-
In the current implementation ([leejet/stable-diffusion.cpp@`4a6e36e`/stable-diffusion.h#L121](https://github.com/leejet/stable-diffusion.cpp/blob/4a6e36edc586779918535e12b4fbe0583044ee6f/stable-diffu…
-
**Is your feature request related to a problem? Please describe.**
The current implementation of the dreambooth trainings for both loras and finetuning is very memory intensive.
**Describe the sol…
-
I am not a programmer, I am a digital artist and I like to know and experience the tools that add a ComfyUI
I would like to be able to do these. Many times I use websites that have these tool the res…
-
FIrst of all, maybe it is my misunderstanding of the paper, so hope somebody could explain it for me, thanks! :
---
in the paper, the loss is defined as
![Screenshot from 2022-08-30 11-52-26](…
-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Contact Details
_No response_
### What should this feature add?
Support SDXL Checkpoint VAEs. For example, t…
-
Hi,
In the VAE paper (https://arxiv.org/pdf/1312.6114.pdf), the VAE loss function has no additional weight parameter for the KLD loss:
![vae_loss](https://user-images.githubusercontent.com/49433…
-
Hi, I am a maintainer of the [sd-webui-comfyui](https://github.com/ModelSurge/sd-webui-comfyui) extension for a1111.
The extension provides a comfyui node that reuses the webui checkpoint already l…
ljleb updated
2 weeks ago
-
### Describe the issue
FP16 model inference is slower compared to FP32. Does FP16 inference require additional configuration or just need to convert the model to FP16
### To reproduce
convert onnx …
-
Nice implementation of VAE tacotron paper though I havn't read paper yet ,but start model training with LJSpeech. As I see this code is extending the Tacotron 1 implementation so I am bit curious that…
-
### 1. Tasks before VAE development
--------
1A) Sigmoid for Vel/Timing
- [ ] Use sigmoid for velocity/micro-timing and use BCE loss instead of MSE
- [ ] Retrain the best four models (without …