Comments (5)
Was getting the same but I think the README example does not show the correct config to use?
Here is an example that worked for me without noise. Download the 512-base-ema.ckpt from here
python scripts/img2img.py --prompt "A fantasy landscape, trending on artstation" --init-img ~/Pictures/example.png --ckpt 512-base-ema.ckpt --config configs/stable-diffusion/v2-inference.yaml
Or with the 768 ckpt:
python scripts/img2img.py --prompt "A fantasy landscape, trending on artstation" --init-img ~/Pictures/example.png --ckpt 768-v-ema.ckpt --config configs/stable-diffusion/v2-inference-v.yaml
from stablediffusion.
try changing you sampler from PLMS to DPM and see if you get sane results. I worked for me. you can do that at the command line by adding "--dpm" to your txt2img line
from stablediffusion.
Sorry, I missed you saying img2img in title. yes, doesn't look like img2img allows you to specify another sampler. if you look in the code (scripts\img2img.py) you should see:
device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
model = model.to(device)
**sampler = DDIMSampler(model)**
I guess you could experiment with substituting another sampler invocation there - if you are getting the excessive noise from img2img, but I also wonder if its another problem. I have not seen oddities from img2img, only txt2img with PLMS sampler.
from stablediffusion.
try changing you sampler from PLMS to DPM and see if you get sane results. I worked for me. you can do that at the command line by adding "--dpm" to your txt2img line
Do you know why the PLMS sampler does not work in this case? I also found this same issue with the txt2img.py script. I get noise unless I use DPM or the default ones.
from stablediffusion.
Try adding the config flag --config configs/stable-diffusion/v2-inference.yaml
. It made the output much better for me.
from stablediffusion.
Related Issues (20)
- Stable diffusion does not start
- Repo
- Missing imports
- x4-upscaler get very bad results
- Unable to Locate 'imwatermark' Module HOT 1
- ControlNet wot working after update
- RuntimeError: Input type (c10::Half) and bias type (float) should be the same
- FutureWarning: 'resume_download' error
- How to use DDPM to convert a low-resolution image into a high-resolution image?
- [algorithm suggestion] "Foveated" image representation when sampling?
- Generated images
- Switching between Inpaint maked and not masked is broken
- update to 3.0
- English Resume
- English Resume
- Parameterization "v" Related Questions in LoRA Training
- 自由驰骋,塔斯马尼亚自驾探险之旅
- What is the script for fine-tuning on one's own dataset? HOT 1
- КРИТИЧЕСКИЕ ОШИБКИ В Python И Yaml
- SinNimbre
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from stablediffusion.