metadata
dataset_info:
features:
- name: id
dtype: string
- name: headline
dtype: string
- name: article
dtype: string
- name: lang
dtype: string
- name: image_caption_separated
dtype: string
- name: topic_word_separated
dtype: string
- name: image_based_top_3
dtype: string
- name: caption_based_top_3
dtype: string
- name: image_based_top_5
dtype: string
- name: caption_based_top_5
dtype: string
- name: image_based_top_10
dtype: string
- name: caption_based_top_10
dtype: string
- name: image_based_top_15
dtype: string
- name: caption_based_top_15
dtype: string
- name: topic_word_separated_new
dtype: string
- name: topic_word_count_new
dtype: int64
- name: prompt_type
dtype: string
- name: article_prompt
dtype: string
splits:
- name: train
num_bytes: 9136949083
num_examples: 394353
- name: valid
num_bytes: 121366337
num_examples: 5187
- name: test
num_bytes: 358666498
num_examples: 15577
download_size: 5317632829
dataset_size: 9616981918