metadata
dataset_info:
features:
- name: unique_id
dtype: string
- name: outlet
dtype: string
- name: headline
dtype: string
- name: article_text
dtype: string
- name: image_description
dtype: string
- name: image
dtype: image
- name: date_published
dtype: string
- name: source_url
dtype: string
- name: canonical_link
dtype: string
- name: new_categories
dtype: string
- name: news_categories_confidence_scores
dtype: string
- name: text_label
dtype: string
- name: multimodal_label
dtype: string
splits:
- name: train
num_bytes: 564156961.625
num_examples: 31339
download_size: 457879914
dataset_size: 564156961.625
configs:
- config_name: default
data_files:
- split: train
path: data/train-*