Mirrored_Repos/InvokeAI
1
0
Fork 0
mirror of https://github.com/invoke-ai/InvokeAI synced 2024-08-30 20:32:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update l2i invoke and seamless to support AutoencoderTiny, remove attention processors if no mid_block is detected

Browse Source
This commit is contained in:
Brandon Rising 2024-03-12 12:00:24 -04:00
parent 43948e0758
commit 7b393656de
2 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
invokeai/app/invocations/latent.py
View File

@ -837,14 +837,15 @@ class LatentsToImageInvocation(BaseInvocation, WithMetadata, WithBoard):
latents = context.tensors.load(self.latents.latents_name) latents = context.tensors.load(self.latents.latents_name)
vae_info = context.models.load(self.vae.vae) vae_info = context.models.load(self.vae.vae)
assert isinstance(vae_info.model, (UNet2DConditionModel, AutoencoderKL)) assert isinstance(vae_info.model, (UNet2DConditionModel, AutoencoderKL, AutoencoderTiny))
with set_seamless(vae_info.model, self.vae.seamless_axes), vae_info as vae: with set_seamless(vae_info.model, self.vae.seamless_axes), vae_info as vae:
assert isinstance(vae, torch.nn.Module) assert isinstance(vae, torch.nn.Module)
latents = latents.to(vae.device) latents = latents.to(vae.device)
if self.fp32: if self.fp32:
vae.to(dtype=torch.float32) vae.to(dtype=torch.float32)
use_torch_2_0_or_xformers = isinstance( # AutoencoderTiny doesn't contain a mid_block property or appear to accept attn processors
use_torch_2_0_or_xformers = hasattr(vae.decoder, "mid_block") and isinstance(
vae.decoder.mid_block.attentions[0].processor, vae.decoder.mid_block.attentions[0].processor,
( (
AttnProcessor2_0, AttnProcessor2_0,

3
invokeai/backend/stable_diffusion/seamless.py
View File

@ -5,6 +5,7 @@ from typing import Callable, List, Union
import torch.nn as nn import torch.nn as nn
from diffusers.models.autoencoders.autoencoder_kl import AutoencoderKL from diffusers.models.autoencoders.autoencoder_kl import AutoencoderKL
from diffusers.models.autoencoders.autoencoder_tiny import AutoencoderTiny
from diffusers.models.unets.unet_2d_condition import UNet2DConditionModel from diffusers.models.unets.unet_2d_condition import UNet2DConditionModel
@ -26,7 +27,7 @@ def _conv_forward_asymmetric(self, input, weight, bias):
@contextmanager @contextmanager
def set_seamless(model: Union[UNet2DConditionModel, AutoencoderKL], seamless_axes: List[str]): def set_seamless(model: Union[UNet2DConditionModel, AutoencoderKL, AutoencoderTiny], seamless_axes: List[str]):
# Callable: (input: Tensor, weight: Tensor, bias: Optional[Tensor]) -> Tensor # Callable: (input: Tensor, weight: Tensor, bias: Optional[Tensor]) -> Tensor
to_restore: list[tuple[nn.Conv2d | nn.ConvTranspose2d, Callable]] = [] to_restore: list[tuple[nn.Conv2d | nn.ConvTranspose2d, Callable]] = []
try: try: