909 lines
37 KiB
Python
909 lines
37 KiB
Python
# Copyright 2023 The HuggingFace Team. All rights reserved.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
import torch
|
|
import torch.nn.functional as F
|
|
from torch import nn
|
|
|
|
from diffusers.utils.torch_utils import maybe_allow_in_graph
|
|
from diffusers.models.activations import get_activation
|
|
from diffusers.models.attention_processor import Attention
|
|
from diffusers.models.embeddings import CombinedTimestepLabelEmbeddings
|
|
|
|
from dataclasses import dataclass
|
|
from ..utils.base import BaseModule
|
|
from ..utils.typing import *
|
|
|
|
|
|
class MemoryEfficientAttentionMixin:
|
|
def enable_xformers_memory_efficient_attention(
|
|
self, attention_op: Optional[Callable] = None
|
|
):
|
|
r"""
|
|
Enable memory efficient attention from [xFormers](https://facebookresearch.github.io/xformers/). When this
|
|
option is enabled, you should observe lower GPU memory usage and a potential speed up during inference. Speed
|
|
up during training is not guaranteed.
|
|
|
|
<Tip warning={true}>
|
|
|
|
⚠️ When memory efficient attention and sliced attention are both enabled, memory efficient attention takes
|
|
precedent.
|
|
|
|
</Tip>
|
|
|
|
Parameters:
|
|
attention_op (`Callable`, *optional*):
|
|
Override the default `None` operator for use as `op` argument to the
|
|
[`memory_efficient_attention()`](https://facebookresearch.github.io/xformers/components/ops.html#xformers.ops.memory_efficient_attention)
|
|
function of xFormers.
|
|
|
|
Examples:
|
|
|
|
```py
|
|
>>> import torch
|
|
>>> from diffusers import DiffusionPipeline
|
|
>>> from xformers.ops import MemoryEfficientAttentionFlashAttentionOp
|
|
|
|
>>> pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-2-1", torch_dtype=torch.float16)
|
|
>>> pipe = pipe.to("cuda")
|
|
>>> pipe.enable_xformers_memory_efficient_attention(attention_op=MemoryEfficientAttentionFlashAttentionOp)
|
|
>>> # Workaround for not accepting attention shape using VAE for Flash Attention
|
|
>>> pipe.vae.enable_xformers_memory_efficient_attention(attention_op=None)
|
|
```
|
|
"""
|
|
self.set_use_memory_efficient_attention_xformers(True, attention_op)
|
|
|
|
def disable_xformers_memory_efficient_attention(self):
|
|
r"""
|
|
Disable memory efficient attention from [xFormers](https://facebookresearch.github.io/xformers/).
|
|
"""
|
|
self.set_use_memory_efficient_attention_xformers(False)
|
|
|
|
def set_use_memory_efficient_attention_xformers(
|
|
self, valid: bool, attention_op: Optional[Callable] = None
|
|
) -> None:
|
|
# Recursively walk through all the children.
|
|
# Any children which exposes the set_use_memory_efficient_attention_xformers method
|
|
# gets the message
|
|
def fn_recursive_set_mem_eff(module: torch.nn.Module):
|
|
if hasattr(module, "set_use_memory_efficient_attention_xformers"):
|
|
module.set_use_memory_efficient_attention_xformers(valid, attention_op)
|
|
|
|
for child in module.children():
|
|
fn_recursive_set_mem_eff(child)
|
|
|
|
for module in self.children():
|
|
if isinstance(module, torch.nn.Module):
|
|
fn_recursive_set_mem_eff(module)
|
|
|
|
|
|
@maybe_allow_in_graph
|
|
class GatedSelfAttentionDense(nn.Module):
|
|
r"""
|
|
A gated self-attention dense layer that combines visual features and object features.
|
|
|
|
Parameters:
|
|
query_dim (`int`): The number of channels in the query.
|
|
context_dim (`int`): The number of channels in the context.
|
|
n_heads (`int`): The number of heads to use for attention.
|
|
d_head (`int`): The number of channels in each head.
|
|
"""
|
|
|
|
def __init__(self, query_dim: int, context_dim: int, n_heads: int, d_head: int):
|
|
super().__init__()
|
|
|
|
# we need a linear projection since we need cat visual feature and obj feature
|
|
self.linear = nn.Linear(context_dim, query_dim)
|
|
|
|
self.attn = Attention(query_dim=query_dim, heads=n_heads, dim_head=d_head)
|
|
self.ff = FeedForward(query_dim, activation_fn="geglu")
|
|
|
|
self.norm1 = nn.LayerNorm(query_dim)
|
|
self.norm2 = nn.LayerNorm(query_dim)
|
|
|
|
self.register_parameter("alpha_attn", nn.Parameter(torch.tensor(0.0)))
|
|
self.register_parameter("alpha_dense", nn.Parameter(torch.tensor(0.0)))
|
|
|
|
self.enabled = True
|
|
|
|
def forward(self, x: torch.Tensor, objs: torch.Tensor) -> torch.Tensor:
|
|
if not self.enabled:
|
|
return x
|
|
|
|
n_visual = x.shape[1]
|
|
objs = self.linear(objs)
|
|
|
|
x = (
|
|
x
|
|
+ self.alpha_attn.tanh()
|
|
* self.attn(self.norm1(torch.cat([x, objs], dim=1)))[:, :n_visual, :]
|
|
)
|
|
x = x + self.alpha_dense.tanh() * self.ff(self.norm2(x))
|
|
|
|
return x
|
|
|
|
|
|
@maybe_allow_in_graph
|
|
class BasicTransformerBlock(nn.Module, MemoryEfficientAttentionMixin):
|
|
r"""
|
|
A basic Transformer block.
