metadata
dataset_info:
features:
- name: text
dtype: string
- name: from
dtype: string
- name: category
dtype: string
splits:
- name: train
num_bytes: 4628371619.233266
num_examples: 6322516
- name: valid
num_bytes: 366022.9265717371
num_examples: 500
- name: test
num_bytes: 366022.9265717371
num_examples: 500
download_size: 8316922290
dataset_size: 4629103665.08641
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*