Adds options, removes static data

[ddrager]

The biggest changes I made here:

1. Removes data directories, since this is data imported from Kaggle
2. Adds mandatory option -d to specify which data directory to use.
3. Adds -y option for year, defaults to current year
4. Updates deprecated sklearn function.

Thank you for this! New to ML, this was a neat learning experiment.

[harvitronix]

This is great, @ddrager. Thanks for the PR.

For the instructions and usage, I think you should add the -y ### as a suggestion so people know that's an arg without having to look at the code.

Cheers.