-
Notifications
You must be signed in to change notification settings - Fork 0
/
decoder.py
159 lines (124 loc) · 5.02 KB
/
decoder.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
158
159
import torch
import torch.nn as nn
import torch.nn.functional as F
from modules import MultiHeadAttentionLayer, FeedforwardLayer
class TransformerDecoderLayer(nn.Module):
def __init__(self,
hid_dim,
n_heads,
pf_dim,
dropout,
device):
super(TransformerDecoderLayer, self).__init__()
self.self_attn_layer_norm = nn.LayerNorm(hid_dim)
self.enc_attn_layer_norm = nn.LayerNorm(hid_dim)
self.ff_layer_norm = nn.LayerNorm(hid_dim)
self.self_attention = MultiHeadAttentionLayer(
hid_dim, n_heads, dropout, device)
self.encoder_attention = MultiHeadAttentionLayer(
hid_dim, n_heads, dropout, device)
self.positionwise_feedforward = PositionwiseFeedforwardLayer(
hid_dim, pf_dim, dropout)
self.dropout = nn.Dropout(dropout)
def forward(self,
trg,
enc_src,
trg_mask,
src_mask):
# trg = [batch size, trg len, hid dim]
# enc_src = [batch size, src len, hid dim]
# trg_mask = [batch size, trg len]
# src_mask = [batch size, src len]
# self attention
_trg, _ = self.self_attention(trg, trg, trg, trg_mask)
# dropout, residual connection and layer norm
trg = self.self_attn_layer_norm(trg + self.dropout(_trg))
# trg = [batch size, trg len, hid dim]
# encoder attention
_trg, attention = self.encoder_attention(
trg, enc_src, enc_src, src_mask)
# dropout, residual connection and layer norm
trg = self.enc_attn_layer_norm(trg + self.dropout(_trg))
# trg = [batch size, trg len, hid dim]
# positionwise feedforward
_trg = self.positionwise_feedforward(trg)
# dropout, residual and layer norm
trg = self.ff_layer_norm(trg + self.dropout(_trg))
# trg = [batch size, trg len, hid dim]
# attention = [batch size, n heads, trg len, src len]
return trg, attention
class seesDecoder(nn.Module):
def __init__(self,
hid_dim,
out_dim,
seq_len,
device,
dropout=0.2,
activation=F.relu):
super(seesDecoder, self).__init__()
self.device = device
self.hid_dim = hid_dim
self.dropout = nn.Dropout(p=dropout)
self.activation = activation
self.FC = nn.Linear(hid_dim * seq_len, out_dim, bias=True)
def forward(self, src):
# src, (batch_size, seq_len, hid_dim)
src = src.view(src.shape[0], src.shape[1]*src.shape[2])
# src, (batch_size, seq_len * hid_dim)
src = self.FC(src)
# src, (batch_size, out_dim)
return src
class seesDecoderWithAttInputs(nn.Module):
def __init__(self,
hid_dim,
out_dim,
seq_len,
device,
dropout=0.2,
activation=F.relu):
super(seesDecoderWithAttInputs, self).__init__()
self.device = device
self.hid_dim = hid_dim
self.dropout = nn.Dropout(p=dropout)
self.activation = activation
self.FC1 = nn.Linear(hid_dim * 2, hid_dim, bias=True)
self.FC2 = nn.Linear(hid_dim * seq_len, out_dim, bias=True)
def forward(self, src):
# src, (batch_size, seq_len, hid_dim * 2)
src = self.FC1(src)
# src, (batch_size, seq_len, hid_dim)
src = src.view(src.shape[0], src.shape[1]*src.shape[2])
# src, (batch_size, seq_len * hid_dim)
src = self.FC2(src)
# src, (batch_size, out_dim)
return src
class seesDecoderWithMultiOut(nn.Module):
def __init__(self,
hid_dim,
out_dim,
seq_len,
device,
dropout=0.2,
activation=F.relu):
super(seesDecoderWithMultiOut, self).__init__()
self.device = device
self.hid_dim = hid_dim
self.seq_len = seq_len
self.dropout = nn.Dropout(p=dropout)
self.activation = activation
self.FC = nn.Linear(hid_dim * 2, hid_dim, bias=True)
self.futureFC = nn.ModuleList(nn.Linear(hid_dim * seq_len,
out_dim,
bias=True) for _ in range(seq_len))
def forward(self, src):
# src, (batch_size, seq_len, hid_dim * 2)
src = self.FC(src)
# src, (batch_size, seq_len, hid_dim)
src = src.view(src.shape[0], src.shape[1]*src.shape[2])
# src, (batch_size, seq_len * hid_dim)
out = []
for layer in self.futureFC:
out.append(layer(src))
out = torch.cat(out, dim=1)
# out, (batch_size, seq_len)
return out