metadata
dataset_info:
features:
- name: transcript
dtype: string
- name: image
dtype: image
splits:
- name: train
num_bytes: 44875402.092
num_examples: 1233
- name: test
num_bytes: 10995611
num_examples: 309
download_size: 55631583
dataset_size: 55871013.092
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
task_categories:
- image-to-text
- image-text-to-text
language:
- ha
size_categories:
- 1K<n<10K