Steven Liu
Steven Liu
This PR combines "Distilled Stable Diffusion inference" with the "Speed up inference" doc. It also: - updates the table with only "speed" results (adds tf32 and combined results) and not...
I noticed that if you use the backticks for code format in a header, you can still see them in the right nav bar. Should I avoid using code format...
This PR attempts to clarify the difference between file types (`safetensors`, `ckpt`, `bin`) and storage layout (multifolder, single file) to avoid confusion when discussing "Diffusers" format and "single file" format....
Investigates why the `VideoProcessor` doc fails to build. I think it may have to do with how the `video` parameter is formatted in the docstring (related to #7937).
This PR adapts @asomoza's outpaint [blog post 3](https://huggingface.co/blog/OzzyGT/outpainting-inpaint-model) for the docs.
Fix and clarify the docs around: 1. FSDP only supports floating data type in `quant_storage` 2. show a more common way to access and configure this option with `BitsAndBytesConfig` in...
Continuation of #7817 (see comment [here](https://github.com/huggingface/diffusers/pull/7817#pullrequestreview-2048708715)) that refactors scheduler features for inference to their own doc. It includes: - custom `timesteps` and `sigmas` showcasing AYS - Karras sigmas - `rescale_betas_zero_snr`
Fixes #7734 where some training info for DeepFloyd IF got lost when I refactored the DreamBooth docs.
The main goal of this PR is to redesign the Transformers docs to: 1. Be more developer-friendly. 2. Improve navigation by replacing the existing structure with a more organic one...