-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathdataset.py
115 lines (90 loc) · 3.71 KB
/
dataset.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
from sklearn.model_selection import train_test_split
from torch.utils.data import DataLoader, Dataset
from utils import *
def get_reference_data(config, data_type):
raw_data = []
exec(f"raw_data = read_data(config.{data_type}_data_path")
assert raw_data != []
raw_data = np.array(raw_data)
data = torch.tensor(np.array([raw_data for _ in range(config.batch_size)]))
data = data.reshape(config.batch_size, -1, 500, 500)
data = data.to(config.device)
return data
def get_query_dataloader(config):
traffic_data_list = [os.path.join(config.DATA_PATH, f'traffic_speed/traffic_speed_{i}.pkl') for i in
range(0, 23374)]
test_data_list_1 = [os.path.join(config.DATA_PATH, f'traffic_speed/traffic_speed_{i}.pkl') for i in
range(23374, 23772)]
test_data_list_2, rest_data_list = train_test_split(traffic_data_list, train_size=0.18325, random_state=100)
valid_data_list, train_data_list, = train_test_split(rest_data_list, train_size=0.25, random_state=100)
test_data_list = test_data_list_1 + test_data_list_2
if config.task == 'train':
train_dataset = MaskDataset(read_data, train_data_list)
valid_dataset = MaskDataset(read_data, valid_data_list)
train_loader = DataLoader(train_dataset, batch_size=config.batch_size, shuffle=False, num_workers=0, drop_last=True)
valid_loader = DataLoader(valid_dataset, batch_size=config.batch_size, shuffle=False, num_workers=0, drop_last=True)
return train_loader, valid_loader
else:
test_dataset = MaskDataset(read_data, test_data_list)
test_loader = DataLoader(test_dataset, batch_size=config.batch_size, shuffle=False, num_workers=0, drop_last=True)
return test_loader
class SpeedDataset(Dataset):
def __init__(self, loader, file_list):
self.inputs = file_list
self.loader = loader
def __getitem__(self, index):
file = self.inputs[index]
x = self.loader(file)['point']
y = self.loader(file)['congestion']
z = self.loader(file)['grid']
return x, y, z
def __len__(self):
return len(self.inputs)
class SpeedClassDataset(Dataset):
def __init__(self, loader, file_list):
self.inputs = file_list
self.loader = loader
def __getitem__(self, index):
file = self.inputs[index]
x = self.loader(file)['point']
y = self.loader(file)['speed']
return x, y
def __len__(self):
return len(self.inputs)
class MaskDataset(Dataset):
def __init__(self, loader, file_list):
self.inputs = file_list
self.loader = loader
def __getitem__(self, index):
file = self.inputs[index]
point = self.loader(file)['point']
x = generate_mask(point)
y = self.loader(file)['congestion']
z = self.loader(file)['grid']
return x, y, z, point
def __len__(self):
return len(self.inputs)
class OnehotDataset(Dataset):
def __init__(self, loader, file_list):
self.inputs = file_list
self.loader = loader
def __getitem__(self, index):
file = self.inputs[index]
point = self.loader(file)['point']
x = generate_onehot_mask(point)
y = self.loader(file)['congestion']
z = self.loader(file)['grid']
return x, y, z, point
def __len__(self):
return len(self.inputs)
class BertDataset(Dataset):
def __init__(self, loader, file_list):
self.inputs = file_list
self.loader = loader
def __getitem__(self, index):
file = self.inputs[index]
x = self.loader(file)['data']
y = self.loader(file)['data']
return x, y
def __len__(self):
return len(self.inputs)