diffusers
diffusers copied to clipboard
huggingface
🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.
**Problem:** We often generate images with a batch_size >1. However, images in the batch (after the first image) by default have an **seed that is unknown to the user**, so...
**Is your feature request related to a problem? Please describe.** I'm trying to convert portions of unet into CoreML. However, CrossAttention fails to compile to the Apple Neural Engine. **Describe...
### Describe the bug I'm using the following code: ``` !pip install diffusers !pip install transformers scipy ftfy pipe = StableDiffusionPipeline.from_pretrained("CompVis/stable-diffusion-v1-4", revision="fp16", torch_dtype=torch.float16, use_auth_token=True) prompt = "Pineapple on a white...
I'd like to propose an idea analagous to https://github.com/huggingface/diffusers/issues/369. The current fine tuning script for textual inversion initialises the new `placeholder_token`'s embedding with an existing `initializer_token` (and enforces that the...
It would be nice to discuss a bit the push to hub design of the library. IMO we have two different use cases for `push_to_hub`. 1. The complement of `from_pretrained(...)`....
### Describe the bug As we discuss in #266 > The original textual inversion support [using more than one vector](https://github.com/rinongal/textual_inversion/blob/main/ldm/modules/embedding_manager.py#L39) to represent the learnt concept. For the current implementation, if...
### Describe the bug Thanks for releasing this great work, it really makes using diffusion a easy thing! But later when I tried to train a `UNet2DConditionModel` from scratch, I...
I used multiple tokens to represent a concept by adding num_vec_per_token number of tokens to the tokenizer which can be initialized with the initial_token. The tokens would be labeled as...
The [Textual Inversion training notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/sd_textual_inversion_training.ipynb) in Google Colab only takes in remote images, instead of also being able to take in local images either uploaded from the desktop or linked...
This standardizes the use of the argument `image` in all pipelines instead of a mix of `init_image` and `image`. Resolves #1257
Metadata
Owner
Metadata
🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.