metadata
dataset_info:
features:
- name: id
dtype: string
- name: annotators
sequence:
- name: label
dtype:
class_label:
names:
'0': hatespeech
'1': normal
'2': offensive
- name: annotator_id
dtype: int32
- name: target
sequence: string
- name: rationales
sequence:
sequence: int32
- name: post_tokens
sequence: string
splits:
- name: train
num_bytes: 7114650
num_examples: 15383
- name: validation
num_bytes: 884900
num_examples: 1922
- name: test
num_bytes: 884744
num_examples: 1924
download_size: 2094913
dataset_size: 8884294
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*