metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: label
dtype: float32
- name: tokens
dtype: string
- name: tree
dtype: string
- name: keywords
dtype: string
- name: keywords_embeddings
sequence: float32
splits:
- name: train
num_bytes: 29316285
num_examples: 8544
- name: validation
num_bytes: 3780849
num_examples: 1101
- name: test
num_bytes: 7584117
num_examples: 2210
download_size: 46990512
dataset_size: 40681251
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*