HALCON Reference Manual 10.0.2
Table of Contents / Classification / Neural Nets ClassesClassesClasses | | | Operators

evaluate_class_mlpT_evaluate_class_mlpevaluate_class_mlpEvaluateClassMlpEvaluateClassMlp (Operator)

Name

evaluate_class_mlpT_evaluate_class_mlpevaluate_class_mlpEvaluateClassMlpEvaluateClassMlp — Calculate the evaluation of a feature vector by a multilayer perceptron.

Signature

evaluate_class_mlp( : : MLPHandle, Features : Result)

Herror T_evaluate_class_mlp(const Htuple MLPHandle, const Htuple Features, Htuple* Result)

Herror evaluate_class_mlp(const HTuple& MLPHandle, const HTuple& Features, HTuple* Result)

HTuple HClassMlp::EvaluateClassMlp(const HTuple& Features) const

void HOperatorSetX.EvaluateClassMlp(
[in] VARIANT MLPHandle, [in] VARIANT Features, [out] VARIANT* Result)

VARIANT HClassMlpX.EvaluateClassMlp([in] VARIANT Features)

static void HOperatorSet.EvaluateClassMlp(HTuple MLPHandle, HTuple features, out HTuple result)

HTuple HClassMlp.EvaluateClassMlp(HTuple features)

Description

evaluate_class_mlpevaluate_class_mlpevaluate_class_mlpEvaluateClassMlpEvaluateClassMlp computes the result ResultResultResultResultresult of evaluating the feature vector FeaturesFeaturesFeaturesFeaturesfeatures with the multilayer perceptron (MLP) MLPHandleMLPHandleMLPHandleMLPHandleMLPHandle. The formulas used for the evaluation are described with create_class_mlpcreate_class_mlpcreate_class_mlpCreateClassMlpCreateClassMlp. Before calling evaluate_class_mlpevaluate_class_mlpevaluate_class_mlpEvaluateClassMlpEvaluateClassMlp, the MLP must be trained with train_class_mlptrain_class_mlptrain_class_mlpTrainClassMlpTrainClassMlp.

If the MLP is used for regression (function approximation), i.e., if (OutputFunction = 'linear'"linear""linear""linear""linear"), ResultResultResultResultresult is the value of the function at the coordinate FeaturesFeaturesFeaturesFeaturesfeatures. For OutputFunction = 'logistic'"logistic""logistic""logistic""logistic" and 'softmax'"softmax""softmax""softmax""softmax", the values in ResultResultResultResultresult can be interpreted as probabilities. Hence, for OutputFunction = 'logistic'"logistic""logistic""logistic""logistic" the elements of ResultResultResultResultresult represent the probabilities of the presence of the respective independent attributes. Typically, a threshold of 0.5 is used to decide whether the attribute is present or not. Depending on the application, other thresholds may be used as well. For OutputFunction = 'softmax'"softmax""softmax""softmax""softmax" usually the position of the maximum value of ResultResultResultResultresult is interpreted as the class of the feature vector, and the corresponding value as the probability of the class. In this case, classify_class_mlpclassify_class_mlpclassify_class_mlpClassifyClassMlpClassifyClassMlp should be used instead of evaluate_class_mlpevaluate_class_mlpevaluate_class_mlpEvaluateClassMlpEvaluateClassMlp because classify_class_mlpclassify_class_mlpclassify_class_mlpClassifyClassMlpClassifyClassMlp directly returns the class and corresponding probability.

Parallelization

Parameters

MLPHandleMLPHandleMLPHandleMLPHandleMLPHandle (input_control)  class_mlp → HClassMlp, HTupleHClassMlp, HTupleHClassMlpX, VARIANTHtuple (integer) (IntPtr) (Hlong) (Hlong) (Hlong)

MLP handle.

FeaturesFeaturesFeaturesFeaturesfeatures (input_control)  real-array → HTupleHTupleVARIANTHtuple (real) (double) (double) (double) (double)

Feature vector.

ResultResultResultResultresult (output_control)  real-array → HTupleHTupleVARIANTHtuple (real) (double) (double) (double) (double)

Result of evaluating the feature vector with the MLP.

Result

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

Possible Predecessors

train_class_mlptrain_class_mlptrain_class_mlpTrainClassMlpTrainClassMlp, read_class_mlpread_class_mlpread_class_mlpReadClassMlpReadClassMlp

Alternatives

classify_class_mlpclassify_class_mlpclassify_class_mlpClassifyClassMlpClassifyClassMlp

See also

create_class_mlpcreate_class_mlpcreate_class_mlpCreateClassMlpCreateClassMlp

References

Christopher M. Bishop: “Neural Networks for Pattern Recognition”; Oxford University Press, Oxford; 1995.
Andrew Webb: “Statistical Pattern Recognition”; Arnold, London; 1999.

Module

Foundation

Table of Contents / Classification / Neural Nets ClassesClassesClasses | | | Operators
HALCON Reference Manual 10.0.2 Copyright © 1996-2011 MVTec Software GmbH