There are 5 python files in this folder:
1) data.py simply loads the LIBSVM a9a dataset
2) function.py defines a stochastic oracle and
a hessian vector product oracle for Logistic
Regression
3) optimizer.py implements the different 
optimization algorithms
4) main.py can generate losses for FedSN for the
provided set of parameters. These get stored inside
the runtime folder.
7) running 'plot.py' stores
the generated figure 6a inside the figures folder
It can also be used to generate figure 6b. To get
the appropriate losses for GIANT use my version of
the GIANT code.