row_count1 = sum(1 for row in open(fn1))
    print(row_count1)
    row_count2 = sum(1 for row in open(fn2))
    sample = (int)(0.4 * row_count1)
    sample2 = (int)(0.4 * row_count2)
    skip = sorted(random.sample(range(row_count1), row_count1 - sample))
    skip2 = sorted(random.sample(range(row_count2), row_count2 - sample2))
    print(pandas.read_csv(fn1, nrows=1).columns)
    data1 = pandas.read_csv(fn1,header=0, skiprows=skip, names=pandas.read_csv(fn1, nrows=1).columns)
    print(data1)
    data2 = pandas.read_csv(fn2,header=0, skiprows=skip2, names=pandas.read_csv(fn2, nrows=1).columns)
    print(data2)