|
|
|
|
Parameters:
|
|
dim (`int`): The number of channels in the input and output.
|
|
num_attention_heads (`int`): The number of heads to use for multi-head attention.
|
|
attention_head_dim (`int`): The number of channels in each head.
|
|
dropout (`float`, *optional*, defaults to 0.0): The dropout probability to use.
|
|
cross_attention_dim (`int`, *optional*): The size of the encoder_hidden_states vector for cross attention.
|
|
activation_fn (`str`, *optional*, defaults to `"geglu"`): Activation function to be used in feed-forward.
|
|
num_embeds_ada_norm (:
|
|
obj: `int`, *optional*): The number of diffusion steps used during training. See `Transformer2DModel`.
|
|
attention_bias (:
|
|
obj: `bool`, *optional*, defaults to `False`): Configure if the attentions should contain a bias parameter.
|
|
only_cross_attention (`bool`, *optional*):
|
|
Whether to use only cross-attention layers. In this case two cross attention layers are used.
|
|
double_self_attention (`bool`, *optional*):
|
|
Whether to use two self-attention layers. In this case no cross attention layers are used.
|
|
upcast_attention (`bool`, *optional*):
|
|
Whether to upcast the attention computation to float32. This is useful for mixed precision training.
|
|
norm_elementwise_affine (`bool`, *optional*, defaults to `True`):
|
|
Whether to use learnable elementwise affine parameters for normalization.
|
|
norm_type (`str`, *optional*, defaults to `"layer_norm"`):
|
|
The normalization layer to use. Can be `"layer_norm"`, `"ada_norm"` or `"ada_norm_zero"`.
|
|
final_dropout (`bool` *optional*, defaults to False):
|
|
Whether to apply a final dropout after the last feed-forward layer.
|
|
attention_type (`str`, *optional*, defaults to `"default"`):
|
|
The type of attention to use. Can be `"default"` or `"gated"` or `"gated-text-image"`.
|
|
"""
|
|
|
|
def __init__(
|
|
self,
|
|
dim: int,
|
|
num_attention_heads: int,
|
|
attention_head_dim: int,
|
|
dropout=0.0,
|
|
cross_attention_dim: Optional[int] = None,
|
|
activation_fn: str = "geglu",
|
|
num_embeds_ada_norm: Optional[int] = None,
|
|
cond_dim_ada_norm_continuous: Optional[int] = None,
|
|
attention_bias: bool = False,
|
|
only_cross_attention: bool = False,
|
|
double_self_attention: bool = False,
|
|
upcast_attention: bool = False,
|
|
norm_elementwise_affine: bool = True,
|
|
norm_type: str = "layer_norm",
|
|
final_dropout: bool = False,
|
|
attention_type: str = "default",
|
|
):
|
|
super().__init__()
|
|
self.only_cross_attention = only_cross_attention
|
|
|
|
self.use_ada_layer_norm_zero = (
|
|
num_embeds_ada_norm is not None
|
|
) and norm_type == "ada_norm_zero"
|
|
self.use_ada_layer_norm = (
|
|
num_embeds_ada_norm is not None
|
|
) and norm_type == "ada_norm"
|
|
self.use_ada_layer_norm_continuous = (
|
|
cond_dim_ada_norm_continuous is not None
|
|
) and norm_type == "ada_norm_continuous"
|
|
|
|
assert (
|
|
int(self.use_ada_layer_norm)
|
|
+ int(self.use_ada_layer_norm_continuous)
|
|
+ int(self.use_ada_layer_norm_zero)
|
|
<= 1
|
|
)
|
|
|
|
if norm_type in ("ada_norm", "ada_norm_zero") and num_embeds_ada_norm is None:
|
|
raise ValueError(
|
|
f"`norm_type` is set to {norm_type}, but `num_embeds_ada_norm` is not defined. Please make sure to"
|
|
f" define `num_embeds_ada_norm` if setting `norm_type` to {norm_type}."
|
|
)
|
|
|
|
# Define 3 blocks. Each block has its own normalization layer.
|
|
# 1. Self-Attn
|
|
if self.use_ada_layer_norm:
|
|
self.norm1 = AdaLayerNorm(dim, num_embeds_ada_norm)
|
|
elif self.use_ada_layer_norm_continuous:
|
|
self.norm1 = AdaLayerNormContinuous(dim, cond_dim_ada_norm_continuous)
|
|
elif self.use_ada_layer_norm_zero:
|
|
self.norm1 = AdaLayerNormZero(dim, num_embeds_ada_norm)
|
|
else:
|
|
self.norm1 = nn.LayerNorm(dim, elementwise_affine=norm_elementwise_affine)
|
|
self.attn1 = Attention(
|
|
query_dim=dim,
|
|
heads=num_attention_heads,
|
|
dim_head=attention_head_dim,
|
|
dropout=dropout,
|
|
bias=attention_bias,
|
|
cross_attention_dim=cross_attention_dim if only_cross_attention else None,
|
|
upcast_attention=upcast_attention,
|
|
)
|
|
|
|
# 2. Cross-Attn
|
|
if cross_attention_dim is not None or double_self_attention:
