metadata
dataset_info:
features:
- name: train
list:
- name: input
sequence:
sequence: int64
- name: output
sequence:
sequence: int64
- name: train_images
sequence: image
- name: test_inputs
list:
- name: input
sequence:
sequence: int64
- name: test_outputs
sequence:
sequence:
sequence: int64
- name: test_images
sequence: image
splits:
- name: train
num_bytes: 22775134
num_examples: 400
download_size: 14470405
dataset_size: 22775134
configs:
- config_name: default
data_files:
- split: train
path: data/train-*