MSD sparse matrix read

MSD sparse matrix read.py

#!/usr/bin/python
#opens the triplets, saves song & user keys, writes out a simpler matrix based on int user, int song, int rating.


dataset_file = open('train_triplets.txt', 'r') #http://labrosa.ee.columbia.edu/millionsong/sites/default/files/challenge/train_triplets.txt.zip
songs_count = 0 
users_count = 0
dataset_count = 0
users_dict = dict()
songs_dict = dict()	
file_out = open('results_int.txt', 'w')
file_songs = open('songs.txt', 'w')
file_users = open('users.txt', 'w')
while 1:
	lines = dataset_file.readlines(1000000)
	if not lines:
		break
	for line in lines:
		line = line.strip()
		(user, song, freq) = line.split()
		
		user = user.strip()
		if (users_dict.has_key(user) == False):
			users_dict[user] = users_count
			file_users.write(user)
			file_users.write('\n')
			users_count = users_count + 1
			
		song = song.strip()
		if (songs_dict.has_key(song) == False):
			songs_dict[song] = songs_count
			file_songs.write(song)
			file_songs.write('\n')
			songs_count = songs_count + 1
			
		freq = freq.strip()
		out_line = [str(users_dict[user]), " ", str(songs_dict[song]), " ", str(freq), '\n']
		file_out.writelines(out_line)
		
dataset_file.close()
file_songs.close()
file_users.close()
file_out.close()