metadata
dataset_info:
features:
- name: image
sequence:
sequence:
sequence: float64
- name: openpose
dtype: image
- name: cloth
dtype: image
- name: caption
dtype: string
- name: embeddings
sequence: float32
splits:
- name: train
num_bytes: 445603693
num_examples: 300
download_size: 34641782
dataset_size: 445603693
configs:
- config_name: default
data_files:
- split: train
path: data/train-*