|
|
# We currently only use AdaLayerNormZero for self attention where there will only be one attention block.
|
|
# I.e. the number of returned modulation chunks from AdaLayerZero would not make sense if returned during
|
|
# the second cross attention block.
|
|
if self.use_ada_layer_norm:
|
|
self.norm2 = AdaLayerNorm(dim, num_embeds_ada_norm)
|
|
elif self.use_ada_layer_norm_continuous:
|
|
self.norm2 = AdaLayerNormContinuous(dim, cond_dim_ada_norm_continuous)
|
|
else:
|
|
self.norm2 = nn.LayerNorm(
|
|
dim, elementwise_affine=norm_elementwise_affine
|
|
)
|
|
|
|
self.attn2 = Attention(
|
|
query_dim=dim,
|
|
cross_attention_dim=cross_attention_dim
|
|
if not double_self_attention
|
|
else None,
|
|
heads=num_attention_heads,
|
|
dim_head=attention_head_dim,
|
|
dropout=dropout,
|
|
bias=attention_bias,
|
|
upcast_attention=upcast_attention,
|
|
) # is self-attn if encoder_hidden_states is none
|
|
else:
|
|
self.norm2 = None
|
|
self.attn2 = None
|
|
|
|
# 3. Feed-forward
|
|
if self.use_ada_layer_norm_continuous:
|
|
self.norm3 = AdaLayerNormContinuous(dim, cond_dim_ada_norm_continuous)
|
|
else:
|
|
self.norm3 = nn.LayerNorm(dim, elementwise_affine=norm_elementwise_affine)
|
|
self.ff = FeedForward(
|
|
dim,
|
|
dropout=dropout,
|
|
activation_fn=activation_fn,
|
|
final_dropout=final_dropout,
|
|
)
|
|
|
|
# 4. Fuser
|
|
if attention_type == "gated" or attention_type == "gated-text-image":
|
|
self.fuser = GatedSelfAttentionDense(
|
|
dim, cross_attention_dim, num_attention_heads, attention_head_dim
|
|
)
|
|
|
|
# let chunk size default to None
|
|
self._chunk_size = None
|
|
self._chunk_dim = 0
|
|
|
|
def set_chunk_feed_forward(self, chunk_size: Optional[int], dim: int):
|
|
# Sets chunk feed-forward
|
|
self._chunk_size = chunk_size
|
|
self._chunk_dim = dim
|
|
|
|
def forward(
|
|
self,
|
|
hidden_states: torch.FloatTensor,
|
|
attention_mask: Optional[torch.FloatTensor] = None,
|
|
encoder_hidden_states: Optional[torch.FloatTensor] = None,
|
|
encoder_attention_mask: Optional[torch.FloatTensor] = None,
|
|
timestep: Optional[torch.LongTensor] = None,
|
|
modulation_cond: Optional[torch.FloatTensor] = None,
|
|
cross_attention_kwargs: Dict[str, Any] = None,
|
|
class_labels: Optional[torch.LongTensor] = None,
|
|
) -> torch.FloatTensor:
|
|
# Notice that normalization is always applied before the real computation in the following blocks.
|
|
# 0. Self-Attention
|
|
if self.use_ada_layer_norm:
|
|
norm_hidden_states = self.norm1(hidden_states, timestep)
|
|
elif self.use_ada_layer_norm_continuous:
|
|
norm_hidden_states = self.norm1(hidden_states, modulation_cond)
|
|
elif self.use_ada_layer_norm_zero:
|
|
norm_hidden_states, gate_msa, shift_mlp, scale_mlp, gate_mlp = self.norm1(
|
|
hidden_states, timestep, class_labels, hidden_dtype=hidden_states.dtype
|
|
)
|
|
else:
|
|
norm_hidden_states = self.norm1(hidden_states)
