metadata
dataset_info:
features:
- name: image
dtype: image
- name: caption
dtype: string
- name: categories
dtype: string
- name: source
dtype: string
- name: language
dtype: string
- name: matched_keywords
dtype: string
- name: is_nigerian
dtype: bool
- name: class_label
dtype: string
- name: image_url
dtype: string
- name: page_title
dtype: string
splits:
- name: train
num_bytes: 22420178
num_examples: 29
download_size: 22421924
dataset_size: 22420178
configs:
- config_name: default
data_files:
- split: train
path: data/train-*