-
Notifications
You must be signed in to change notification settings - Fork 82
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Request for training to use jpg and jpeg dataset source format (in addition to png). #84
Comments
You may change the related code here: Sana/diffusion/data/datasets/sana_data.py Line 52 in 41dcbe9
|
Hey @lawrence-cj , you do a great job, but please, don't answer things like "just do it". It's not funny at all. |
He didn't. He told you exactly what to do. |
Ok. So "img_extension=".png & .jpg & jpeg (please)", would be ok, but it's not. |
So, you mean you want to blend png jpg and jpeg all together to train? |
Hi @lawrence-cj . |
OK. I'll try to figure out a solution and let u know. |
Thank you @lawrence-cj . |
@AfterHAL The fix is passed and merged. The new meta_data.json is like: Sana/asset/example_data/meta_data.json Lines 1 to 6 in 2799bdc
You can specify the specific image name into the img_names list.Let me know if it helps to solve your problem. |
Hi.
I would change it myself if I could, but I have no clue where to do it in the code.
Thanks.
The text was updated successfully, but these errors were encountered: