hollowstrawberry/kohya-colab
Accessible Google Colab notebooks for Stable Diffusion Lora training, based on the work of kohya-ss and Linaqruf
PythonGPL-3.0
Issues
- 5
Lora trainer result just black
#99 opened - 102
- 5
- 9
the problem is now
#96 opened - 9
Lora Trainer cannot be used, not supported between instances of 'TorchVersion' and 'Version'
#95 opened - 25
- 5
returned non-zero exit status 1.
#93 opened - 3
- 4
error when start tagging images
#91 opened - 3
- 1
Problem with xformer and cuda
#89 opened - 2
Add custom yaml for checkpoint
#88 opened - 2
[FeatureRequest] add custom VAE
#87 opened - 1
Request: Image Generation after Training
#86 opened - 2
add diffuser format as pretrained model
#85 opened - 0
Add v parameterization option
#84 opened - 1
LOHA training.
#83 opened - 1
CUDA backend failed
#82 opened - 1
ImportError during Tagging the Image
#81 opened - 3
- 1
FailedPreconditionError
#79 opened - 3
Proses has stopped
#78 opened - 4
Will there ever be support for SDXL?
#77 opened - 3
- 0
LoRa for XL models?
#75 opened - 1
No tensorflow-python?
#73 opened - 2
Can't use optional model for training
#72 opened - 3
- 1
Error when trying to train -CUDA SETUP-
#70 opened - 25
Cuda not working
#69 opened - 3
Google Drive Disconnects
#68 opened - 1
- 5
- 2
- 3
NameError: name 'resume' is not defined
#64 opened - 7
- 2
- 1
- 3
Full Brok3n
#60 opened - 1
Multiple Safetensor
#59 opened - 1
Maybe symbolic link error?
#58 opened - 0
SDXL
#57 opened - 2
Will Prodigy be added as optimizer ?
#56 opened - 5
Please put together a Paperspace version
#55 opened - 2
Please explain the num_repeat step
#54 opened - 2
- 1
Failes at step 4 everytime
#52 opened - 1
Tagging downloads every time,
#51 opened - 7
SDXL support?
#50 opened - 2