File size: 8,444 Bytes
e4df51f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
import torch
import torch.nn.functional as F
from torch.nn.utils.rnn import pad_sequence

try:
    from flash_attn import flash_attn_varlen_func
    FLASH_ATTN_AVALIABLE = True
except:
    FLASH_ATTN_AVALIABLE = False


def apply_rotary_emb(
    x: torch.Tensor,
    freqs_cis,
    use_real = True,
    use_real_unbind_dim = -1,
):
    """
    Apply rotary embeddings to input tensors using the given frequency tensor. This function applies rotary embeddings
    to the given query or key 'x' tensors using the provided frequency tensor 'freqs_cis'. The input tensors are
    reshaped as complex numbers, and the frequency tensor is reshaped for broadcasting compatibility. The resulting
    tensors contain rotary embeddings and are returned as real tensors.

    Args:
        x (`torch.Tensor`):
            Query or key tensor to apply rotary embeddings. [B, H, S, D] xk (torch.Tensor): Key tensor to apply
        freqs_cis (`Tuple[torch.Tensor]`): Precomputed frequency tensor for complex exponentials. ([B, S, D], [B, S, D],)

    Returns:
        Tuple[torch.Tensor, torch.Tensor]: Tuple of modified query tensor and key tensor with rotary embeddings.
    """
    if use_real:
        B, H, S, D = x.size()
        cos, sin = freqs_cis[..., 0], freqs_cis[..., 1]
        cos = cos.unsqueeze(1)  
        sin = sin.unsqueeze(1)
        cos, sin = cos.to(x.device), sin.to(x.device)

        if use_real_unbind_dim == -1:
            # Used for flux, cogvideox, hunyuan-dit
            x_real, x_imag = x.reshape(*x.shape[:-1], -1, 2).unbind(-1)  # [B, S, H, D//2]
            x_rotated = torch.stack([-x_imag, x_real], dim=-1).flatten(3)
        elif use_real_unbind_dim == -2:
            # Used for Stable Audio
            x_real, x_imag = x.reshape(*x.shape[:-1], 2, -1).unbind(-2)  # [B, S, H, D//2]
            x_rotated = torch.cat([-x_imag, x_real], dim=-1)
        else:
            raise ValueError(f"`use_real_unbind_dim={use_real_unbind_dim}` but should be -1 or -2.")

        out = (x.float() * cos + x_rotated.float() * sin).to(x.dtype)

        return out
    else:
        # used for lumina
        x_rotated = torch.view_as_complex(x.float().reshape(*x.shape[:-1], -1, 2))
        freqs_cis = freqs_cis.unsqueeze(2)
        x_out = torch.view_as_real(x_rotated * freqs_cis).flatten(3)

        return x_out.type_as(x)


class FluxAttnProcessor2_0:
    """Attention processor used typically in processing the SD3-like self-attention projections."""

    def __init__(self):
        if not hasattr(F, "scaled_dot_product_attention"):
            raise ImportError("FluxAttnProcessor2_0 requires PyTorch 2.0, to use it, please upgrade PyTorch to 2.0.")

    def __call__(
        self,
        attn,
        hidden_states,
        encoder_hidden_states=None,
        attention_mask=None,
        image_rotary_emb=None,
        lens=None,
    ) -> torch.FloatTensor:
        batch_size, _, _ = hidden_states.shape if encoder_hidden_states is None else encoder_hidden_states.shape

        # `sample` projections.
        query = attn.to_q(hidden_states)
        key = attn.to_k(hidden_states)
        value = attn.to_v(hidden_states)

        inner_dim = key.shape[-1]
        head_dim = inner_dim // attn.heads

        query = query.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
        key = key.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
        value = value.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)

        if attn.norm_q is not None:
            query = attn.norm_q(query)
        if attn.norm_k is not None:
            key = attn.norm_k(key)

        # the attention in FluxSingleTransformerBlock does not use `encoder_hidden_states`
        if encoder_hidden_states is not None:
            # `context` projections.
            encoder_hidden_states_query_proj = attn.add_q_proj(encoder_hidden_states)
            encoder_hidden_states_key_proj = attn.add_k_proj(encoder_hidden_states)
            encoder_hidden_states_value_proj = attn.add_v_proj(encoder_hidden_states)

            encoder_hidden_states_query_proj = encoder_hidden_states_query_proj.view(
                batch_size, -1, attn.heads, head_dim
            ).transpose(1, 2)
            encoder_hidden_states_key_proj = encoder_hidden_states_key_proj.view(
                batch_size, -1, attn.heads, head_dim
            ).transpose(1, 2)
            encoder_hidden_states_value_proj = encoder_hidden_states_value_proj.view(
                batch_size, -1, attn.heads, head_dim
            ).transpose(1, 2)

            if attn.norm_added_q is not None:
                encoder_hidden_states_query_proj = attn.norm_added_q(encoder_hidden_states_query_proj)
            if attn.norm_added_k is not None:
                encoder_hidden_states_key_proj = attn.norm_added_k(encoder_hidden_states_key_proj)

