datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
CelebHQ-face-segmentationface_segmentation_20This is the dataset used for FluxKontext or Qwen-Image-Edit training with the task image-text2image.
This is a good adtaset that can very fine the effect of finetuning, where the base model is not good at
Refer this for the lora finetune model qwen-image-edit-lora-face-segmentation
Usage of the dataset
dd = load_editing_dataset("TsienDragon/face_segmentation_20")
sample = dd["test"][0]
Example of the data structure
DATASET_ROOT/train
├── control_images
│ ├── 060002_4_028450_FEMALE_30.png
│… See the full description on the dataset page: https://huggingface.co/datasets/TsienDragon/face_segmentation_20.