SMO (Sequential Minimal Optimization)

Reference: http://cs229.stanford.edu/materials/smo.pdf
*This demo is the implementation of the Algorithm in above-mentioned reference.
SMO:
If we want to allow a variable threshold the updates must be made on a pair of data points, an approach that results in the SMO algorithm. The rate of convergence of the algorithm is strongly affected by the order in which the data points are chosen for updating. Heuristic measures such as the degree of violation of the KKT conditions can be used to ensure very effective convergence rates in practice.

Refer to: Platt, John. Fast Training of Support Vector Machines using Sequential Minimal Optimization,
in Advances in Kernel Methods – Support Vector Learning, B. Scholkopf, C. Burges,
A. Smola, eds., MIT Press (1998).