metadata
dataset_info:
features:
- name: image_path
dtype: image
- name: caption
sequence: string
- name: MeSH
dtype: string
- name: Problems
dtype: string
- name: Labels
sequence: string
splits:
- name: train
num_bytes: 12746800649.607
num_examples: 6719
- name: validation
num_bytes: 429296586
num_examples: 224
- name: test
num_bytes: 984473478
num_examples: 523
download_size: 14179074607
dataset_size: 14160570713.607
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*