-
Notifications
You must be signed in to change notification settings - Fork 0
/
movie_lens.py
227 lines (156 loc) · 7.72 KB
/
movie_lens.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
import csv
import time
from trie import trie_node
from hash import hash
def ids_trie(data_path):
movie_ids = trie_node()
with open(data_path, 'r') as movies_csv:
csv_reader = csv.reader(movies_csv)
next(csv_reader, None)
for row in csv_reader:
movie_ids.add(row[1],int(row[0]))
return movie_ids
def titles_hash(data_path):
movie_titles = hash(2**15)
with open(data_path, 'r') as movies_csv:
csv_reader = csv.reader(movies_csv)
next(csv_reader, None)
for row in csv_reader:
movie_titles.add(int(row[0]), row[1])
return movie_titles
def read_ratings(movies_data, ratings_data):
movies_info = hash(2**15)
users_info = hash(2**18)
with open(movies_data, 'r') as movies_csv:
csv_reader = csv.reader(movies_csv)
next(csv_reader, None)
for row in csv_reader:
movies_info.add(int(row[0]),[row[2].split('|'),[],0])
with open(ratings_data, 'r') as ratings_csv:
csv_reader = csv.reader(ratings_csv)
next(csv_reader, None)
for row in csv_reader:
#movies_info operations
data = movies_info.get(int(row[1]))[0]
data[1].append(float(row[2]))
data[2] += 1
movies_info.update(int(row[1]), data)
#users_info operations
data = users_info.get(int(row[0]))[0]
if not data:
users_info.add(int(row[0]), [[int(row[1])], [float(row[2])]])
else:
data[0].append(row[1])
data[1].append(row[2])
users_info.update(int(row[0]), data)
for cell in movies_info.data:
if cell != None:
if cell[1][2]:
sum = 0
for rating in cell[1][1]:
sum += rating
cell[1][1] = sum / cell[1][2]
else:
cell[1][1] = 0
return movies_info, users_info
def read_tags(data_path, movie_titles):
tags = trie_node()
with open(data_path, 'r') as tags_csv:
csv_reader = csv.reader(tags_csv)
next(csv_reader, None)
for row in csv_reader:
movies = tags.get(row[2])
if movies:
movie = movie_titles.get(int(row[1]))[0]
if movie not in movies:
movies.append(movie)
tags.add(row[2],movies)
else:
movie = movie_titles.get(int(row[1]))[0]
tags.add(row[2],[movie])
return tags
def partition(ids, low, high, movies_info):
i = low - 1
pivot_id = ids[high]
for j in range(low,high):
if movies_info.get(ids[j])[0][1] >= movies_info.get(pivot_id)[0][1]:
i += 1
ids[i],ids[j] = ids[j],ids[i]
ids[i+1],ids[high] = ids[high],ids[i+1]
return i+1
def quicksort(ids, low, high, movies_info):
if low < high:
pivot_idx = partition(ids, low, high, movies_info)
quicksort(ids, low, pivot_idx-1, movies_info)
quicksort(ids, pivot_idx+1, high, movies_info)
if __name__ == '__main__':
start_time = time.time()
#INICIALIZACAO DAS ESTRUTURAS DE DADOS
#1 - TRIE
movie_ids = ids_trie('Dados/movie.csv')
movie_titles = titles_hash('Dados/movie.csv')
#2 - HASH TABLE AND 3 - ANY STRUCTURE
movies_info, users_info = read_ratings('Dados/movie.csv', 'Dados/rating.csv')
#EXTRA - TRIE PARA TAGS
tags = read_tags('Dados/tag.csv', movie_titles)
print("--- %s seconds ---" % (time.time() - start_time))
#MODO CONSOLE
sorted_ids = []
query = input('$ ')
while(query):
#movie <prefix> -----------------------------------------------------------------------------------------------
if query.split(' ')[0] == 'movie':
results = movie_ids.search(query[6:])
print("{0:<7} | {1:<40} | {2:<40} | {3:<7} | {4:<6}".format('movieid','title','genres','ratings','count'))
for result in results:
print("{0:7} | {1:40.40} | {2:40.40} | {3:7.6} | {4:6}".format(result[1],
result[0],
'|'.join(movies_info.get(result[1])[0][0]),
movies_info.get(result[1])[0][1],
movies_info.get(result[1])[0][2]))
#user <id> ----------------------------------------------------------------------------------------------------
if query.split(' ')[0] == 'user':
user_info = users_info.get(int(query.split(' ')[1]))[0]
if user_info != None:
print('{0:11} | {1:40} | {2:13} | {3:6}'.format('user_rating','title','global_rating','count'))
for i in range(len(user_info[0])):
print('{0:>11.1f} | {1:40.40} | {2:>13.6f} | {3:<5}'.format(float(user_info[1][i]),
movie_titles.get(int(user_info[0][i]))[0],
float(movies_info.get(int(user_info[0][i]))[0][1]),
movies_info.get(int(user_info[0][i]))[0][2]))
else:
print('No matching results')
#topN <genre> -----------------------------------------------------------------------------------------------
if query[:3] == 'top':
N = int(query.split(' ')[0][3:])
genre = query.split(' ')[1].replace("'",'')
if not sorted_ids:
sorted_ids = [x[0] for x in [y for y in movies_info.data if y != None] if x[1][2] >= 1000]
quicksort(sorted_ids, 0, len(sorted_ids)-1, movies_info)
print("{0:<40} | {1:<40} | {2:<7} | {3:<6}".format('title', 'genres', 'ratings', 'count'))
for id in sorted_ids:
if genre in movies_info.get(id)[0][0]:
print("{0:40.40} | {1:40.40} | {2:7.6} | {3:6}".format(movie_titles.get(id)[0],
'|'.join(movies_info.get(id)[0][0]),
movies_info.get(id)[0][1],
movies_info.get(id)[0][2]))
N -= 1
if N == 0:
break
#tags <list of tags> -----------------------------------------------------------------------------------------
if query.split(' ')[0] == 'tags':
tag_list = [x for x in query[5:].split("'") if x != '' and x != ' ']
movies = tags.get(tag_list[0])
if not movies:
movies = []
for tag in tag_list[1:]:
movies = [movie for movie in movies if movie in tags.get(tag)]
print('{0:40} | {1:<40} | {2:<7} | {3:<6}'.format('title','genres','rating','count'))
for movie in movies:
id = movie_ids.get(movie)
print('{0:40.40s} | {1:40.40s} | {2:.5f} | {3:<6}'.format(movie,
"|".join(movies_info.get(id)[0][0]),
float(movies_info.get(id)[0][1]),
movies_info.get(id)[0][2]))
#LE QUERY ----------------------------------------------------------------------------------------------------
query = input('$ ')