metadata
dataset_info:
features:
- name: topic
dtype: string
- name: syntagm
dtype: string
- name: labels
dtype: int64
- name: root
dtype: string
splits:
- name: train
num_bytes: 2330816
num_examples: 34295
- name: test
num_bytes: 496594
num_examples: 7350
- name: eval
num_bytes: 499052
num_examples: 7349
download_size: 1104598
dataset_size: 3326462
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: eval
path: data/eval-*