create_class_knn — Create a k-nearest neighbors (k-NN) classifier.
create_class_knn creates a k-nearest neighbors (k-NN) data structure. This can be either used to classify data or to approximately locate nearest neighbors in a NumDim-dimensional space.
Most of the operators described in Classification/K-Nearest-Neighbor use the resulting handle KNNHandle.
The k-NN classifies by searching approximately the nearest neighbors and returning their classes as result. With the used approximation, the search time is logarithmically to the number of samples and dimensions.
The dimension of the feature vectors is the only parameter that necessarily has to be set in NumDim.
Number of dimensions of the feature.
Default value: 10
Handle of the k-NN classifier.
If the parameters are valid, the operator create_class_knn returns the value 2 (H_MSG_TRUE). If necessary, an exception is raised.
Marius Muja, David G. Lowe: “Fast Approximate Nearest Neighbors with Automatic Algorithm Configuration”; International Conference on Computer Vision Theory and Applications (VISAPP 09); 2009.