Skip to content

How to use custon dataset in train_dreambooth_flux.py. #10655

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

Closed
rooooc opened this issue Jan 26, 2025 · 3 comments · Fixed by #10664
Closed

How to use custon dataset in train_dreambooth_flux.py. #10655

rooooc opened this issue Jan 26, 2025 · 3 comments · Fixed by #10664

Comments

@rooooc
Copy link

rooooc commented Jan 26, 2025

Hi. what if i want to train two images with two different prompts. somethink like m1.jpeg , m1.txt ; m2.jpeg, m2.txt.
the default example only shows all images share one instant prompt. thanks for the help!

@hlky
Copy link
Contributor

hlky commented Jan 27, 2025

Look at options --dataset_name --image_column --caption_column, you may need to prepare the dataset so that datasets understands it.

cc @linoytsaban do you have any examples for this?

@linoytsaban
Copy link
Collaborator

Yes exactly you can use --dataset_name --image_column --caption_column to pass images with correspond captions,
this example should make it clear :)
let us know if you have further questions

@hlky
Copy link
Contributor

hlky commented Jan 27, 2025

Thanks @linoytsaban

@rooooc I've added documentation and a script to create metadata.jsonl from your image text pairs here #10664

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants