Datasets:
metadata
language:
- mn
task_categories:
- image-to-text
dataset_info:
features:
- name: image
dtype: image
- name: text
dtype: string
splits:
- name: train
num_bytes: 1132342857.712
num_examples: 8091
download_size: 1114644837
dataset_size: 1132342857.712
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info: features: - name: image dtype: image - name: text dtype: string splits: - name: train