            # attention
            query = torch.cat([encoder_hidden_states_query_proj, query], dim=2)
            key = torch.cat([encoder_hidden_states_key_proj, key], dim=2)
            value = torch.cat([encoder_hidden_states_value_proj, value], dim=2)

        if image_rotary_emb is not None:
            query = apply_rotary_emb(query, image_rotary_emb)
            key = apply_rotary_emb(key, image_rotary_emb)

        # supporting sequence length
        q_lens = lens.clone() if lens is not None else torch.LongTensor([query.shape[2]] * batch_size).to(query.device)
        k_lens = lens.clone() if lens is not None else torch.LongTensor([key.shape[2]] * batch_size).to(key.device)
        
        # hacked: shared attention   
        txt_len = 512
        context_key = [
            torch.cat([key[0], key[1, :, txt_len:]], dim=1).permute(1, 0, 2),
            key[1].permute(1, 0, 2)
        ]
        context_value = [
            torch.cat([value[0], value[1, :, txt_len:]], dim=1).permute(1, 0, 2),
            value[1].permute(1, 0, 2)
        ]
        k_lens = torch.LongTensor([k.size(0) for k in context_key]).to(query.device)
        key = pad_sequence(context_key, batch_first=True).permute(0, 2, 1, 3)
        value = pad_sequence(context_value, batch_first=True).permute(0, 2, 1, 3)

        # core attention
        if FLASH_ATTN_AVALIABLE:
            query = query.permute(0, 2, 1, 3)   # batch, sequence, num_head, head_dim
            key = key.permute(0, 2, 1, 3)
            value = value.permute(0, 2, 1, 3)
            
            query = torch.cat([u[:l] for u, l in zip(query, q_lens)], dim=0)
            key = torch.cat([u[:l] for u, l in zip(key, k_lens)], dim=0)
            value = torch.cat([u[:l] for u, l in zip(value, k_lens)], dim=0)
            cu_seqlens_q = F.pad(q_lens.cumsum(dim=0), (1, 0)).to(torch.int32)
            cu_seqlens_k = F.pad(k_lens.cumsum(dim=0), (1, 0)).to(torch.int32)
            max_seqlen_q = torch.max(q_lens).item()
            max_seqlen_k = torch.max(k_lens).item()

            hidden_states = flash_attn_varlen_func(query, key, value, cu_seqlens_q, cu_seqlens_k, max_seqlen_q, max_seqlen_k)
            hidden_states = pad_sequence([
                hidden_states[start: end]
                for start, end in zip(cu_seqlens_q[:-1], cu_seqlens_q[1:])
            ], batch_first=True)
            hidden_states = hidden_states.reshape(batch_size, -1, attn.heads * head_dim)

        else:
            attn_mask = torch.zeros((query.size(0), 1, query.size(2), key.size(2)), dtype=torch.bool).to(query)
            for i, (q_len, k_len) in enumerate(zip(q_lens, k_lens)):
                attn_mask[i, :, :q_len, :k_len] = True

            hidden_states = F.scaled_dot_product_attention(query, key, value, attn_mask=attn_mask, dropout_p=0.0, is_causal=False)
            hidden_states = hidden_states.transpose(1, 2).reshape(batch_size, -1, attn.heads * head_dim)

        hidden_states = hidden_states.to(query.dtype)

        if encoder_hidden_states is not None:
            encoder_hidden_states, hidden_states = (
                hidden_states[:, : encoder_hidden_states.shape[1]],
                hidden_states[:, encoder_hidden_states.shape[1] :],
            )

            # linear proj
            hidden_states = attn.to_out[0](hidden_states)
            # dropout
            hidden_states = attn.to_out[1](hidden_states)
            encoder_hidden_states = attn.to_add_out(encoder_hidden_states)

            return hidden_states, encoder_hidden_states
        else:
            return hidden_states