metadata
dataset_info:
features:
- name: image
dtype: image
- name: english
dtype: string
- name: nepali
dtype: string
- name: mixed
dtype: string
- name: label
dtype: string
splits:
- name: train
num_bytes: 11701149
num_examples: 5000
download_size: 11441064
dataset_size: 11701149
configs:
- config_name: default
data_files:
- split: train
path: data/train-*