-
WIP HERE: https://github.com/lstein/stable-diffusion/compare/development...tmm1:dev-train-m1
---
I started experimenting with running main.py on M1 and wanted to document some immediate issues.
…
-
**Is your feature request related to a problem? Please describe.**
I frequently change the base and max size setting when using txt2img and img2img.
Right now I have to dedicate a huge portion of th…
-
Not sure if it's a bug, but training for 8000 steps with "subject_filewords.txt" didn't produce anything remotely connected with training data (using 6 imgs dataset), while using default "style_filewo…
-
Okay, so I've seen @lstein has added
`x = x.contiguous() if x.device.type == 'mps' else x`
to ldm/modules/attention.py in the doggettx-optimizations branch
but there's another error happening how
…
-
Ran `docker-compose up` and getting an error `ModuleNotFoundError: No module named 'k_diffusion'. Full output here:
```sh
$ docker-compose up
[+] Running 1/0
⠿ Container sd Created …
-
### 🐛 Describe the bug
the context for this is that it's part of cross-attention in stable-diffusion:
https://github.com/CompVis/stable-diffusion/blob/69ae4b35e0a0f6ee1af8bb9a5d0016ccb27e36dc/ld…
-
This issue is deprecated.
Please submit here: https://www.gptshunter.com/submit-gpt
## Thanks to our sponsor: FlowGPT
Click to view more.
![image](https://github.com/airyland/gptshunter.com/…
-
### Describe the bug
I have some ckpt files and try to convert it to diffusers in google colab.
with this command
`!python /content/diffusers/scripts/convert_original_stable_diffusion_to_dif…