|
|
|
|
# 1. Retrieve lora scale.
|
|
lora_scale = (
|
|
cross_attention_kwargs.get("scale", 1.0)
|
|
if cross_attention_kwargs is not None
|
|
else 1.0
|
|
)
|
|
|
|
# 2. Prepare GLIGEN inputs
|
|
cross_attention_kwargs = (
|
|
cross_attention_kwargs.copy() if cross_attention_kwargs is not None else {}
|
|
)
|
|
gligen_kwargs = cross_attention_kwargs.pop("gligen", None)
|
|
|
|
attn_output = self.attn1(
|
|
norm_hidden_states,
|
|
encoder_hidden_states=encoder_hidden_states
|
|
if self.only_cross_attention
|
|
else None,
|
|
attention_mask=attention_mask,
|
|
**cross_attention_kwargs,
|
|
)
|
|
if self.use_ada_layer_norm_zero:
|
|
attn_output = gate_msa.unsqueeze(1) * attn_output
|
|
hidden_states = attn_output + hidden_states
|
|
|
|
# 2.5 GLIGEN Control
|
|
if gligen_kwargs is not None:
|
|
hidden_states = self.fuser(hidden_states, gligen_kwargs["objs"])
|
|
# 2.5 ends
|
|
|
|
# 3. Cross-Attention
|
|
if self.attn2 is not None:
|
|
if self.use_ada_layer_norm:
|
|
norm_hidden_states = self.norm2(hidden_states, timestep)
|
|
elif self.use_ada_layer_norm_continuous:
|
|
norm_hidden_states = self.norm2(hidden_states, modulation_cond)
|
|
else:
|
|
norm_hidden_states = self.norm2(hidden_states)
|
|
|
|
attn_output = self.attn2(
|
|
norm_hidden_states,
|
|
encoder_hidden_states=encoder_hidden_states,
|
|
attention_mask=encoder_attention_mask,
|
|
**cross_attention_kwargs,
|
|
)
|
|
hidden_states = attn_output + hidden_states
|
|
|
|
# 4. Feed-forward
|
|
if self.use_ada_layer_norm_continuous:
|
|
norm_hidden_states = self.norm3(hidden_states, modulation_cond)
|
|
else:
|
|
norm_hidden_states = self.norm3(hidden_states)
|
|
|
|
if self.use_ada_layer_norm_zero:
|
|
norm_hidden_states = (
|
|
norm_hidden_states * (1 + scale_mlp[:, None]) + shift_mlp[:, None]
|
|
)
|
|
|
|
if self._chunk_size is not None:
|
|
# "feed_forward_chunk_size" can be used to save memory
|
|
if norm_hidden_states.shape[self._chunk_dim] % self._chunk_size != 0:
|
|
raise ValueError(
|
|
f"`hidden_states` dimension to be chunked: {norm_hidden_states.shape[self._chunk_dim]} has to be divisible by chunk size: {self._chunk_size}. Make sure to set an appropriate `chunk_size` when calling `unet.enable_forward_chunking`."
|
|
)
|
|
|
|
num_chunks = norm_hidden_states.shape[self._chunk_dim] // self._chunk_size
|
|
ff_output = torch.cat(
|
|
[
|
|
self.ff(hid_slice, scale=lora_scale)
|
|
for hid_slice in norm_hidden_states.chunk(
|
|
num_chunks, dim=self._chunk_dim
|
|
)
|
|
],
|
|
dim=self._chunk_dim,
|
|
)
|
|
else:
|
|
ff_output = self.ff(norm_hidden_states, scale=lora_scale)
|
|
|
|
if self.use_ada_layer_norm_zero:
|
|
ff_output = gate_mlp.unsqueeze(1) * ff_output
|
|
|
|
hidden_states = ff_output + hidden_states
|
|
|
|
return hidden_states
|
|
|
|
|
|
class FeedForward(nn.Module):
|
|
r"""
|
|
A feed-forward layer.
|
|
|
|
Parameters:
|
|
dim (`int`): The number of channels in the input.
|
|
dim_out (`int`, *optional*): The number of channels in the output. If not given, defaults to `dim`.
|
|
mult (`int`, *optional*, defaults to 4): The multiplier to use for the hidden dimension.
|
|
dropout (`float`, *optional*, defaults to 0.0): The dropout probability to use.
|
|
activation_fn (`str`, *optional*, defaults to `"geglu"`): Activation function to be used in feed-forward.
|
|
final_dropout (`bool` *optional*, defaults to False): Apply a final dropout.
|
|
"""
|
|
|
|
def __init__(
|
|
self,
|
|
dim: int,
|
|
dim_out: Optional[int] = None,
|
|
mult: int = 4,
|
|
dropout: float = 0.0,
|
|
activation_fn: str = "geglu",
|
|
final_dropout: bool = False,
|
|
):
|
|
super().__init__()
|
|
inner_dim = int(dim * mult)
|
|
dim_out = dim_out if dim_out is not None else dim
|
|
linear_cls = nn.Linear
|
|
|
|
if activation_fn == "gelu":
|
|
act_fn = GELU(dim, inner_dim)
|
|
if activation_fn == "gelu-approximate":
|
|
act_fn = GELU(dim, inner_dim, approximate="tanh")
|
|
elif activation_fn == "geglu":
|
|
act_fn = GEGLU(dim, inner_dim)
|
|
elif activation_fn == "geglu-approximate":
|
|
act_fn = ApproximateGELU(dim, inner_dim)
|
|
|
|
self.net = nn.ModuleList([])
|
|
# project in
|
|
self.net.append(act_fn)
|
|
# project dropout
|
|
self.net.append(nn.Dropout(dropout))
|
|
# project out
|
|
self.net.append(linear_cls(inner_dim, dim_out))
|
|
# FF as used in Vision Transformer, MLP-Mixer, etc. have a final dropout
|
|
if final_dropout:
|
|
self.net.append(nn.Dropout(dropout))
|
|
|
|
def forward(self, hidden_states: torch.Tensor, scale: float = 1.0) -> torch.Tensor:
|
|
for module in self.net:
|
|
hidden_states = module(hidden_states)
|
|
return hidden_states
|
|
|
|
|
|
class GELU(nn.Module):
|
|
r"""
|
|
GELU activation function with tanh approximation support with `approximate="tanh"`.
|
|
|
|
Parameters:
|
|
dim_in (`int`): The number of channels in the input.
|
|
dim_out (`int`): The number of channels in the output.
|
|
approximate (`str`, *optional*, defaults to `"none"`): If `"tanh"`, use tanh approximation.
