Spaces:
Runtime error
Runtime error
Commit History
-
Update diffusers version to support changing LoRA alpha
3fef5a7 -
Add a sample model
791824f -
Allow changing LoRA scaling alpha
c93c7f0
-
Let users jointhe LoRA Library organization
a3bc505 -
Change default values
f256f81 -
Update default values and add some explanations
9d220f6 -
Add changes from https://github.com/huggingface/diffusers/pull/2106
609aa31
-
Add changes from https://github.com/huggingface/diffusers/pull/2103
6ff910e -
Update requirements.txt
1164e72 -
Fix
15d58b2 -
Fix
c85d28f -
Update for lora-library org
8e506ce -
Save test images with train_dreambooth_lora.py
2873c82 -
Use Uploader to upload models in training time
Using two different upload methods was not a good idea. So, stop using upload method provided by train_dreambooth_lora.py and use Uploader class in this repo. Also, to make it easier to port updates for train_dreambooth_lora.py from the diffusers library, reset changes.8b149b2 -
Refactor
307f13f -
Fix
c11895e -
Update requirements.txt
f2dba90 -
Fix
65f2128 -
Change the default value for uploading
0e35688 -
Rename
99b9db8 -
Show warning when HF_TOKEN is not specified
4cc28d8 -
Update default model name
d049258 -
Hide the "remove GPU after training" button.
Training of a LoRA DreamBooth model often only takes about 10 minutes and this feature is not very useful for this Space.21bd5b5 -
Enable to remove GPU after model training is done
926d3f3 -
Update
31ac894 -
Update
2d7cbc5 -
Add tensorboard to requirements.txt
dd877cc -
Update how to use
96315e7 -
Add note on how to use
c750343 -
Disable the use_wandb button when WANDB_API_KEY is not set
f4da866
-
Rename
8956d86
-
Update to Diffusers
db1e5fb -
Duplicate from hysts/LoRA-SD-training
b083a19