-
Notifications
You must be signed in to change notification settings - Fork 32
/
preprocess_movie.py
105 lines (96 loc) · 4.82 KB
/
preprocess_movie.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
import json
import pandas as pd
f = open('u.data', 'r')
data = f.readlines()
f = open('u.item', 'r', encoding='ISO-8859-1')
movies = f.readlines()
f = open('u.user', 'r')
users = f.readlines()
movie_names = [_.split('|')[1] for _ in movies] # movie_names[0] = 'Toy Story (1995)'
user_ids = [_.split('|')[0] for _ in users] # user_ids[0] = '1'
movie_ids = [_.split('|')[0] for _ in movies] # movie_ids[0] = '1'
interaction_dicts = dict()
for line in data:
user_id, movie_id, rating, timestamp = line.split('\t')
if user_id not in interaction_dicts:
interaction_dicts[user_id] = {
'movie_id': [],
'rating': [],
'timestamp': [],
}
interaction_dicts[user_id]['movie_id'].append(movie_id)
interaction_dicts[user_id]['rating'].append(int(int(rating) > 3))
interaction_dicts[user_id]['timestamp'].append(timestamp)
with open('item_mapping.csv', 'w') as f:
import csv
writer = csv.writer(f)
writer.writerow(['movie_id', 'movie_name'])
for i, name in enumerate(movie_names):
writer.writerow([i + 1, name])
sequential_interaction_list = []
seq_len = 10
for user_id in interaction_dicts:
temp = zip(interaction_dicts[user_id]['movie_id'], interaction_dicts[user_id]['rating'], interaction_dicts[user_id]['timestamp'])
temp = sorted(temp, key=lambda x: x[2])
result = zip(*temp)
interaction_dicts[user_id]['movie_id'], interaction_dicts[user_id]['rating'], interaction_dicts[user_id]['timestamp'] = [list(_) for _ in result]
for i in range(10, len(interaction_dicts[user_id]['movie_id'])):
sequential_interaction_list.append(
[user_id, interaction_dicts[user_id]['movie_id'][i-seq_len:i], interaction_dicts[user_id]['rating'][i-seq_len:i], interaction_dicts[user_id]['movie_id'][i], interaction_dicts[user_id]['rating'][i], interaction_dicts[user_id]['timestamp'][i].strip('\n')]
)
sequential_interaction_list = sequential_interaction_list[-10000:] # 10000 records
import csv
# save the csv file for baselines
with open('./data/train.csv', 'w') as f:
writer = csv.writer(f)
writer.writerow(['user_id', 'history_movie_id', 'history_rating', 'movie_id', 'rating', 'timestamp'])
writer.writerows(sequential_interaction_list[:int(len(sequential_interaction_list)*0.8)])
with open('./data/valid.csv', 'w') as f:
writer = csv.writer(f)
writer.writerow(['user_id', 'history_movie_id', 'history_rating', 'movie_id', 'rating', 'timestamp'])
writer.writerows(sequential_interaction_list[int(len(sequential_interaction_list)*0.8):int(len(sequential_interaction_list)*0.9)])
with open('./data/test.csv', 'w') as f:
writer = csv.writer(f)
writer.writerow(['user_id', 'history_movie_id', 'history_rating', 'movie_id', 'rating', 'timestamp'])
writer.writerows(sequential_interaction_list[int(len(sequential_interaction_list)*0.9):])
def csv_to_json(input_path, output_path):
data = pd.read_csv(input_path)
json_list = []
for index, row in data.iterrows():
row['history_movie_id'] = eval(row['history_movie_id'])
row['history_rating'] = eval(row['history_rating'])
L = len(row['history_movie_id'])
preference = []
unpreference = []
for i in range(L):
if int(row['history_rating'][i]) == 1:
preference.append(movie_names[int(row['history_movie_id'][i]) - 1])
else:
unpreference.append(movie_names[int(row['history_movie_id'][i]) - 1])
target_movie = movie_names[int(row['movie_id']) - 1]
preference_str = ""
unpreference_str = ""
for i in range(len(preference)):
if i == 0:
preference_str += "\"" + preference[i] + "\""
else:
preference_str += ", \"" + preference[i] + "\""
for i in range(len(unpreference)):
if i == 0:
unpreference_str += "\"" + unpreference[i] + "\""
else:
unpreference_str += ", \"" + unpreference[i] + "\""
target_preference = int(row['rating'])
target_movie_str = "\"" + target_movie + "\""
target_preference_str = "Yes." if target_preference == 1 else "No."
json_list.append({
"instruction": "Given the user's preference and unpreference, identify whether the user will like the target movie by answering \"Yes.\" or \"No.\".",
"input": f"User Preference: {preference_str}\nUser Unpreference: {unpreference_str}\nWhether the user will like the target movie {target_movie_str}?",
"output": target_preference_str,
})
with open(output_path, 'w') as f:
json.dump(json_list, f, indent=4)
# generate the json file for the TALLRec
csv_to_json('./data/train.csv', './data/train.json')
csv_to_json('./data/valid.csv', './data/valid.json')
csv_to_json('./data/test.csv', './data/test.json')