I am not sure everything should go into master, hence the different commits.
What has been done:
Various fixes in the notebooks (348434d ). Most important is that the smoothing is fixed. It was totally wrong before.
update the preprocessing script (d4c93f6 ). The smoothing is also done as a preprocessing step. This makes the code slightly faster. We do not have to do that for the final version. Coming back should be easy.
Adding two scripts that perform the full batch of experiments (347ff42 )
Add a notebook to produce all the plots for the paper (d3e322e )
Add a notebook to produce the solution for the SVM (3cc75f1 )
I think at least 348434d and d4c93f6 should go to master.
I am not sure everything should go into master, hence the different commits.
What has been done:
I think at least 348434d and d4c93f6 should go to master.