Solves a linear system of the form (K + \lambda * I) * W = Y using Gauss-Seidel based Block Coordinate Descent as described in http://arxiv.
Solves a linear system of the form (K + \lambda * I) * W = Y using Gauss-Seidel based Block Coordinate Descent as described in http://arxiv.org/abs/1602.05310
K is assumed to be a symmetric kernel matrix generated using a kernel generator.
training labels RDD
L2 regularization parameter
number of columns per block of Gauss-Seidel solve
number of passes of co-ordinate descent to run
seed to use for permuting column blocks
frequency at which intermediate data should be checkpointed
a model that can be applied on test data.