Let -path be the folder with all the data

To run:

1. python kdd_2014_data_model1.py -path
2. Run kdd_2014_model1.R (change folder <- path)
3. download ESLI data from http://nces.ed.gov/ccd/elsi/tableGenerator.aspx (Public School, Years 2011-2012, columns school id and school type)
4. Run kdd_2014_model2.R (change folder <- path) until line 126
5. python kdd_2014_data_model2.py -path
6. Run kdd_2014_model2.R from line 126 until the end
7. Final prediction is (0.5model1+0.5model2)*discount