Spaces:
Paused
Paused
File size: 5,543 Bytes
4eb0627 | 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 | import torch
from typing import Optional
from einops import rearrange
from xfuser.core.distributed import (get_sequence_parallel_rank,
get_sequence_parallel_world_size,
get_sp_group)
# from xfuser.core.long_ctx_attention import xFuserLongContextAttention
from yunchang import LongContextAttention
def sinusoidal_embedding_1d(dim, position):
sinusoid = torch.outer(position.type(torch.float64), torch.pow(
10000, -torch.arange(dim//2, dtype=torch.float64, device=position.device).div(dim//2)))
x = torch.cat([torch.cos(sinusoid), torch.sin(sinusoid)], dim=1)
return x.to(position.dtype)
def pad_freqs(original_tensor, target_len):
seq_len, s1, s2 = original_tensor.shape
pad_size = target_len - seq_len
padding_tensor = torch.ones(
pad_size,
s1,
s2,
dtype=original_tensor.dtype,
device=original_tensor.device)
padded_tensor = torch.cat([original_tensor, padding_tensor], dim=0)
return padded_tensor
def rope_apply(x, freqs, num_heads):
x = rearrange(x, "b s (n d) -> b s n d", n=num_heads)
s_per_rank = x.shape[1]
x_out = torch.view_as_complex(x.to(torch.float64).reshape(
x.shape[0], x.shape[1], x.shape[2], -1, 2))
sp_size = get_sequence_parallel_world_size()
sp_rank = get_sequence_parallel_rank()
freqs = pad_freqs(freqs, s_per_rank * sp_size)
freqs_rank = freqs[(sp_rank * s_per_rank):((sp_rank + 1) * s_per_rank), :, :]
x_out = torch.view_as_real(x_out * freqs_rank).flatten(2)
return x_out.to(x.dtype)
def sp_gather_with_grad(x):
"""直接调用gather会导致梯度丢失。这里先gather，再把gather后的向量通过cat和x链接，
从而实现保梯度的gather操作.
Args:
x (_type_): _description_
Returns:
_type_: _description_
"""
from xfuser.core.distributed import (get_sequence_parallel_rank,
get_sequence_parallel_world_size,
get_sp_group)
x_gather = get_sp_group().all_gather(x, dim=1)
total = get_sequence_parallel_world_size() * x.shape[1]
start = x.shape[1] * get_sequence_parallel_rank()
end = x.shape[1] * (get_sequence_parallel_rank() + 1)
x = torch.cat([x_gather[:, :start], x, x_gather[:, end:]], dim=1)
return x
def usp_dit_forward(self,
x: torch.Tensor,
timestep: torch.Tensor,
context: torch.Tensor,
clip_feature: Optional[torch.Tensor] = None,
y: Optional[torch.Tensor] = None,
use_gradient_checkpointing: bool = False,
use_gradient_checkpointing_offload: bool = False,
**kwargs,
):
t = self.time_embedding(
sinusoidal_embedding_1d(self.freq_dim, timestep))
t_mod = self.time_projection(t).unflatten(1, (6, self.dim))
context = self.text_embedding(context)
if self.has_image_input:
x = torch.cat([x, y], dim=1) # (b, c_x + c_y, f, h, w)
clip_embdding = self.img_emb(clip_feature)
context = torch.cat([clip_embdding, context], dim=1)
x, (f, h, w) = self.patchify(x)
freqs = torch.cat([
self.freqs[0][:f].view(f, 1, 1, -1).expand(f, h, w, -1),
self.freqs[1][:h].view(1, h, 1, -1).expand(f, h, w, -1),
self.freqs[2][:w].view(1, 1, w, -1).expand(f, h, w, -1)
], dim=-1).reshape(f * h * w, 1, -1).to(x.device)
def create_custom_forward(module):
def custom_forward(*inputs):
return module(*inputs)
return custom_forward
# Context Parallel
x = torch.chunk(
x, get_sequence_parallel_world_size(),
dim=1)[get_sequence_parallel_rank()]
for block in self.blocks:
if self.training and use_gradient_checkpointing:
if use_gradient_checkpointing_offload:
with torch.autograd.graph.save_on_cpu():
x = torch.utils.checkpoint.checkpoint(
create_custom_forward(block),
x, context, t_mod, freqs,
use_reentrant=False,
)
else:
x = torch.utils.checkpoint.checkpoint(
create_custom_forward(block),
x, context, t_mod, freqs,
use_reentrant=False,
)
else:
x = block(x, context, t_mod, freqs)
x = self.head(x, t)
# Context Parallel
# x = get_sp_group().all_gather(x, dim=1)
x = sp_gather_with_grad(x)
# unpatchify
x = self.unpatchify(x, (f, h, w))
return x
def usp_attn_forward(self, x, freqs):
q = self.norm_q(self.q(x))
k = self.norm_k(self.k(x))
v = self.v(x)
q = rope_apply(q, freqs, self.num_heads)
k = rope_apply(k, freqs, self.num_heads)
q = rearrange(q, "b s (n d) -> b s n d", n=self.num_heads)
k = rearrange(k, "b s (n d) -> b s n d", n=self.num_heads)
v = rearrange(v, "b s (n d) -> b s n d", n=self.num_heads)
# x = xFuserLongContextAttention()(
# None,
# query=q,
# key=k,
# value=v,
# )
x = LongContextAttention()(
query=q,
key=k,
value=v,
)
x = x.flatten(2)
del q, k, v
torch.cuda.empty_cache()
return self.o(x) |