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

Comments fix

Browse Source 
Co-Authored-By: Ryan Dick <14897797+RyanJDick@users.noreply.github.com>
This commit is contained in:
Sergey Borisov 2024-07-30 00:39:50 +03:00
parent 2227a2357f
commit 1fd9631f2d
2 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
invokeai/backend/stable_diffusion/extensions/base.py
View File

@ -65,7 +65,7 @@ class ExtensionBase:
Args:
unet (UNet2DConditionModel): The UNet model on execution device to patch.
cached_weights (Dict[str, torch.Tensor]]): A read-only copy of the model's original weights in CPU, for
original_weights (Dict[str, torch.Tensor]]): A read-only copy of the model's original weights in CPU, for
unpatching purposes. Extension can save tensor which being modified, if it is not saved yet, or can
access original weight value.
"""

2
invokeai/backend/stable_diffusion/extensions/lora.py
View File

@ -57,7 +57,7 @@ class LoRAExt(ExtensionBase):
:param lora: LoRA model to patch in.
:param lora_weight: LoRA patch weight.
:param prefix: A string prefix that precedes keys used in the LoRAs weight layers.
:param original_weights: TODO:
:param original_weights: Dict of original weights, filled by weights which lora patches, used for unpatching.
"""
# assert lora.device.type == "cpu"