metadata
dataset_info:
features:
- name: text
dtype: string
- name: from
dtype: string
- name: category
dtype: string
splits:
- name: train
num_bytes: 22196900245.088894
num_examples: 27045120
- name: valid
num_bytes: 410367.9378218491
num_examples: 500
- name: test
num_bytes: 410367.9378218491
num_examples: 500
download_size: 27246460798
dataset_size: 22197720980.964535
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*