File size: 2,529 Bytes
2568013
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
# MIT License

# Copyright (c) 2022 Karl Stelzner

# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:

# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.

# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.

# This file comes from https://github.com/stelzner/srt

import torch
from einops import rearrange
from torch import nn


class Attention(nn.Module):
    def __init__(
        self, dim, heads=8, dim_head=64, dropout=0.0, selfatt=True, kv_dim=None
    ):
        super().__init__()
        inner_dim = dim_head * heads
        project_out = not (heads == 1 and dim_head == dim)

        self.heads = heads
        self.scale = dim_head**-0.5

        self.attend = nn.Softmax(dim=-1)
        if selfatt:
            self.to_qkv = nn.Linear(dim, inner_dim * 3, bias=False)
        else:
            self.to_q = nn.Linear(dim, inner_dim, bias=False)
            self.to_kv = nn.Linear(kv_dim, inner_dim * 2, bias=False)

        self.to_out = (
            nn.Sequential(nn.Linear(inner_dim, dim), nn.Dropout(dropout))
            if project_out
            else nn.Identity()
        )

    def forward(self, x, z=None):
        if z is None:
            qkv = self.to_qkv(x).chunk(3, dim=-1)
        else:
            q = self.to_q(x)
            k, v = self.to_kv(z).chunk(2, dim=-1)
            qkv = (q, k, v)

        q, k, v = map(lambda t: rearrange(t, "b n (h d) -> b h n d", h=self.heads), qkv)

        dots = torch.matmul(q, k.transpose(-1, -2)) * self.scale

        attn = self.attend(dots)

        out = torch.matmul(attn, v)
        out = rearrange(out, "b h n d -> b n (h d)")
        return self.to_out(out)