metadata
dataset_info:
features:
- name: episode
dtype: string
- name: subtitle
dtype: string
- name: place
dtype: string
- name: person
dtype: string
- name: text
dtype: string
- name: text_original
dtype: string
splits:
- name: train
num_bytes: 740179
num_examples: 3344
download_size: 286909
dataset_size: 740179
configs:
- config_name: default
data_files:
- split: train
path: data/train-*