NB This is research code. This implementation of Gappletron is in 
no way optimized, and runs very slowly because it is written in Python. 

To run the experiments the following python packages are required:
- pandas
- numpy
- copy
- glob
- os

these R packages are required:
- library


To run all experiments from the experiments section of the paper, one
can run "RunSynthethic.py".

The code to generate the data can be found in "synsep.py".

To combine the results from all experiments one can run the files
"read_results.py".

To produce the plots from the paper one can run the files "plots.R".

To see the implementation of Gappletron see "Gappletron.py". 
The OCO algorithm we use can be found in "scalefreeMD.py" and the 
loss functions can be found in "loss.py".

The files "runclass.py" and "RunExperiment.py" contain wrappers 
for our Gappletron implementation as well as for the other classifiers 
used. These wrappers allow the user to more easily run individual 
experiments.

Our implementations of the other classifiers can be found
in "classifiers.py". 