diffusers

diffusers copied to clipboard

Conversion Script: Update to Stable Diffusion 1.5 (EMA and non-EMA)

16

Since stable diffusion 1.5 was out, I was hoping to use it with the imagic colab. Could you upgrade diffusers to use the new model instead of the old one?

redwraith2

training sd with dreambooth/textual inversion with two different class of images

13

### Describe the bug Hi, I am following stable diffusion training with the dreambooth example at local machine. Although the following link gives every details about training sd, however I...

hamzafar

Update FlaxLMSDiscreteScheduler

1

- Add the missing `scale_model_input` method to `FlaxLMSDiscreteScheduler` - Use `jnp.append` for appending to `state.derivatives` - Use `jnp.delete` to pop from `state.derivatives`

dzlab

Add StableDiffusion repaint pipeline

12

nathanielherman

### Describe the bug clip guided generates a very poor result，     this is my code: `logger = logging.get_logger(__name__) # pylint: disable=invalid-name class CLIPGuidedStableDiffusionPipeline(DiffusionPipeline): r""" Pipeline for...

ScottishFold007

safety_checker and pipe checking and how to disable warning safety checker

4

### Describe the bug When disabling safety checker, diffusers spits out a wall of text, every time. This is an annoyance. We get it, you don't want people making boobies...

WASasquatch

ValueError: The tokenizer already contains the token ... Please pass a different `placeholder_token` that is not already in the tokenizer.

3

Just a quick question regarding tokens: I have a model trained on a new token and wish to add more detail and add extra run cycles to it. I get...

OzzyD

RePaintPipeline always returns black image when pipe.to("mps") is called

5

### Describe the bug When I try to do the inpainting with RePaintPipeline on the MPS backend it only returns black image. It works fine with CPU, but very slow....

GaidamakUA

### Describe the bug Dreambooth example uses multiprocess DataLoader (`num_workers=1`) and it causes error when pickling `collate_fn` because it is not top-level function. #### Workaround Training works if I disable...

Dragollla

Right set of UNet hyperparameters when training DDPM

2

Hi there ! I am currently training a DDPM model on a custom image dataset following the cool [unconditional_image_generation](https://github.com/huggingface/diffusers/blob/main/examples/unconditional_image_generation/train_unconditional.py) example script. Since I don't have the compute to perform comprehensive...

leopoldmaillard