metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 1583711
num_examples: 5
download_size: 1584991
dataset_size: 1583711
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
This dataset is compiled from google/dreambooth and is used to demonstrate the fine-tuning feature of mlx-flux. For details, see fine-tuning.