metadata
dataset_info:
features:
- name: text
dtype: string
- name: label
dtype:
class_label:
names:
'0': World
'1': Sports
'2': Business
'3': Sci/Tech
- name: keywords
dtype: string
- name: keywords_embeddings
sequence: float32
splits:
- name: train
num_bytes: 401783299
num_examples: 120000
- name: test
num_bytes: 25437883
num_examples: 7600
download_size: 493380668
dataset_size: 427221182
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*