Comments (17)
We don't support full text to image fine-tuning. However, you can refer to which provides this support: https://github.com/bghira/SimpleTuner.
from diffusers.
I also encounter NAN problem when adapting SD3 to another task
from diffusers.
please try --max_grad_norm=1
from diffusers.
also #8592 might help
from diffusers.
@bghira I do use these strategies, however, nan still exists.
from diffusers.
can you try adding QK Norm blocks back?
from diffusers.
this will extend the model to 3B parameters for higher performance (the 2B seems limited) and reintroduces the missing qk norm blocks
prepare for a longer training session
from diffusers.
Hi @bghira, thank you for your suggestion! It seems be hard to train for many users by introducing extra 1B parameters, i.e. extensive computing power and longer training iterations.
from diffusers.
I mean in order to fine-tuning a model, it is not wise to introduce at least 1B parameters. That should not be the correct direction for SD3.
from diffusers.
someone has to!
from diffusers.
I mean in order to fine-tuning a model, it is not wise to introduce at least 1B parameters. That should not be the correct direction for SD3.
You can just use the model there with the original number of blocks and copy in the parameters except QK norms, then init them as I had.
from diffusers.
My question is what happens when you try a lower learning rate or try to overfit a single batch of training data? In any case, this seems more like a discussion to me and not an "issue".
from diffusers.
I also encounter NAN problem when tuning the SD3 for image editing. Is there any new updates?
from diffusers.
Did you try the "logit_normal" weighting scheme?
from diffusers.
try --mixed_precision="bf16"
may help
from diffusers.
Where is this file located? train_text_to_imagesd3.py, I couldn't find it.Thanks
from diffusers.
Closing this since switching to "logit_normal" weighting scheme for loss resolves this issue in most cases. If not, please re-open and I will turn it into a discussion.
from diffusers.
Related Issues (20)
- how to unload a pipeline HOT 3
- Error(s) in initializing SD3ControlNetModel by from_transformer HOT 9
- Text encoder errors in both SD3 dreambooth training scripts HOT 6
- Rethinking the `encode_prompt()` method in pipelines
- Using `torchsde.BrownianInterval` instead of `torchsde.BrownianTree` in class `BatchedBrownianTree` HOT 2
- Unexpected validation image on webdataset SDXL controlnet training script HOT 2
- Implementation of DDIM, why taking Xt and (t-1) as input? HOT 1
- Batched Text to image generation really slow with pipeline HOT 7
- Advanced training SD1.5 has an issue when saving checkpoints HOT 5
- `SD3Transformer2DModel` doesn't get configured HOT 7
- Broken diffusers from version 0.29.0 LoRAAttnProcessor2_0 is removed project that might depend on this class will error HOT 2
- use load from single file for advance model training HOT 4
- SD3 cannot finetunes a better model (hand and face deformation)? HOT 3
- StableDiffusionXLPipeline.from_single_file can't use in offline server HOT 3
- AutoEncoderKL only works for 2D shaped inputs HOT 3
- SD3 - image_processor.py:111: RuntimeWarning: invalid value encountered in cast images = (images * 255).round().astype("uint8") HOT 4
- Clarification Needed on Hardcoded Value in Conditional Statement in LeditPP
- Removing LoRAAttnProcessor causes many dependencies to fail HOT 1
- Model and input data type is not same
- Integrate HiDiffusion
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 diffusers.