-
Notifications
You must be signed in to change notification settings - Fork 4
/
Copy pathDataset_CDAE.py
160 lines (127 loc) · 5.98 KB
/
Dataset_CDAE.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
import numpy as np
import os
import pickle
import random
from tqdm import tqdm
from sklearn.utils import shuffle
class Dataset(object):
def __init__(self, BATCH_SIZE, dataset='douban'):
self.batch_size = BATCH_SIZE
self.peo2item_movie, self.peo2item_book, self.item2peo_movie, self.item2peo_book= self.load_data(dataset)
self.movie_set = self.item2peo_movie.keys()
self.book_set = self.item2peo_book.keys()
self.num_user = len(self.peo2item_movie)
self.num_movie = len(self.item2peo_movie)
self.num_book = len(self.item2peo_book)
self.movie_vali, self.movie_test, self.movie_nega, self.book_vali, self.book_test, self.book_nega = self.get_test_data(dataset)
print('load data down')
self.test_count = 0
self.batch_count = 0
self.count = 0
self.epoch = 1
def load_data(self, dataset):
print('loading data...{}')
peo2movie = pickle.load(open(os.path.join('processed_data', dataset,'peo2movie_id.pkl'), 'rb'))
peo2book = pickle.load(open(os.path.join('processed_data', dataset,'peo2book_id.pkl'), 'rb'))
movie2peo = pickle.load(open(os.path.join('processed_data', dataset,'movie2peo_id.pkl'), 'rb'))
book2peo = pickle.load(open(os.path.join('processed_data', dataset,'book2peo_id.pkl'), 'rb'))
return peo2movie, peo2book, movie2peo, book2peo
# def get_train_indices(self, domain):
# row, col = [], []
#
# if domain == 'movie':
# dict = self.peo2item_movie
# vali = self.movie_vali
# elif domain == 'book':
# dict = self.peo2item_book
# vali = self.book_vali
# else:
# return
#
# for user in dict:
# dict[user].remove(vali[user])
# for each in dict[user]:
# row.append(user)
# col.append(each)
#
# row = np.array(row)
# col = np.array(col)
#
# return row, col
def get_part_train_indices(self, domain, percent):
row, col = [], []
if domain == 'movie':
dict = self.peo2item_movie
vali = self.movie_vali
test = self.movie_test
elif domain == 'book':
dict = self.peo2item_book
vali = self.book_vali
test = self.book_test
else:
return
for user in dict:
if len(dict[user])>2:
dict[user].remove(vali[user])
dict[user].remove(test[user])
else:
dict[user].remove(vali[user])
dict[user] = shuffle(dict[user], random_state=72)
num_item = int(round(percent * len(dict[user])))
for i in range(num_item):
row.append(user)
col.append(dict[user][i])
row = np.array(row)
col = np.array(col)
return row, col
def get_test_data(self, dataset):
if not os.path.exists(os.path.join(os.getcwd(), 'processed_data', dataset, 'movie_vali.pkl')) or \
not os.path.exists(os.path.join(os.getcwd(), 'processed_data', dataset, 'movie_test.pkl')):
movie_vali = {}
movie_test = {}
book_vali = {}
book_test = {}
movie_nega = {}
book_nega = {}
for i in tqdm(self.peo2item_movie):
items = self.peo2item_movie[i]
if len(items)>=2:
items_choices = shuffle(items, random_state=2020)[:2]
movie_vali[i] = items_choices[0]
movie_test[i] = items_choices[1]
else:
movie_vali[i] = items[0]
movie_test[i] = items[0]
all_nega_movies = list((set(list(range(self.num_movie))) - set(items)))
movie_nega[i] = shuffle(all_nega_movies, random_state=2020)[:99]
for i in tqdm(self.peo2item_book):
items = self.peo2item_book[i]
if len(items)>=2:
items_choices = shuffle(items, random_state=2020)[:2]
book_vali[i] = items_choices[0]
book_test[i] = items_choices[1]
else:
book_vali[i] = items[0]
book_test[i] = items[0]
all_nega_books = list((set(list(range(self.num_book))) - set(items)))
book_nega[i] = shuffle(all_nega_books, random_state=2020)[:99]
pickle.dump(movie_vali, open(os.path.join('processed_data', dataset, 'movie_vali.pkl'), 'wb'))
pickle.dump(book_vali, open(os.path.join('processed_data', dataset, 'book_vali.pkl'), 'wb'))
pickle.dump(movie_test, open(os.path.join('processed_data', dataset, 'movie_test.pkl'), 'wb'))
pickle.dump(book_test, open(os.path.join('processed_data', dataset, 'book_test.pkl'), 'wb'))
pickle.dump(movie_nega, open(os.path.join('processed_data', dataset, 'movie_nega.pkl'), 'wb'))
pickle.dump(book_nega, open(os.path.join('processed_data', dataset, 'book_nega.pkl'), 'wb'))
else:
movie_vali = pickle.load(open(os.path.join('processed_data', dataset, 'movie_vali.pkl'), 'rb'))
book_vali = pickle.load(open(os.path.join('processed_data', dataset, 'book_vali.pkl'), 'rb'))
movie_test = pickle.load(open(os.path.join('processed_data', dataset, 'movie_test.pkl'), 'rb'))
book_test = pickle.load(open(os.path.join('processed_data', dataset, 'book_test.pkl'), 'rb'))
movie_nega = pickle.load(open(os.path.join('processed_data', dataset, 'movie_nega.pkl'), 'rb'))
book_nega = pickle.load(open(os.path.join('processed_data', dataset, 'book_nega.pkl'), 'rb'))
return movie_vali, movie_test, movie_nega, book_vali, book_test, book_nega
if __name__ == '__main__':
dataset = Dataset(1000, 4, True)
i = 1
vali = dataset.movie_vali[i]
nega = dataset.movie_nega[i]
nega_vali = np.concatenate([nega, vali],1)