Cut down on the amount of data in data set.

simplify_statistics.py

import csv
import codecs

fin = 'quick-statistics.csv'
threshold = 40

lines_to_keep = []
with open(fin, 'r') as f:
    keep_line = 0
    reader = csv.reader(f)
    lines_to_keep.append(reader.next())
    for line in reader:
        if keep_line == threshold:
            lines_to_keep.append(line)
            keep_line = 0
        else:
            keep_line +=1

fout = '{}-out.csv'.format(fin.split('.')[0])
with codecs.open(fout, mode='w+', encoding='utf-8') as f:
    for line in lines_to_keep:
        l = ','.join(line) + '\n'
        f.write(l)

print "Complete"