File size: 2,830 Bytes
4450790
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
# Copyright (c) Meta Platforms, Inc. and affiliates.
# All rights reserved.
#
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.

# References:
#   https://github.com/facebookresearch/dino/blob/master/vision_transformer.py
#   https://github.com/rwightman/pytorch-image-models/tree/master/timm/models/vision_transformer.py

import logging

from torch import Tensor
from torch import nn
import comfy.ops
ops = comfy.ops.manual_cast
from comfy.ldm.modules.attention import optimized_attention

logger = logging.getLogger("dinov2")

try:
    from xformers.ops import memory_efficient_attention, unbind

    XFORMERS_AVAILABLE = True
except ImportError:
    logger.warning("xFormers not available")
    XFORMERS_AVAILABLE = False


class Attention(nn.Module):
    def __init__(
        self,
        dim: int,
        num_heads: int = 8,
        qkv_bias: bool = False,
        proj_bias: bool = True,
        attn_drop: float = 0.0,
        proj_drop: float = 0.0,
    ) -> None:
        super().__init__()
        self.num_heads = num_heads
        self.head_dim = dim // num_heads
        self.scale = self.head_dim**-0.5

        self.qkv = ops.Linear(dim, dim * 3, bias=qkv_bias)
        self.attn_drop = nn.Dropout(attn_drop)
        self.proj = ops.Linear(dim, dim, bias=proj_bias)
        self.proj_drop = nn.Dropout(proj_drop)
        

    def forward(self, x: Tensor) -> Tensor:
        # B, N, C = x.shape
        # qkv = self.qkv(x).reshape(B, N, 3, self.num_heads, C // self.num_heads).permute(2, 0, 3, 1, 4)

        # q, k, v = qkv[0] * self.scale, qkv[1], qkv[2]
        # attn = q @ k.transpose(-2, -1)

        # attn = attn.softmax(dim=-1)
        # #attn = self.attn_drop(attn)

        # x = (attn @ v).transpose(1, 2).reshape(B, N, C)
        # x = self.proj(x)
        # #x = self.proj_drop(x)
        # return x
        # print("x shape: ", x.shape)
       
        B, N, C = x.shape
        q, k, v = self.qkv(x).reshape(B, N, 3, self.num_heads, C // self.num_heads).permute(2, 0, 3, 1, 4)
        out = optimized_attention(q, k, v, self.num_heads, skip_reshape=True)

        out= self.proj(out)
        out = self.proj_drop(out)
        return out
        

class MemEffAttention(Attention):
    def forward(self, x: Tensor, attn_bias=None) -> Tensor:
        if not XFORMERS_AVAILABLE:
            assert attn_bias is None, "xFormers is required for nested tensors usage"
            return super().forward(x)

        B, N, C = x.shape
        qkv = self.qkv(x).reshape(B, N, 3, self.num_heads, C // self.num_heads)

        q, k, v = unbind(qkv, 2)

        x = memory_efficient_attention(q, k, v, attn_bias=attn_bias)
        x = x.reshape([B, N, C])

        x = self.proj(x)
        x = self.proj_drop(x)
        return x