metadata
dataset_info:
features:
- name: text
dtype: string
- name: from
dtype: string
- name: category
dtype: string
splits:
- name: train
num_bytes: 28297947976.691166
num_examples: 13947605
- name: valid
num_bytes: 1014437.5316296656
num_examples: 500
- name: test
num_bytes: 1014437.5316296656
num_examples: 500
download_size: 19959212243
dataset_size: 28299976851.75443
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*