metadata
dataset_info:
features:
- name: sentences
dtype: string
- name: tokens
sequence: string
- name: anim_tags
sequence:
class_label:
names:
'0': 'N'
'1': A
'2': H
- name: target-indexes
sequence: int64
splits:
- name: train
num_bytes: 16009926.20369589
num_examples: 44997
- name: test
num_bytes: 2442173.208751686
num_examples: 6669
- name: validation
num_bytes: 2101097.67878176
num_examples: 5941
download_size: 9561679
dataset_size: 20553197.091229334
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*