognition''; Arnold, London; 1999.
Foundation
