wolffd@0: wolffd@0: wolffd@0: wolffd@0: Netlab Reference Manual knnfwd wolffd@0: wolffd@0: wolffd@0: wolffd@0:

knnfwd wolffd@0:

wolffd@0:

wolffd@0: Purpose wolffd@0:

wolffd@0: Forward propagation through a K-nearest-neighbour classifier. wolffd@0: wolffd@0:

wolffd@0: Synopsis wolffd@0:

wolffd@0:

wolffd@0: 
wolffd@0: [y, l] = knnfwd(net, x)
wolffd@0:

wolffd@0: wolffd@0: wolffd@0:

wolffd@0: Description wolffd@0:

wolffd@0: [y, l] = knnfwd(net, x) takes a matrix x wolffd@0: of input vectors (one vector per row) wolffd@0: and uses the k-nearest-neighbour rule on the training data contained wolffd@0: in net to wolffd@0: produce wolffd@0: a matrix y of outputs and a matrix l of classification wolffd@0: labels. wolffd@0: The nearest neighbours are determined using Euclidean distance. wolffd@0: The ijth entry of y counts the number of occurrences that wolffd@0: an example from class j is among the k closest training wolffd@0: examples to example i from x. wolffd@0: The matrix l contains the predicted class labels wolffd@0: as an index 1..N, not as 1-of-N coding. wolffd@0: wolffd@0:

wolffd@0: Example wolffd@0:

wolffd@0:

wolffd@0: 
wolffd@0: net = knn(size(xtrain, 2), size(t_train, 2), 3, xtrain, t_train);
wolffd@0: y = knnfwd(net, xtest);
wolffd@0: conffig(y, t_test);
wolffd@0:

wolffd@0: wolffd@0: Creates a 3 nearest neighbour model net and then applies it to wolffd@0: the data xtest. The results are plotted as a confusion matrix with wolffd@0: conffig. wolffd@0: wolffd@0:

wolffd@0: See Also wolffd@0:

wolffd@0: kmeans, knn

wolffd@0: Pages: wolffd@0: Index wolffd@0:

wolffd@0: