Classification Library (ClassificationLib)


getErrorRatio
to calculate error ratio from classification test




FUNCTIONS

double - getErrorRatio(int[] resultclass, int[] correctclass)




EXAMPLES

Example

VectorLib vlib=new VectorLib(); ClassificationLib clib=new ClassificationLib(); Dataset dataset=new Dataset("ruspini", "supervised"); int n_test = 5; int[][] testdata = new int[n_test][dataset.dimensi]; int[] resultclass = new int[n_test]; int[] originalclass = new int[n_test]; int k = 4; for (int i=0; i<n_test; i++) {      int rand_row = vlib.getRandom(0, dataset.n);      originalclass[i]=dataset.label[rand_row];      testdata[i] = vlib.getRow(dataset.data, rand_row);      resultclass[i] = clib.NearestNeighbors(dataset.data, dataset.label, testdata[i], k); } double error=clib.getErrorRatio(resultclass, originalclass); vlib.view("Test data", testdata); vlib.view("k of Nearest Neighbors", k); vlib.view("Error ratio", error);

Output

Test data = 13.0      69.0 117.0      115.0 15.0      75.0 72.0      31.0 108.0      116.0 k of Nearest Neighbors = 4 Error ratio = 0.0