Implementation of Kernel non-negative matrix factorization with multiplicative updates in PySpark. The multiplication with Kernel matrices is done via distributed block matrix multiplication operation.
For easy local instalation of Spark see: https://github.com/paulovn/ml-vm-notebook