-
Notifications
You must be signed in to change notification settings - Fork 0
/
data.py
49 lines (34 loc) · 1.21 KB
/
data.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
import itertools
import random
import torch as th
from torch.utils.data.dataset import Dataset
from torch.utils.data.dataloader import DataLoader
def mod_div(a, b, m):
for c in range(m):
if (b * c) % m == a:
return c
return -1
def make_data(base):
return [[a, b, mod_div(a, b, base)]
for a, b in itertools.product(range(base), range(1, base))]
class SimpleDataset(Dataset):
def __init__(self, data):
super().__init__()
self.data = data
def __len__(self):
return len(self.data)
def __getitem__(self, item):
return self.data[item]
def make_dataset(base, train_frac=0.8, batch_size=32):
data = make_data(base)
random.shuffle(data)
split = int(train_frac*len(data))
data = th.tensor(data)
op = th.full((data.shape[0],), base)
eq = th.full((data.shape[0],), base+1)
data = th.vstack([data[:, 0], op, data[:, 1], eq, data[:, 2]]).T
train_data = SimpleDataset(data[:split])
test_data = SimpleDataset(data[split:])
train_loader = DataLoader(train_data, batch_size=batch_size, shuffle=True)
test_loader = DataLoader(test_data, batch_size=batch_size, shuffle=True)
return train_loader, test_loader