diffusers

diffusers copied to clipboard

Training dreambooth on multiple GPUs with deepspeed

3

### Describe the bug Hi, I am distributing dream booth training on 8 Tesla V100 (16GB) GPUs using deep speed. I have configured accelerate as below(output of default_config.yaml): `command_file: null...

hamzafar

dreambooth: add support for .txt captions for instance images

4

Not sure if instance captions are desired on the main dreambooth example, but this is what I've been using locally. FWIW, I haven't gotten great results with captions so far,...

timh

integrate ort into textual-inversion and text-to-image examples

1

prathikr

Strange behaviour and issues with SD 2.1 on multi-gpu (4090) and xformers attention

4

### Describe the bug Hello! Today I tried to setup SD 2.1 but have some trouble with more than one issue. ``` pipe = StableDiffusionPipeline.from_pretrained( "stabilityai/stable-diffusion-2-1", torch_dtype=torch.float16, ).to("cuda") # or...

Marcophono2

AttributeError: Can't pickle local object 'main.<locals>.<lambda>'

6

### Describe the bug Under Windows and conda, I get this error on the current diffusers version when I start examples/dreambooth/train_dreambooth.py `AttributeError: Can't pickle local object 'main..'` I solved setting...

djdookie

Forward pass of ScoreSdeVeScheduler scheduler is missing

3

### Describe the bug Hi, Forward pass is missing for this scheduler https://github.com/huggingface/diffusers/blob/v0.10.2/src/diffusers/schedulers/scheduling_sde_ve.py#L45 It would be great if this could be added. Could you give me some pointers how I...

cameronalonso2

MPS: models require an initial pass for reproducibility

10

I have no idea what's causing this, I suspect something related to RNGs, as we know they don't work properly in `mps` yet. This is a pending task from our...

pcuenca

Fix lr-scaling store_true & default=True cli argument for textual_inversion training.

5

`action="store_true"` means that the default action will be to set the value to `False`, and then if provided, set the value to `True`. Currently the default is manually set as...

aredden

May I ask what the T_max parameter means and what is its default value?

2


nikoharo

Remove duplicates in training scripts

2

As title.

haofanwang