-
Notifications
You must be signed in to change notification settings - Fork 0
/
my_net.py
60 lines (46 loc) · 1.36 KB
/
my_net.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
import torch
from torch import nn
import torch.nn.functional as F
import torch.nn.init as init
def kaiming_init(m):
if isinstance(m, (nn.Linear, nn.Conv2d)):
init.kaiming_normal(m.weight)
if m.bias is not None:
m.bias.data.fill_(0)
elif isinstance(m, (nn.BatchNorm1d, nn.BatchNorm2d)):
m.weight.data.fill_(1)
if m.bias is not None:
m.bias.data.fill_(0)
class nlp_nn(nn.Module):
def __init__(self, embedding_dim, L1, L2, L3, L4, L5):
super(nlp_nn, self).__init__()
self.Flatten = nn.Flatten()
self.Linear1 = nn.Linear(embedding_dim, L1)
self.Linear2 = nn.Linear(L1, L2)
self.Linear3 = nn.Linear(L2, L3)
self.Linear4 = nn.Linear(L3, L4)
self.Linear5 = nn.Linear(L4, L5)
self.LinearFinal = nn.Linear(L1, 2)
self.Drop0 = nn.Dropout(0.4)
self.Drop1 = nn.Dropout(0.4)
self.Drop2 = nn.Dropout(0.4)
self.Drop3 = nn.Dropout(0.4)
self.Drop4 = nn.Dropout(0.4)
self.Drop5 = nn.Dropout(0.4)
def forward(self, x):
act = nn.ReLU()
soft = nn.Softmax(dim=0)
x = self.Flatten(x)
x = self.Drop0(x)
x = act(self.Linear1(x))
x = self.Drop1(x)
#x = act(self.Linear2(x))
#x = self.Drop2(x)
#x = act(self.Linear3(x))
#x = self.Drop3(x)
#x = act(self.Linear4(x))
#x = self.Drop4(x)
#x = act(self.Linear5(x))
#x = self.Drop5(x)
x = self.LinearFinal(x)
return x