pre-procesamiento.py

import time
import datetime


with open('file1.csv', 'w') as output:
    with open('combined_data_1.txt', 'r') as f:

        currentMovieID = 0
        for line in f:
            line = line.strip()
            if (line[len(line) - 1] == ":"):
                currentMovieID = int(line[0: len(line) - 1])
            else:
                (userID, rating, rating_date) = line.split(',')
                rating_date = int(time.mktime(datetime.datetime.strptime(rating_date, "%Y-%m-%d").timetuple()))
                output.write('%s,%s,%s,%s\n' % (currentMovieID, userID, rating_date, rating))