metadata
dataset_info:
features:
- name: message_id
dtype: string
- name: subject
dtype: string
- name: from
dtype: string
- name: to
sequence: string
- name: cc
sequence: string
- name: bcc
sequence: string
- name: date
dtype: timestamp[us, tz=UTC]
- name: body
dtype: string
- name: file_name
dtype: string
splits:
- name: train
num_bytes: 1145885954
num_examples: 517401
download_size: 493482449
dataset_size: 1145885954
configs:
- config_name: default
data_files:
- split: train
path: data/train-*