|
|
"""
|
|
|
|
def __init__(self, dim_in: int, dim_out: int, approximate: str = "none"):
|
|
super().__init__()
|
|
self.proj = nn.Linear(dim_in, dim_out)
|
|
self.approximate = approximate
|
|
|
|
def gelu(self, gate: torch.Tensor) -> torch.Tensor:
|
|
if gate.device.type != "mps":
|
|
return F.gelu(gate, approximate=self.approximate)
|
|
# mps: gelu is not implemented for float16
|
|
return F.gelu(gate.to(dtype=torch.float32), approximate=self.approximate).to(
|
|
dtype=gate.dtype
|
|
)
|
|
|
|
def forward(self, hidden_states):
|
|
hidden_states = self.proj(hidden_states)
|
|
hidden_states = self.gelu(hidden_states)
|
|
return hidden_states
|
|
|
|
|
|
class GEGLU(nn.Module):
|
|
r"""
|
|
A variant of the gated linear unit activation function from https://arxiv.org/abs/2002.05202.
|
|
|
|
Parameters:
|
|
dim_in (`int`): The number of channels in the input.
|
|
dim_out (`int`): The number of channels in the output.
|
|
"""
|
|
|
|
def __init__(self, dim_in: int, dim_out: int):
|
|
super().__init__()
|
|
linear_cls = nn.Linear
|
|
|
|
self.proj = linear_cls(dim_in, dim_out * 2)
|
|
|
|
def gelu(self, gate: torch.Tensor) -> torch.Tensor:
|
|
if gate.device.type != "mps":
|
|
return F.gelu(gate)
|
|
# mps: gelu is not implemented for float16
|
|
return F.gelu(gate.to(dtype=torch.float32)).to(dtype=gate.dtype)
|
|
|
|
def forward(self, hidden_states, scale: float = 1.0):
|
|
args = ()
|
|
hidden_states, gate = self.proj(hidden_states, *args).chunk(2, dim=-1)
|
|
return hidden_states * self.gelu(gate)
|
|
|
|
|
|
class ApproximateGELU(nn.Module):
|
|
r"""
|
|
The approximate form of Gaussian Error Linear Unit (GELU). For more details, see section 2:
|
|
https://arxiv.org/abs/1606.08415.
|
|
|
|
Parameters:
|
|
dim_in (`int`): The number of channels in the input.
|
|
dim_out (`int`): The number of channels in the output.
|
|
"""
|
|
|
|
def __init__(self, dim_in: int, dim_out: int):
|
|
super().__init__()
|
|
self.proj = nn.Linear(dim_in, dim_out)
|
|
|
|
def forward(self, x: torch.Tensor) -> torch.Tensor:
|
|
x = self.proj(x)
|
|
return x * torch.sigmoid(1.702 * x)
|
|
|
|
|
|
class AdaLayerNorm(nn.Module):
|
|
r"""
|
|
Norm layer modified to incorporate timestep embeddings.
|
|
|
|
Parameters:
|
|
embedding_dim (`int`): The size of each embedding vector.
|
|
num_embeddings (`int`): The size of the dictionary of embeddings.
|
|
"""
|
|
|
|
def __init__(self, embedding_dim: int, num_embeddings: int):
|
|
super().__init__()
|
|
self.emb = nn.Embedding(num_embeddings, embedding_dim)
|
|
self.silu = nn.SiLU()
|
|
self.linear = nn.Linear(embedding_dim, embedding_dim * 2)
|
|
self.norm = nn.LayerNorm(embedding_dim, elementwise_affine=False)
|
|
|
|
def forward(self, x: torch.Tensor, timestep: torch.Tensor) -> torch.Tensor:
|
|
emb = self.linear(self.silu(self.emb(timestep)))
|
|
scale, shift = torch.chunk(emb, 2, dim=1)
|
|
x = self.norm(x) * (1 + scale.unsqueeze(1)) + shift.unsqueeze(1)
|
|
return x
|
|
|
|
|
|
class AdaLayerNormContinuous(nn.Module):
|
|
r"""
|
|
Norm layer modified to incorporate arbitrary continuous embeddings.
|
|
|
|
Parameters:
|
|
embedding_dim (`int`): The size of each embedding vector.
|
|
"""
|
|
|
|
def __init__(self, embedding_dim: int, condition_dim: int):
|
|
super().__init__()
|
|
self.silu = nn.SiLU()
|
|
self.linear1 = nn.Linear(condition_dim, condition_dim)
|
|
self.linear2 = nn.Linear(condition_dim, embedding_dim * 2)
|
|
self.norm = nn.LayerNorm(embedding_dim, elementwise_affine=False)
|
|
|
|
def forward(self, x: torch.Tensor, condition: torch.Tensor) -> torch.Tensor:
|
|
emb = self.linear2(self.silu(self.linear1(condition)))
|
|
scale, shift = torch.chunk(emb, 2, dim=1)
|
|
x = self.norm(x) * (1 + scale.unsqueeze(1)) + shift.unsqueeze(1)
|
|
return x
|
|
|
|
|
|
class Modulation(nn.Module):
|
|
def __init__(self, embedding_dim: int, condition_dim: int, zero_init: bool = False, single_layer: bool = False):
|
|
super().__init__()
|
|
self.silu = nn.SiLU()
|
|
if single_layer:
|
|
self.linear1 = nn.Identity()
|
|
else:
|
|
self.linear1 = nn.Linear(condition_dim, condition_dim)
|
|
|
|
self.linear2 = nn.Linear(condition_dim, embedding_dim * 2)
|
|
|
|
# Only zero init the last linear layer
|
|
if zero_init:
|
|
nn.init.zeros_(self.linear2.weight)
|
|
nn.init.zeros_(self.linear2.bias)
|
|
|
|
def forward(self, x: torch.Tensor, condition: torch.Tensor) -> torch.Tensor:
|
|
emb = self.linear2(self.silu(self.linear1(condition)))
|
|
scale, shift = torch.chunk(emb, 2, dim=1)
|
|
x = x * (1 + scale.unsqueeze(1)) + shift.unsqueeze(1)
|
|
return x
|
|
|
|
|
|
class AdaLayerNormZero(nn.Module):
|
|
r"""
|
|
Norm layer adaptive layer norm zero (adaLN-Zero).
|
|
|
|
Parameters:
|
|
embedding_dim (`int`): The size of each embedding vector.
|
|
num_embeddings (`int`): The size of the dictionary of embeddings.
|
|
"""
|
|
|
|
def __init__(self, embedding_dim: int, num_embeddings: int):
|
|
super().__init__()
|
|
|
|
self.emb = CombinedTimestepLabelEmbeddings(num_embeddings, embedding_dim)
|
|
|
|
self.silu = nn.SiLU()
|
|
self.linear = nn.Linear(embedding_dim, 6 * embedding_dim, bias=True)
|
|
self.norm = nn.LayerNorm(embedding_dim, elementwise_affine=False, eps=1e-6)
|
|
|
|
def forward(
|
|
self,
|
|
x: torch.Tensor,
|
|
timestep: torch.Tensor,
|
|
class_labels: torch.LongTensor,
|
|
hidden_dtype: Optional[torch.dtype] = None,
|
|
) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor]:
|
|
emb = self.linear(
|
|
self.silu(self.emb(timestep, class_labels, hidden_dtype=hidden_dtype))
|
|
)
|
|
shift_msa, scale_msa, gate_msa, shift_mlp, scale_mlp, gate_mlp = emb.chunk(
|
|
6, dim=1
|
|
)
|
|
x = self.norm(x) * (1 + scale_msa[:, None]) + shift_msa[:, None]
|
|
return x, gate_msa, shift_mlp, scale_mlp, gate_mlp
|
|
|
|
|
|
class AdaGroupNorm(nn.Module):
|
|
r"""
|
|
GroupNorm layer modified to incorporate timestep embeddings.
|
|
|
|
Parameters:
|
|
embedding_dim (`int`): The size of each embedding vector.
|
|
num_embeddings (`int`): The size of the dictionary of embeddings.
|
|
num_groups (`int`): The number of groups to separate the channels into.
|
|
act_fn (`str`, *optional*, defaults to `None`): The activation function to use.
|
|
eps (`float`, *optional*, defaults to `1e-5`): The epsilon value to use for numerical stability.
|
|
"""
|
|
|
|
def __init__(
|
|
self,
|
|
embedding_dim: int,
|
|
out_dim: int,
|
|
num_groups: int,
|
|
act_fn: Optional[str] = None,
|
|
eps: float = 1e-5,
|
|
):
|
|
super().__init__()
|
|
self.num_groups = num_groups
|
|
self.eps = eps
|
|
|
|
if act_fn is None:
|
|
self.act = None
|
|
else:
|
|
self.act = get_activation(act_fn)
|
|
|
|
self.linear = nn.Linear(embedding_dim, out_dim * 2)
|
|
|
|
def forward(self, x: torch.Tensor, emb: torch.Tensor) -> torch.Tensor:
|
|
if self.act:
|
|
emb = self.act(emb)
|
|
emb = self.linear(emb)
|
|
emb = emb[:, :, None, None]
|
|
scale, shift = emb.chunk(2, dim=1)
|
|
|
|
x = F.group_norm(x, self.num_groups, eps=self.eps)
|
|
x = x * (1 + scale) + shift
|
|
return x
|
|
|
|
class Transformer1D(BaseModule, MemoryEfficientAttentionMixin):
|
|
"""
|
|
A 1D Transformer model for sequence data.
|
|
|
|
Parameters:
|
|
num_attention_heads (`int`, *optional*, defaults to 16): The number of heads to use for multi-head attention.
|
|
attention_head_dim (`int`, *optional*, defaults to 88): The number of channels in each head.
|
|
in_channels (`int`, *optional*):
|
|
The number of channels in the input and output (specify if the input is **continuous**).
|
|
num_layers (`int`, *optional*, defaults to 1): The number of layers of Transformer blocks to use.
|
|
dropout (`float`, *optional*, defaults to 0.0): The dropout probability to use.
|
|
cross_attention_dim (`int`, *optional*): The number of `encoder_hidden_states` dimensions to use.
|
|
activation_fn (`str`, *optional*, defaults to `"geglu"`): Activation function to use in feed-forward.
|
|
num_embeds_ada_norm ( `int`, *optional*):
|
|
The number of diffusion steps used during training. Pass if at least one of the norm_layers is
|
|
`AdaLayerNorm`. This is fixed during training since it is used to learn a number of embeddings that are
|
|
added to the hidden states.
