metadata
dataset_info:
features:
- name: image
dtype: image
- name: filename
dtype: string
- name: subtype_label
dtype: string
- name: modifier
dtype: bool
- name: transcription
dtype: string
- name: metadata_json
dtype: string
- name: series_title
dtype: string
- name: episode_no
dtype: string
- name: program_title
dtype: string
- name: episode_title
dtype: string
- name: director
dtype: string
- name: producer
dtype: string
splits:
- name: train
num_bytes: 8277937
num_examples: 452
- name: test
num_bytes: 968768
num_examples: 51
download_size: 8981265
dataset_size: 9246705
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*