First of all, thank you for your wonderful work!
I would like to fine-tune the flux2-klein-base-9B model on the Light100K dataset by myself.
I followed the tutorial, which instructs me to organize the dataset like this:
datasets/flux2klein_alpha_interp5_20260501_unified_edgeexp/
control/
mask_normrgb_l01/
mask_normrgb_l02/
mask_normrgb_l03/
mask_normrgb_l04/
mask_normrgb_l05/
target_l01/
target_l02/
target_l03/
target_l04/
However, the dataset released on Hugging Face does not contain the mask_normrgb_l0* folders.
Could you please either release them on Hugging Face, or update the tutorial with instructions on how to generate these masks on my own?
First of all, thank you for your wonderful work!
I would like to fine-tune the flux2-klein-base-9B model on the Light100K dataset by myself.
I followed the tutorial, which instructs me to organize the dataset like this:
However, the dataset released on Hugging Face does not contain the
mask_normrgb_l0*folders.Could you please either release them on Hugging Face, or update the tutorial with instructions on how to generate these masks on my own?