|
|
|
|
During inference, you can denoise for up to but not more steps than `num_embeds_ada_norm`.
|
|
attention_bias (`bool`, *optional*):
|
|
Configure if the `TransformerBlocks` attention should contain a bias parameter.
|
|
"""
|
|
|
|
@dataclass
|
|
class Config(BaseModule.Config):
|
|
num_attention_heads: int = 16
|
|
attention_head_dim: int = 88
|
|
in_channels: Optional[int] = None
|
|
out_channels: Optional[int] = None
|
|
num_layers: int = 1
|
|
dropout: float = 0.0
|
|
norm_num_groups: int = 32
|
|
cross_attention_dim: Optional[int] = None
|
|
attention_bias: bool = False
|
|
activation_fn: str = "geglu"
|
|
num_embeds_ada_norm: Optional[int] = None
|
|
cond_dim_ada_norm_continuous: Optional[int] = None
|
|
only_cross_attention: bool = False
|
|
double_self_attention: bool = False
|
|
upcast_attention: bool = False
|
|
norm_type: str = "layer_norm"
|
|
norm_elementwise_affine: bool = True
|
|
attention_type: str = "default"
|
|
enable_memory_efficient_attention: bool = False
|
|
gradient_checkpointing: bool = False
|
|
|
|
cfg: Config
|
|
|
|
def configure(self) -> None:
|
|
super().configure()
|
|
|
|
self.num_attention_heads = self.cfg.num_attention_heads
|
|
self.attention_head_dim = self.cfg.attention_head_dim
|
|
inner_dim = self.num_attention_heads * self.attention_head_dim
|
|
|
|
linear_cls = nn.Linear
|
|
|
|
if self.cfg.norm_type == "layer_norm" and (
|
|
self.cfg.num_embeds_ada_norm is not None
|
|
or self.cfg.cond_dim_ada_norm_continuous is not None
|
|
):
|
|
raise ValueError("Incorrect norm_type.")
|
|
|
|
# 2. Define input layers
|
|
self.in_channels = self.cfg.in_channels
|
|
|
|
self.norm = torch.nn.GroupNorm(
|
|
num_groups=self.cfg.norm_num_groups,
|
|
num_channels=self.cfg.in_channels,
|
|
eps=1e-6,
|
|
affine=True,
|
|
)
|
|
self.proj_in = linear_cls(self.cfg.in_channels, inner_dim)
|
|
|
|
# 3. Define transformers blocks
|
|
self.transformer_blocks = nn.ModuleList(
|
|
[
|
|
BasicTransformerBlock(
|
|
inner_dim,
|
|
self.num_attention_heads,
|
|
self.attention_head_dim,
|
|
dropout=self.cfg.dropout,
|
|
cross_attention_dim=self.cfg.cross_attention_dim,
|
|
activation_fn=self.cfg.activation_fn,
|
|
num_embeds_ada_norm=self.cfg.num_embeds_ada_norm,
|
|
cond_dim_ada_norm_continuous=self.cfg.cond_dim_ada_norm_continuous,
|
|
attention_bias=self.cfg.attention_bias,
|
|
only_cross_attention=self.cfg.only_cross_attention,
|
|
double_self_attention=self.cfg.double_self_attention,
|
|
upcast_attention=self.cfg.upcast_attention,
|
|
norm_type=self.cfg.norm_type,
|
|
norm_elementwise_affine=self.cfg.norm_elementwise_affine,
|
|
attention_type=self.cfg.attention_type,
|
|
)
|
|
for d in range(self.cfg.num_layers)
|
|
]
|
|
)
|
|
|
|
# 4. Define output layers
|
|
self.out_channels = (
|
|
self.cfg.in_channels
|
|
if self.cfg.out_channels is None
|
|
else self.cfg.out_channels
|
|
)
|
|
|
|
self.proj_out = linear_cls(inner_dim, self.cfg.in_channels)
|
|
|
|
self.gradient_checkpointing = self.cfg.gradient_checkpointing
|
|
|
|
self.set_use_memory_efficient_attention_xformers(
|
|
self.cfg.enable_memory_efficient_attention
|
|
)
|
|
|
|
def forward(
|
|
self,
|
|
hidden_states: torch.Tensor,
|
|
encoder_hidden_states: Optional[torch.Tensor] = None,
|
|
timestep: Optional[torch.LongTensor] = None,
|
|
modulation_cond: Optional[torch.FloatTensor] = None,
|
|
class_labels: Optional[torch.LongTensor] = None,
|
|
cross_attention_kwargs: Dict[str, Any] = None,
|
|
attention_mask: Optional[torch.Tensor] = None,
|
|
encoder_attention_mask: Optional[torch.Tensor] = None,
|
|
):
|
|
"""
|
|
The [`Transformer1DModel`] forward method.
|
|
|
|
Args:
|
|
hidden_states (`torch.LongTensor` of shape `(batch size, num latent pixels)` if discrete, `torch.FloatTensor` of shape `(batch size, channel, height, width)` if continuous):
|
|
Input `hidden_states`.
|
|
encoder_hidden_states ( `torch.FloatTensor` of shape `(batch size, sequence len, embed dims)`, *optional*):
|
|
Conditional embeddings for cross attention layer. If not given, cross-attention defaults to
|
|
self-attention.
