-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathdataset_load.py
157 lines (151 loc) · 7.72 KB
/
dataset_load.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
import glob
import os
import time
import numpy as np
import pandas as pd
import pickle
import random
def get_data_partition(partition_file):
vid2partition, partition2vid = {}, {}
df = pd.read_csv(partition_file)
for row in df.values:
vid, partition = str("%03d"%row[0]), row[1]
vid2partition[vid] = partition
if partition not in partition2vid:
partition2vid[partition] = []
if vid not in partition2vid[partition]:
partition2vid[partition].append(vid)
return vid2partition, partition2vid
import re
def extract_number(filename):
# 使用正则表达式匹配中间的数字
match = re.search(r'-(\d+)-', filename)
if match:
return int(match.group(1)) # 返回匹配的数字
else:
return 0 # 如果没有匹配到数字，则返回 0
def load_data(args):
feature_path = os.path.join(args.dataset_file_path,"features",args.feature_set)
label_path = os.path.join(args.dataset_file_path,"labels")
data_file_name = f'data_{args.feature_set}_{args.fea_dim}.pkl'
data_file = os.path.join(f'./data_cache/{args.task}/', data_file_name)
if args.use_personality ==True and args.use_emotion==False:
data_file = os.path.join(f'./data_cache/personality_{args.task}/', data_file_name)
elif args.use_emotion==True and args.use_personality ==False:
data_file = os.path.join(f'./data_cache/emotion/', data_file_name)
elif args.use_emotion==True and args.use_personality ==True:
data_file = os.path.join(f'./data_cache/emotion_personality/', data_file_name)
else:
data_file = os.path.join(f'./data_cache/{args.task}/', data_file_name)
if os.path.exists(data_file) and args.use_emotion==True: # check if file of preprocessed data exists
print(f'Find cached data "{os.path.basename(data_file)}".')
with open(data_file, 'rb') as f:
data = pickle.load(f)
return data
elif os.path.exists(data_file) and args.use_emotion==False:
print(f'Find cached data "{os.path.basename(data_file)}".')
with open(data_file, 'rb') as f:
data = pickle.load(f)
return data
print('Constructing data from scratch ...')
data = {'train': {'feature': [], 'label': []},
'val': {'feature': [], 'label': []},
'test': {'feature': [], 'label': []}}
if args.use_personality==True and args.use_emotion==False:
data['train']['personality']=[]
data['val']['personality']=[]
data['test']['personality']=[]
personality = pd.read_csv(os.path.join(args.dataset_file_path,"personality.csv"))
if args.use_emotion==True and args.use_personality==False:
data['train']['emotion']=[]
data['val']['emotion']=[]
data['test']['emotion']=[]
emotion = pd.read_csv(os.path.join(args.dataset_file_path,"emotion.csv"), header=None)
if args.use_emotion==True and args.use_personality==True:
data['train']['personality']=[]
data['val']['personality']=[]
data['test']['personality']=[]
personality = pd.read_csv(os.path.join(args.dataset_file_path,"personality.csv"))
data['train']['emotion']=[]
data['val']['emotion']=[]
data['test']['emotion']=[]
emotion = pd.read_csv(os.path.join(args.dataset_file_path,"emotion.csv"), header=None)
for vid in sorted(os.listdir(feature_path)):
lie = 2
nolie = 3
label_file = os.path.join(label_path, vid + '.csv')
label = (pd.read_csv(label_file).iloc[:, 3]).to_numpy()
files = os.listdir(os.path.join(feature_path,vid))
files = sorted(files, key=extract_number)
random.seed(int(vid))
random.shuffle(files) #乱序遍历文件夹
for file in files:
number = int(file.split('-')[1])
if file.endswith("csv"):
feature = pd.read_csv(os.path.join(feature_path, vid, file), header=None).to_numpy()
elif file.endswith("npy"):
feature = np.load(os.path.join(feature_path, vid, file))
if label[number-1]>0 and lie>0:
lie = lie - 1
data["val"]['label'].append(label[number - 1])
data["val"]['feature'].append(feature)
if args.use_personality==True and args.use_emotion==False:
personality_data = personality.iloc[:,1:61].to_numpy()[int(vid)-1]
data["val"]['personality'].append(personality_data)
if args.use_emotion==True and args.use_personality==False:
emotion_data = emotion.to_numpy()[int(vid)-1]
data["val"]['emotion'].append(emotion_data)
if args.use_emotion==True and args.use_personality==True:
personality_data = personality.iloc[:,1:61].to_numpy()[int(vid)-1]
data["val"]['personality'].append(personality_data)
emotion_data = emotion.to_numpy()[int(vid)-1]
data["val"]['emotion'].append(emotion_data)
elif label[number-1]==0 and nolie>0:
nolie = nolie - 1
data["val"]['label'].append(label[number - 1])
data["val"]['feature'].append(feature)
if args.use_personality==True and args.use_emotion==False:
personality_data = personality.iloc[:,1:61].to_numpy()[int(vid)-1]
data["val"]['personality'].append(personality_data)
if args.use_emotion==True and args.use_personality==False:
emotion_data = emotion.to_numpy()[int(vid)-1]
data["val"]['emotion'].append(emotion_data)
if args.use_emotion==True and args.use_personality==True:
personality_data = personality.iloc[:,1:61].to_numpy()[int(vid)-1]
data["val"]['personality'].append(personality_data)
emotion_data = emotion.to_numpy()[int(vid)-1]
data["val"]['emotion'].append(emotion_data)
else:
data["train"]['label'].append(label[number - 1])
data["train"]['feature'].append(feature)
if args.use_personality==True and args.use_emotion==False:
personality_data = personality.iloc[:,1:61].to_numpy()[int(vid)-1]
data["train"]['personality'].append(personality_data)
if args.use_emotion==True and args.use_personality==False:
emotion_data = emotion.to_numpy()[int(vid)-1]
data["train"]['emotion'].append(emotion_data)
if args.use_emotion==True and args.use_personality==True:
personality_data = personality.iloc[:,1:61].to_numpy()[int(vid)-1]
data["train"]['personality'].append(personality_data)
emotion_data = emotion.to_numpy()[int(vid)-1]
data["train"]['emotion'].append(emotion_data)
print(vid)
print(len(data["train"]['label']))
print(len(data["val"]['label']))
print("数据加载完成")
if args.task=="deception":
if args.use_personality==True:
if not os.path.exists("./data_cache/personality_deception"):
os.mkdir("./data_cache/personality_deception")
else:
if not os.path.exists("./data_cache/deception"):
os.mkdir("./data_cache/deception")
elif args.task=="emotion":
if args.use_personality==True:
if not os.path.exists("./data_cache/personality_emotion"):
os.mkdir("./data_cache/personality_emotion")
else:
if not os.path.exists("./data_cache/emotion"):
os.mkdir("./data_cache/emotion")
pickle.dump(data, open(data_file, 'wb'))
return data