Added LORA, StableTuner

Signed-off-by: Georg Zoeller <georgzoeller@users.noreply.github.com>
This commit is contained in:
Georg Zoeller 2022-12-23 22:55:19 +08:00 committed by GitHub
parent 9b4a293ccb
commit 0a4194d1a1
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -109,8 +109,13 @@ Tools and models for use in conjuction with Stable Diffusion
## Training
* [Dreambooth-Stable-Diffusion](https://github.com/XavierXiao/Dreambooth-Stable-Diffusion) - Implementation of [Google's DreamBooth](https://arxiv.org/abs/2208.12242) for stable diffusion, allowing fine-tuning of the model for specific concepts.
* [LORA](https://github.com/cloneofsimo/lora) - LORA (Low-rank Adaptation) for Faster Text-to-Image Diffusion Fine-tuning (UNET+CLIP), a faster, more efficient Fine tuning method requiring less RAM than Dreambooth. [Huggingface demo here](https://huggingface.co/spaces/ysharma/Low-rank-Adaptation).
* [Stable Tuner](https://github.com/devilismyfriend/StableTuner) - Stable Tuner, an easy to install Dreambooth trainer with a very comfortable user interface.
* [Stable Diffusion Trainer](https://github.com/CCRcmcpe/scal-sdt) - Stable Diffusion trainer with scalable dataset size and hardware usage. Requires 10G of VRAM.
* [textual-inversion](https://github.com/rinongal/textual_inversion) - Addition of personalized content to Stable Diffusion without retraining the model ([Paper](https://textual-inversion.github.io/), [Paper2](https://dreambooth.github.io/)).