File size: 6,267 Bytes
226675b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
import torch
import torch.nn.functional as F
from einops import rearrange
from torch import nn

class TwoLayerConv2d(nn.Sequential):
    def __init__(self, in_channels, out_channels, kernel_size=3):
        super().__init__(nn.Conv2d(in_channels, in_channels, kernel_size=kernel_size,
                            padding=kernel_size // 2, stride=1, bias=False),
                         nn.BatchNorm2d(in_channels),
                         nn.ReLU(),
                         nn.Conv2d(in_channels, out_channels, kernel_size=kernel_size,
                            padding=kernel_size // 2, stride=1)
                         )

class Transformer(nn.Module):
    def __init__(self, dim, depth, heads, dim_head, mlp_dim, dropout):
        super().__init__()
        self.layers = nn.ModuleList([])
        for _ in range(depth):
            self.layers.append(nn.ModuleList([
                Residual(PreNorm(dim, Attention(dim, heads = heads, dim_head = dim_head, dropout = dropout))),
                Residual(PreNorm(dim, FeedForward(dim, mlp_dim, dropout = dropout)))
            ]))
    def forward(self, x, mask = None):
        for attn, ff in self.layers:
            x = attn(x, mask = mask)
            x = ff(x)
        return x

class TransformerDecoder(nn.Module):
    def __init__(self, dim, depth, heads, dim_head, mlp_dim, dropout, softmax=True):
        super().__init__()
        self.layers = nn.ModuleList([])
        for _ in range(depth):
            self.layers.append(nn.ModuleList([
                Residual2(PreNorm2(dim, Cross_Attention(dim, heads = heads,
                                                        dim_head = dim_head, dropout = dropout,
                                                        softmax=softmax))),
                Residual(PreNorm(dim, FeedForward(dim, mlp_dim, dropout = dropout)))
            ]))
    def forward(self, x, m, mask = None):
        """target(query), memory"""
        for attn, ff in self.layers:
            x = attn(x, m, mask = mask)
            x = ff(x)
        return x

class PreNorm(nn.Module):
    def __init__(self, dim, fn):
        super().__init__()
        self.norm = nn.LayerNorm(dim)
        self.fn = fn
    def forward(self, x, **kwargs):
        return self.fn(self.norm(x), **kwargs)

class PreNorm2(nn.Module):
    def __init__(self, dim, fn):
        super().__init__()
        self.norm = nn.LayerNorm(dim)
        self.fn = fn
    def forward(self, x, x2, **kwargs):
        return self.fn(self.norm(x), self.norm(x2), **kwargs)

class Residual(nn.Module):
    def __init__(self, fn):
        super().__init__()
        self.fn = fn
    def forward(self, x, **kwargs):
        return self.fn(x, **kwargs) + x
    
class Residual2(nn.Module):
    def __init__(self, fn):
        super().__init__()
        self.fn = fn
    def forward(self, x, x2, **kwargs):
        return self.fn(x, x2, **kwargs) + x
    
class FeedForward(nn.Module):
    def __init__(self, dim, hidden_dim, dropout = 0.):
        super().__init__()
        self.net = nn.Sequential(
            nn.Linear(dim, hidden_dim),
            nn.GELU(),
            nn.Dropout(dropout),
            nn.Linear(hidden_dim, dim),
            nn.Dropout(dropout)
        )
    def forward(self, x):
        return self.net(x)

class Attention(nn.Module):
    def __init__(self, dim, heads = 8, dim_head = 64, dropout = 0.):
        super().__init__()
        inner_dim = dim_head *  heads
        self.heads = heads
        self.scale = dim ** -0.5

        self.to_qkv = nn.Linear(dim, inner_dim * 3, bias = False)
        self.to_out = nn.Sequential(
            nn.Linear(inner_dim, dim),
            nn.Dropout(dropout)
        )

    def forward(self, x, mask = None):
        b, n, _, h = *x.shape, self.heads
        qkv = self.to_qkv(x).chunk(3, dim = -1)
        q, k, v = map(lambda t: rearrange(t, 'b n (h d) -> b h n d', h = h), qkv)

        dots = torch.einsum('bhid,bhjd->bhij', q, k) * self.scale
        mask_value = -torch.finfo(dots.dtype).max

        if mask is not None:
            mask = F.pad(mask.flatten(1), (1, 0), value = True)
            assert mask.shape[-1] == dots.shape[-1], 'mask has incorrect dimensions'
            mask = mask[:, None, :] * mask[:, :, None]
            dots.masked_fill_(~mask, mask_value)
            del mask

        attn = dots.softmax(dim=-1)


        out = torch.einsum('bhij,bhjd->bhid', attn, v)
        out = rearrange(out, 'b h n d -> b n (h d)')
        out = self.to_out(out)
        return out

class Cross_Attention(nn.Module):
    def __init__(self, dim, heads = 8, dim_head = 64, dropout = 0., softmax=True):
        super().__init__()
        inner_dim = dim_head * heads
        self.heads = heads
        self.scale = dim ** -0.5

        self.softmax = softmax
        self.to_q = nn.Linear(dim, inner_dim, bias=False)
        self.to_k = nn.Linear(dim, inner_dim, bias=False)
        self.to_v = nn.Linear(dim, inner_dim, bias=False)

        self.to_out = nn.Sequential(
            nn.Linear(inner_dim, dim),
            nn.Dropout(dropout)
        )

    def forward(self, x, m, mask = None):

        b, n, _, h = *x.shape, self.heads
        q = self.to_q(x)
        k = self.to_k(m)
        v = self.to_v(m)

        q, k, v = map(lambda t: rearrange(t, 'b n (h d) -> b h n d', h = h), [q,k,v])

        dots = torch.einsum('bhid,bhjd->bhij', q, k) * self.scale
        mask_value = -torch.finfo(dots.dtype).max

        if mask is not None:
            mask = F.pad(mask.flatten(1), (1, 0), value = True)
            assert mask.shape[-1] == dots.shape[-1], 'mask has incorrect dimensions'
            mask = mask[:, None, :] * mask[:, :, None]
            dots.masked_fill_(~mask, mask_value)
            del mask

        if self.softmax:
            attn = dots.softmax(dim=-1)
        else:
            attn = dots
        # attn = dots
        # vis_tmp(dots)

        out = torch.einsum('bhij,bhjd->bhid', attn, v)
        out = rearrange(out, 'b h n d -> b n (h d)')
        out = self.to_out(out)
        # vis_tmp2(out)

        return out