metadata
dataset_info:
features:
- name: description
dtype: string
- name: original_dataset
dtype: string
- name: original_dimensions
dtype: string
- name: id
dtype: int64
splits:
- name: train
num_bytes: 10627268
num_examples: 70000
download_size: 3422117
dataset_size: 10627268
configs:
- config_name: default
data_files:
- split: train
path: data/train-*