metadata
dataset_info:
features:
- name: id
dtype: int32
- name: superimposed_image
dtype: image
- name: next_frame
dtype: image
splits:
- name: train
num_bytes: 7567984298.96
num_examples: 42920
download_size: 7431454261
dataset_size: 7567984298.96
configs:
- config_name: default
data_files:
- split: train
path: data/train-*