|
|
timestep ( `torch.LongTensor`, *optional*):
|
|
Used to indicate denoising step. Optional timestep to be applied as an embedding in `AdaLayerNorm`.
|
|
class_labels ( `torch.LongTensor` of shape `(batch size, num classes)`, *optional*):
|
|
Used to indicate class labels conditioning. Optional class labels to be applied as an embedding in
|
|
`AdaLayerZeroNorm`.
|
|
cross_attention_kwargs ( `Dict[str, Any]`, *optional*):
|
|
A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under
|
|
`self.processor` in
|
|
[diffusers.models.attention_processor](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
|
|
attention_mask ( `torch.Tensor`, *optional*):
|
|
An attention mask of shape `(batch, key_tokens)` is applied to `encoder_hidden_states`. If `1` the mask
|
|
is kept, otherwise if `0` it is discarded. Mask will be converted into a bias, which adds large
|
|
negative values to the attention scores corresponding to "discard" tokens.
|
|
encoder_attention_mask ( `torch.Tensor`, *optional*):
|
|
Cross-attention mask applied to `encoder_hidden_states`. Two formats supported:
|
|
|
|
* Mask `(batch, sequence_length)` True = keep, False = discard.
|
|
* Bias `(batch, 1, sequence_length)` 0 = keep, -10000 = discard.
|
|
|
|
If `ndim == 2`: will be interpreted as a mask, then converted into a bias consistent with the format
|
|
above. This bias will be added to the cross-attention scores.
|
|
return_dict (`bool`, *optional*, defaults to `True`):
|
|
Whether or not to return a [`~models.unet_2d_condition.UNet2DConditionOutput`] instead of a plain
|
|
tuple.
|
|
|
|
Returns:
|
|
If `return_dict` is True, an [`~models.transformer_2d.Transformer2DModelOutput`] is returned, otherwise a
|
|
`tuple` where the first element is the sample tensor.
|
|
"""
|
|
# ensure attention_mask is a bias, and give it a singleton query_tokens dimension.
|
|
# we may have done this conversion already, e.g. if we came here via UNet2DConditionModel#forward.
|
|
# we can tell by counting dims; if ndim == 2: it's a mask rather than a bias.
|
|
# expects mask of shape:
|
|
# [batch, key_tokens]
|
|
# adds singleton query_tokens dimension:
|
|
# [batch, 1, key_tokens]
|
|
# this helps to broadcast it as a bias over attention scores, which will be in one of the following shapes:
|
|
# [batch, heads, query_tokens, key_tokens] (e.g. torch sdp attn)
|
|
# [batch * heads, query_tokens, key_tokens] (e.g. xformers or classic attn)
|
|
if attention_mask is not None and attention_mask.ndim == 2:
|
|
# assume that mask is expressed as:
|
|
# (1 = keep, 0 = discard)
|
|
# convert mask into a bias that can be added to attention scores:
|
|
# (keep = +0, discard = -10000.0)
|
|
attention_mask = (1 - attention_mask.to(hidden_states.dtype)) * -10000.0
|
|
attention_mask = attention_mask.unsqueeze(1)
|
|
|
|
# convert encoder_attention_mask to a bias the same way we do for attention_mask
|
|
if encoder_attention_mask is not None and encoder_attention_mask.ndim == 2:
|
|
encoder_attention_mask = (
|
|
1 - encoder_attention_mask.to(hidden_states.dtype)
|
|
) * -10000.0
|
|
encoder_attention_mask = encoder_attention_mask.unsqueeze(1)
|
|
|
|
# 1. Input
|
|
batch, _, seq_len = hidden_states.shape
|
|
residual = hidden_states
|
|
|
|
hidden_states = self.norm(hidden_states)
|
|
inner_dim = hidden_states.shape[1]
|
|
hidden_states = hidden_states.permute(0, 2, 1).reshape(
|
|
batch, seq_len, inner_dim
|
|
)
|
|
hidden_states = self.proj_in(hidden_states)
|
|
|
|
# 2. Blocks
|
|
for block in self.transformer_blocks:
|
|
if self.training and self.gradient_checkpointing:
|
|
hidden_states = torch.utils.checkpoint.checkpoint(
|
|
block,
|
|
hidden_states,
|
|
attention_mask,
|
|
encoder_hidden_states,
|
|
encoder_attention_mask,
|
|
timestep,
|
|
modulation_cond,
|
|
cross_attention_kwargs,
|
|
class_labels,
|
|
use_reentrant=False,
|
|
)
|
|
else:
|
|
hidden_states = block(
|
|
hidden_states,
|
|
attention_mask=attention_mask,
|
|
encoder_hidden_states=encoder_hidden_states,
|
|
encoder_attention_mask=encoder_attention_mask,
|
|
timestep=timestep,
|
|
modulation_cond=modulation_cond,
|
|
cross_attention_kwargs=cross_attention_kwargs,
|
|
class_labels=class_labels,
|
|
)
|
|
|
|
# 3. Output
|
|
hidden_states = self.proj_out(hidden_states)
|
|
hidden_states = (
|
|
hidden_states.reshape(batch, seq_len, inner_dim)
|
|
.permute(0, 2, 1)
|
|
.contiguous()
|
|
)
|
|
|
|
output = hidden_states + residual
|
|
|
|
return output
|