Data Generators

There is a data generator framework is JKernelMachines to help you test new kernels or learning algorithm with artificial data. 2 classes are currently available, one for binary classification and the other one for multiclass classification.

In these generators, data are sampled from normal distributions centered on each class. The following is a short example of how to generate 100 samples for binary classification:

	GaussianGenerator gg = new GaussianGenerator();
	// generate 100 samples
	List<TrainingSample<double[]>> list = gg.generateList(100);

Parameters like the number of samples in each class, the standard deviation of each class and the number of dimension of the generated samples can be tweaked.