Conversation
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
|
@bot /style |
|
Style bot fixed some files and pushed the changes. |
|
Thanks for the script, do you have an example training run as well? |
|
running a testing train run with this dataset: https://huggingface.co/datasets/Norod78/Yarn-art-style |
sayakpaul
left a comment
sayakpaul
left a comment
There was a problem hiding this comment.
Looks great! Just a few questions :)
| --aspect_ratio_buckets="672,1568;688,1504;720,1456;752,1392;800,1328;832,1248;880,1184;944,1104;1024,1024;1104,944;1184,880;1248,832;1328,800;1392,752;1456,720;1504,688;1568,672" | ||
| ``` | ||
|
|
||
| ### Bilingual Prompts |
| weighting = compute_loss_weighting_for_sd3(weighting_scheme=args.weighting_scheme, sigmas=sigmas) | ||
|
|
||
| # flow matching loss | ||
| target = noise - model_input |
There was a problem hiding this comment.
Should the target be also flipped because of the negation above?
|
another example, trained on poli's tarot card dataset https://huggingface.co/datasets/multimodalart/1920-raider-waite-tarot-public-domain. 
|
|
The RHS perfectly fits my job description. |
|
@sayakpaul last failing test seems unrelated? |
3 participants
add lora training script for https://huggingface.co/Tongyi-MAI/Z-Image