metadata
dataset_info:
features:
- name: img_id
dtype: int64
- name: caption_id
dtype: int64
- name: caption
sequence: string
- name: noun_id
sequence: int64
- name: number
sequence: string
- name: anim
sequence: string
- name: deprel
sequence: string
splits:
- name: train
num_bytes: 1275
num_examples: 5
download_size: 4230
dataset_size: 1275
configs:
- config_name: default
data_files:
- split: train
path: data/train-*