metadata
dataset_info:
features:
- name: image
sequence:
sequence:
sequence: float32
- name: conditioning_image
sequence:
sequence:
sequence: float32
- name: text
dtype: string
splits:
- name: train
num_bytes: 51170058344
num_examples: 27823
- name: test
num_bytes: 1379346000
num_examples: 750
download_size: 10656831248
dataset_size: 52549404344
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*