Finegrain/refiners
1
0
Fork 0
mirror of https://github.com/finegrain-ai/refiners.git synced 2024-11-09 23:12:02 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

simplify even more CrossAttentionAdapter

Browse source 
Following Laurent2916's idea: see #167
This commit is contained in:
limiteinductive 2024-01-11 10:29:21 +01:00 committed by Benjamin Trom
parent 3ab8ed2989
commit deed703617
1 changed files with 27 additions and 18 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

45
src/refiners/foundationals/latent_diffusion/image_prompt.py
View file

@ -9,7 +9,6 @@ import refiners.fluxion.layers as fl
from refiners.fluxion.adapters.adapter import Adapter
from refiners.fluxion.context import Contexts
from refiners.fluxion.layers.attentions import ScaledDotProductAttention
from refiners.fluxion.layers.chain import Distribute
from refiners.fluxion.utils import image_to_tensor, normalize
from refiners.foundationals.clip.image_encoder import CLIPImageEncoderH
@ -234,16 +233,12 @@ class PerceiverResampler(fl.Chain):
return {"perceiver_resampler": {"x": None}}
class InjectionPoint(fl.Chain):
pass
class ImageCrossAttention(fl.Chain):
def __init__(self, text_cross_attention: fl.Attention, scale: float = 1.0) -> None:
self.scale = scale
self._scale = scale
super().__init__(
fl.Distribute(
fl.UseContext(context="ip_adapter", key="query_projection"),
fl.Identity(),
fl.Chain(
fl.UseContext(context="ip_adapter", key="clip_image_embedding"),
fl.Linear(
@ -271,10 +266,14 @@ class ImageCrossAttention(fl.Chain):
fl.Multiply(self.scale),
)
@property
def scale(self) -> float:
return self._scale
class SetQueryProjection(fl.Passthrough):
def __init__(self) -> None:
super().__init__(fl.GetArg(index=0), fl.SetContext(context="ip_adapter", key="query_projection"))
@scale.setter
def scale(self, value: float) -> None:
self._scale = value
self.ensure_find(fl.Multiply).scale = value
class CrossAttentionAdapter(fl.Chain, Adapter[fl.Attention]):
@ -283,15 +282,24 @@ class CrossAttentionAdapter(fl.Chain, Adapter[fl.Attention]):
target: fl.Attention,
scale: float = 1.0,
) -> None:
self._scale = scale
with self.setup_adapter(target):
super().__init__(
fl.Sum(
target[:-1], # original text cross attention
ImageCrossAttention(text_cross_attention=target, scale=scale),
),
target[-1], # projection
clone = target.structural_copy()
scaled_dot_product = clone.ensure_find(ScaledDotProductAttention)
image_cross_attention = ImageCrossAttention(
text_cross_attention=clone,
scale=self.scale,
)
clone.replace(
old_module=scaled_dot_product,
new_module=fl.Sum(
scaled_dot_product,
image_cross_attention,
),
)
super().__init__(
clone,
)
self.ensure_find(fl.Attention).insert_after_type(Distribute, SetQueryProjection())
@property
def image_cross_attention(self) -> ImageCrossAttention:
@ -307,10 +315,11 @@ class CrossAttentionAdapter(fl.Chain, Adapter[fl.Attention]):
@property
def scale(self) -> float:
return self.image_cross_attention.scale
return self._scale
@scale.setter
def scale(self, value: float) -> None:
self._scale = value
self.image_cross_attention.scale = value
def load_weights(self, key_tensor: Tensor, value_tensor: Tensor) -> None: